Development and Characterization of Microhaplotype Databases for Diverse Crop Species

Development and Characterization of Microhaplotype Databases for Diverse Crop Species | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Development and Characterization of Microhaplotype Databases for Diverse Crop Species Dongyan Zhao, Meng Lin, Cristiane H. Taniguti, Alexander M. Sandercock, and 36 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9292361/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 7 You are reading this latest preprint version Abstract Microhaplotypes are short genomic segments that contain multiple tightly linked variants, providing multiallelic data that can enhance genetic resolution compared to traditional biallelic single nucleotide polymorphism (SNP) markers. Here, we present the creation and utilization of separate microhaplotype databases for eight crop species representing diverse genome sizes, ploidy levels, and breeding systems. We developed a standardized, species-agnostic pipeline for processing, filtering, and databasing microhaplotypes generated using the DArTag targeted genotyping platform. To enhance user accessibility, we developed a no-code, user-friendly application, HapApp, that uses an R Shiny front-end interface to allow breeders and researchers to add unique, standardized microhaplotype identities from raw DArTag reports and iteratively update the existing crop-specific database with the newly discovered microhaplotypes. Comparative analyses of these databases highlighted the advantages of microhaplotypes in capturing greater allelic diversity, resolving fine-scale population structures, and improving linkage map construction. This integrated framework provides a reproducible and scalable foundation for managing and exploiting microhaplotype data in plant breeding and genetic research, enabling robust cross-project comparisons and facilitating trait discovery in both simple and complex crop genomes, while enabling comparative genomics and cross-species functional transfer that accelerates genetic gains across all crop species. microhaplotypes crop species HapApp allelic diversity multi-allelic markers Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Key message Standardized microhaplotype databases for eight diverse crops enable multiallelic analyses, comparative genetics, and breeding decisions. Introduction The integration of molecular markers into plant breeding programs has dramatically accelerated genetic improvement by enabling precise germplasm characterization and selection. DNA-based markers in plants have taken many forms over the past several decades, starting with restriction fragment length polymorphisms (RFLPs) in 1980 (Botstein et al. 1980 ), random amplified polymorphic DNA (RAPD) (Williams et al. 1990 ; Welsh and McClelland 1990 ), amplified fragment length polymorphisms (AFLPs) (Vos et al. 1995 ), simple-sequence repeats (SSRs) (Mrázek et al. 2007 ), and single-nucleotide polymorphisms (SNPs) (Mammadov et al. 2012 ). Today, the predominant DNA marker type is the bi-allelic SNP marker due to its abundance and wide distribution across genomes and its suitability for high-throughput genotyping. SNPs can be identified cheaply and quickly using a wide range of detection technologies. For researchers, SNP datasets require minimal storage space and are easier to interpret than many other marker systems. There are several software packages available for researchers to use that aid allele calling from raw data, including proprietary software such as KlusterCaller and SNPviewer (LGC Ltd, Huntingdon, UK) and GenomeStudio (Illumina, San Diego, CA, USA), and free R packages such as polyRAD and updog (Clark et al. 2019 ; Gerard and Ferrão 2020 ). SNPs are often preferred for time-sensitive, data-driven decisions on germplasm in breeding programs. As such, SNPs have become indispensable for assessing genetic diversity and population structure, mapping traits of interest, and implementing marker-assisted selection/genomic selection in plant and animal breeding programs. While bi-allelic SNP calling is often sufficient and accurate for diploids or species with low levels of heterozygosity, it lacks sufficient information to accurately resolve genotypes in highly heterozygous species and species with complex genomes, especially polyploids. Recent advances in DNA sequencing technology have led to the development of various approaches for genetic architecture profiling. Whole-genome sequencing (WGS) provides the most comprehensive view of genetic variation but remains cost-prohibitive to be applied yearly for routine breeding applications. To address this limitation, reduced representation approaches have emerged as cost-effective alternatives. These methods can be broadly categorized into two strategies: random sampling and targeted amplification of genome sequences. Random reduced-representation methods, such as genotyping-by-sequencing (GBS) and Diversity Arrays Technology sequencing (DArTseq), employ restriction enzymes to sample a subset of the genome (Elshire et al. 2011 ; Cruz et al. 2013 ). GBS typically utilizes methylation-sensitive restriction enzymes to reduce genome complexity, thereby enriching for low-copy and genic regions and increasing the likelihood of detecting functional variants. In DArTseq, a species-specific combination of restriction enzymes is used to achieve genome complexity reduction and preferentially amplify low-copy fragments. Both approaches generate reduced-representation libraries and have proven particularly valuable for species with limited or no reference genome resources, enabling rapid and cost-effective genome-wide variant discovery. The downside of both methods is the low reproducibility from one project to another, since a random subset of a reduced representation library is sequenced each time, leading to high missing data rates between project runs. Additionally, the data analysis is computationally intensive, requiring skilled bioinformatic support and ample time to complete. In contrast, targeted genotyping approaches such as DArTag, AgriSeq, FlexSeq, GT-seq, and related platforms are typically designed with primers flanking an anchor SNP at each locus (Campbell et al. 2015 ; Semalaiyappan et al. 2022 ; Blyton et al. 2023 ; Clare et al. 2024 ). These methods offer several advantages over GBS and DArTseq, including higher coverage of target loci, low missing data rates, read depths that are consistent across samples and projects ensures data are comparable across projects, and reduced computational requirements for data analysis (Zhao et al. 2023 , 2024b , a ; Hardigan et al. 2023 ; Sandercock et al. 2025 ). A further key advantage is the ability to generate detailed sequence information surrounding variant sites, as these approaches employ next-generation sequencing of short reads (typically 50–250 bp). Rather than providing only single nucleotide calls at target variant sites, these platforms return data in two formats: raw FASTQ files containing individual sequencing reads, and semi-processed marker reports presenting multi-allelic sequences for each locus. These sequences contain not only the target variant site but also capture adjacent polymorphisms within the amplified regions. The co-occurrence of closely linked variants within these short haplotype blocks can yield three or more allelic combinations, characterizing a multi-allelic microhaplotype (Kidd et al. 2013 ). Due to their short length, variants within a microhaplotype are assumed to remain in strong linkage and be inherited as a unit, and the combination of variants within have known phasing. Multiallelic genetic data can provide more detailed and nuanced insights into genetic diversity, allele dosage, trait expression, and population structure. The comprehensive sequence information also enables more accurate genotype calling through sequence-context evaluation, enhanced discrimination power through microhaplotype-based analysis, and the identification of novel alleles within the amplified regions. The concept of microhaplotypes was first introduced in the forensics field by Kidd and colleagues (Kidd et al. 2013 ) and rapidly gained popularity due to their high information content and analytical efficiency. Subsequent studies in forensic genetics have demonstrated the advantages of using multi-allelic microhaplotypes over individual SNPs in diverse applications, including kinship investigations, ancestry inference, resolution of complex DNA mixture, and recovery of allelic information from degraded DNA specimens (Kidd et al. 2014 ; Oldoni and Podini 2019 ; de Barros Rodrigues et al. 2025 ; Turchi et al. 2026 ). Beyond forensics, the use of microhaplotypes derived from targeted genotyping panels has demonstrated notable advantages in aquaculture. GT-seq microhaplotypes have been successfully applied to improve parentage assignment in oysters, salmon, and rockfishes (Baetscher et al. 2018 ; Thompson et al. 2025 ; Anderson et al. 2025 ). For example, simulation studies in oyster and salmon breeding programs have indicated that incorporating microhaplotypes can increase the accuracy of estimated breeding values (Delomas et al. 2024 ). In plant breeding, the advantages of using multi-allelic markers have been highlighted across diverse applications, including genomic prediction (GP) and genomic selection (GS) as well as genome-wide association studies (GWAS) (Matias et al. 2017 ; Abed and Belzile 2019 ). Traditionally, most multi-allelic markers in crop studies have been derived from GBS or WGS datasets by estimating combinations of SNPs in complete linkage disequilibrium, referred to as haplotype blocks. Such blocks are typically inferred through methods such as confidence interval-based approaches using normalized measures of allelic association, identification of genomic regions with limited recombination events, or read-based phasing approaches (Gabriel et al. 2002 ; Wang et al. 2002 ; Garrison and Marth 2012 ; Martin et al. 2016 ). Simulation studies have shown that GWAS models accommodating multiallelic markers and haplotype blocks in polyploid species achieve higher accuracy than models based on bi-allelic SNPs (Thérèse Navarro et al. 2022 ). More recently, microhaplotypes derived from DArTag panels have been applied to genomic studies in alfalfa, demonstrating higher values of both intra- and inter-population diversity compared to SNP markers (Medina et al. 2025 ). Over the past five years, Breeding Insight (BI; RRID: SCR_026645) and collaborators have developed a series of medium-density DArTag genotyping panels (Zhao et al. 2023 , 2024b , a , 2025 ; Hardigan et al. 2023 ; Endelman et al. 2024 ; Chen et al. 2025 ). These panels were developed using a diverse breeding germplasm and validated across F 1 and/or backcross mapping populations as well as core diversity panels, ensuring robustness across breeding materials. The DArTag genotyping platform produces several data outputs including traditional FASTQ files and a proprietary format called Missing Allele Discovery Counts (MADC) file. The MADC file contains read counts for all microhaplotypes at targeted loci and captures four distinct classes of microhaplotypes. Reference (Ref) and Alternative (Alt) alleles represent the target variants known a priori from panel design and are analogous to SNP marker definitions of the same designation. In contrast to Ref and Alt, RefMatch and AltMatch sequences match either the reference or alternative state at the target site while containing additional polymorphisms in the flanking sequences that were not known at the time of design, thus newly “discovered”. One challenge in utilizing MADC data across projects stems from the lack of unique identifiers for RefMatch and AltMatch alleles from all genotyping projects. To systematically organize the RefMatch and AltMatch information, we have established a species-agnostic pipeline to assign standardized and unique microhaplotype identities (IDs) to serve three essential purposes: 1) enable efficient tracking of marker genetic information, 2) facilitate cross-project and cross-institution comparison of genotyping results, and 3) abide by Findable, Accessible, Interoperable, and Reusable (FAIR) data principles ( https://www.go-fair.org/fair-principles/ ). Here, we present the individual microhaplotype databases for eight plant species (alfalfa, blueberry, cranberry, cucumber, pecan, potato, strawberry, and sweetpotato) and open them to the public such that breeders and researchers can fully leverage the rich information generated via DArTag genotyping through community efforts. Materials and Methods Datasets used to generate the microhaplotype databases Diverse sample sets, including F 1 and backcross populations for each species, were collected by breeders and researchers. These samples were genotyped at DArT ( www.diversityarrays.com ), and the resulting MADC reports were shared with BI to facilitate the generation of microhaplotype databases. Naming convention To ensure naming consistency across DArTag panels developed at different times and by different research groups, we implemented a systematic marker ID standardization process. The original marker IDs, which varied in format and naming conventions, were converted to a uniform structure based on chromosome number and genomic position from the reference genome used for panel design. Genomic positions were padded to nine digits to maintain consistent formatting (e.g., chr1_000589632). This standardization step was performed prior to microhaplotype ID assignment to ensure marker tracking across all panels and downstream analyses. Core database structure and panel-specific considerations The core database forms the foundation of the microhaplotype tracking system, offering consistent reference points for allele comparison and standard templates for sequence alignment. This core database contains both Ref and Alt alleles, with sequence lengths varying according to panel design specifications. The genomic insert size in DArTag panels has lengthened over time as sequencing costs have decreased, resulting in two main design categories. The 54-bp design (~ 75 bp with primer sequences before trimming) was implemented in alfalfa, blueberry, and Phase I markers in potato and strawberry panels. The 81-bp design (~ 100 bp with primer sequences before trimming) was implemented in cranberry, cucumber, pecan, sweetpotato panels, and Phase II markers in potato and strawberry panels. Construction of reference-based core database The construction of the core microhaplotype database for reference-based markers involves an iterative process using probe design files, reference genomes, and analytical scripts. This workflow addresses two key challenges: 1) in the genotyping reports, Ref and Alt amplicons often contain International Union of Pure and Applied Chemistry (IUPAC) codes representing ambiguous base calls, and 2) the 3’ ends of some amplicons may contain sequencing errors. The primary goal is to create a core database of Ref and Alt microhaplotypes that are free of IUPAC codes and mismatches (except for the target variant in Alt) with the reference genome. Another critical consideration is that a DArTag panel, like other amplicon-sequencing technologies, is composed of amplicons derived from both the top and bottom strands of the reference genome. Accurately establishing the strand orientation is crucial for determining the precise positions and nucleotides of the variant sites. To maintain consistency, all microhaplotype databases preserve strand orientation as implemented during panel creation by DArT. The workflow begins by extracting 180–300 bp flanking sequences encompassing the target variants from the corresponding reference genome (Table 1 ). Ref and Alt amplicon sequences are then retrieved from the DArTag MADC report and aligned to the flanking sequences via BLAST (Camacho et al. 2009 ). This alignment step helps determine the correct strand orientation of the amplicons, and those amplicons originating from the bottom strand are exported as a key file for downstream analysis ( Table S1 ). Meanwhile, the coordinates of the Ref and Alt microhaplotypes are determined based on the reference genome, and sequences are fetched to populate the core database. By design, the core database contains twice the number of marker loci in a DArTag panel, as it incorporates both Ref and Alt microhaplotype sequences for each locus. Table 1 Summary of DArTag panels and microhaplotype databases across eight species. Species Ploidy Reference genome Developed by Design length #Markers Database version #Microhaplotypes #Samples Alfalfa 4x (Chen et al. 2020 ) BI 54 bp 3K 50 35,259 15,600 Blueberry 4x (Colle et al. 2019 ) BI 54 bp 3K 20 28,653 8,930 Cranberry 2x (Diaz-Garcia et al. 2021 ) BI 81 bp 3K 5 14,380 4,146 Cucumber 2x (Yang et al. 2012 ) BI 81 bp 3K 10 9,823 8,272 Pecan 2x (Lovell et al. 2021 ) BI 81 bp 3K 9 26,073 6,768 Sweetpotato 6x (Wu et al. 2018 ) BI 81 bp 3K 19 37,593 9,212 Potato 4x (Pham et al. 2020 ) CIP-WU 54–81 bp 4K 15 35,741 3,102 Strawberry a 8x (Hardigan et al. 2021 ) Collaborator 54–81 bp 5K 3 38,227 1,880 a Panel designed for cultivated strawberry (octoploid, 8x) where markers were sub-genome specific, enabling functional diploid analysis. Systematic processing and database integration of novel microhaplotypes We developed a systematic workflow to process MADC reports and assign standardized microhaplotype IDs. The process begins with updating marker IDs to the standard “chromosomeN_000000000” format from the initial MADC report. As an initial quality control, we filter RefMatch and AltMatch sequences to require a minimum presence of either 10 samples or 5% of the total samples in a project (whichever is smaller), with each sample having at least two reads (Fig. 1a). The workflow diverges from here, based on amplicon size. For reports containing sequencing reads longer than genomic insert size by panel design, the reads often contain adapter sequences. In this case, microhaplotypes were first cleaned by removing potential adapter sequences with Cutadapt (Martin 2011 ), and any duplicates identified among RefMatch and AltMatch sequences are marked for exclusion. For reports with reads matching the panel design length, no adapter removal is needed. All sequences underwent BLAST analysis against the existing microhaplotype database. Any previously cataloged sequences in the database were renamed with their previously assigned unique database IDs to produce an updated MADC report with fixed microhaplotype IDs. Any novel sequences that met our inclusion criteria (≥ 90% identity over ≥ 90% coverage of subject microhaplotypes) were incorporated into the database and then assigned new unique identifiers, following the RefMatch and AltMatch classification scheme. These new unique IDs were then written back to the updated MADC report, which then had unique standardized IDs for all observed microhaplotypes. The process yielded two key outputs: (1) MADC report with standardized universal microhaplotypes IDs (MADC_fixedID) and filtered for missing data, and (2) a new version of the microhaplotype database incorporating newly identified alleles with unique IDs. If no new alleles were found, the database remained unchanged, but the MADC report is still generated with universal standardized microhaplotype IDs. Creation of an R Shiny interface for processing standardized microhaplotype ID assignment The workflow described above provides a flexible pipeline for processing MADC files, identifying new unique microhaplotypes, and producing an updated MADC with fixed microhaplotype IDs based on the microhaplotype database. However, its Bash and Python scripting format makes it inaccessible to users without coding experience or who are uncomfortable with their skill level. To improve user accessibility, we created a no-code R Shiny interface that provides a point-and-click method for using the workflow (Chang et al. 2024 ), which has been shown to provide complex plant and animal breeding workflows with a more approachable framework (Sandercock et al. 2025 ). When the original MADC file is uploaded, it is first checked to confirm it is in the correct format. Then the application submits the MADC to the pipeline described above, which performs the computation steps. The updated information is captured and displayed to the user along with any warnings or notifications. The output files include 1) a new version of the microhaplotype FASTA file, and 2) a MADC file with microhaplotypes that passed initial quality control and were assigned a unique standardized ID. Sequence alignment for the alfalfa and blueberry microhaplotype databases The alfalfa microhaplotype database, as of November 2025, is at v50. It was aligned to the haplotype-based ‘XinJiangDaYe’ reference genome (Chen et al. 2020 ) to understand the distribution of potential paralogous amplifications. The alignment was performed using Bowtie2 v2.5.1 (Langmead and Salzberg 2012 ) with default parameters. All possible alignments were returned followed by a preliminary filter to remove any alignment with a total number of insertions/deletions >5bp. A total of 34,891 microhaplotypes were successfully aligned to at least one locus in the reference genome. The syntenic genomic regions (Chen et al. 2020 ) across four haplotypes in each of the eight homologous groups were used to determine the most likely syntenic alignment when a microhaplotype aligned to multiple loci on a single haplotype. A microhaplotype was considered a potential paralogous amplification if it had 1) one or more alignments outside its target homologous group or 2) two or more copies on any of the four haplotypes within the homologous group but outside the target genomic region (Figure S1 C & D). For DArTag loci containing only non-paralogous microhaplotypes, the copy number (ranging from one to four) of each target locus was determined based on the number of aligned positions within the corresponding syntenic regions (Figure S1 a & b). The most up-to-date blueberry microhaplotype database, v20, was characterized for potential paralogous amplification using the same strategy as in alfalfa. The blueberry database was aligned to ‘Draper’ reference genome (Colle et al. 2019 ) and 28,448 microhaplotypes were successfully aligned to at least one locus in the reference genome. Potential paralogous amplifications and copy number of non-paralogous loci in the blueberry microhaplotype database were determined as described above. Comparison of microhaplotypes, target, and off-target markers through linkage map construction in pecan Linkage maps were built using OneMap v3.2.2 (Taniguti et al. 2022 ) based on a previously published pecan ( Carya illinoinensis ) F 1 dataset comprising 188 progeny (Chen et al.). The mapping procedure was applied using three different data set marker types: (i) microhaplotypes, (ii) target SNPs only, and (iii) combined target and off-target SNPs. For the microhaplotype dataset (i), the MADC file was processed using our standard workflow, which assigns fixed microhaplotype IDs based on the pecan microhaplotype database v9. The processed file was converted to a polyRAD (v2.0.0) input object using the readDArTag function, and genotypes were called using the IterateHWE function, which assumes Hardy-Weinberg equilibrium without population structure (Clark et al. 2019 ). We used this approach instead of PipelineMapping2Parents to avoid errors that can arise when parental genotypes are altered or imputed from progeny segregation patterns (Taniguti et al. 2022 ). The resulting genotypes were exported as a Variant Call Format (VCF) file using the RADdata2VCF function with asSNP parameter set to FALSE. For datasets (ii) and (iii), VCF files were generated from the MADC file with the madc2vcf_targets and madc2vcf_all functions from the BIGr (v0.6.2) package (Sandercock et al. 2025 ), respectively. These VCFs contained allele count data and were processed in polyRAD under the same Hardy-Weinberg equilibrium (HWE) genotype calling model used for dataset (i). All datasets were subjected to the same filtering criteria: a minimum genotype depth of six reads, a mean marker depth greater than 30, a maximum of 25% missing genotypes per marker and per individual, removal of redundant markers, and exclusion of markers deviating from expected Mendelian segregation (α = 0.05, Bonferroni-corrected for multiple tests). Additional filtering was performed using the rf_snp_filter_onemap function, which assesses pairwise recombination fractions (rf) and LOD scores to identify informative markers. For each pair of markers, the function retained only comparisons with strong linkage support (LOD > 5) and low recombination (rf < 0.15). Pairs that did not meet these criteria were temporarily masked (set to missing), ensuring that only informative pairwise combinations contribute to the count. The number of remaining informative (non-missing) pairwise comparisons was then counted for each marker, and markers falling below the 5th percentile or above the 95th percentile of the resulted counts distribution were removed. Markers were assigned to linkage groups according to their chromosome in the reference genome. Linkage distances were estimated considering the order of markers using either their physical positions or the multidimensional scaling (MDS) ordering algorithm (Preedy and Hackett 2016 ). When MDS ordering failed due to collinearity, the lower quantile threshold in rf_snp_filter_onemap was relaxed. In dataset (ii), the threshold was raised to 0.40 for linkage group (LG) 2 and to 0.35 for LG 10 and LG 11. In dataset (iii), the threshold was raised to 0.35 for LG 2 and LG 11. The performance of MDS ordering algorithm in each dataset was evaluated by computing the absolute value of Spearman's rank correlation coefficient ρ between the algorithm-derived marker order and the reference genome order. The correlation coefficient was calculated according to Spearman (1904) using the formula \(\:\rho\:=\:1\:-\:\backslash\:frac\left\{6{\sum\:}_{\left\{i=1\right\}}^{\left\{m\right\}{d}_{i}^{2}}\right\}\left\{m\right(m^2-1\left)\right\}\) , where d i denotes the rank difference for marker mi between the estimated and true orders. Comparative analysis of microhaplotype and SNP genotyping data in sweetpotato breeding germplasm Microhaplotype variant data were generated from MADC reports in which alleles had been assigned unique standardized IDs across seven independent sweetpotato DArTag genotyping projects. For each MADC dataset, only Ref and Alt microhaplotypes were retained for marker loci with greater than 15 microhaplotypes since they likely contained some paralogous amplification. Microhaplotype genotype dosage calls were produced as VCF using the readDArTag function in the polyRAD (v2.0.0) package (Clark et al. 2019 ). Individual project VCF files were merged to form a single consolidated VCF and subjected to sequential quality filtering. Variants with low coverage (mean read depth 1000), or failing missingness criteria were filtered using a population-aware strategy. A microhaplotype was retained if it met either the global threshold of ≤ 20% missing data across all samples or the per-population threshold of ≤ 20% missing data in at least one population. This approach ensures that markers informative for specific populations, including low-frequency or private alleles, are not discarded due to varying population size or poor performance in other populations. After dosage calling and marker filtering, samples with ≤ 20% missing data were removed. Given the multi-allelic nature of microhaplotype genotypes, conventional minor allele frequency (MAF) was not an appropriate measure. Instead, a combined alternative allele frequency (AAF) was calculated and used to filter loci, with those having AAF below 0.05 excluded from further analysis. All filtering steps were performed under a hexaploid ploidy assumption, consistent with the known sweetpotato genome structure. The resulting microhaplotype dataset represented the union of high-quality loci across all projects and was used for principal component analysis (PCA) and discriminant analysis of principal components (DAPC) to evaluate population structure and assignment accuracy using the R packages vcfR (v1.15.0) and adegenet (v2.1.11) (Jombart 2008 ; Knaus and Grünwald 2017 ). Two SNP datasets were generated, target-SNPs and all-SNPs (target + off-target SNPs). For SNP analysis, loci absent from the final filtered microhaplotype dataset were first removed from all per-project MADC files to ensure that downstream SNP calls were drawn from the same set of genomic regions. Individual project SNP genotypes were generated using the readDArTag function with the parameter asSNPs set to TRUE in polyRAD (v2.0.0) package. The resulting SNP datasets were merged into a single SNP VCF under the same hexaploid model. Samples not present in the final filtered microhaplotype dataset were excluded from the concatenated SNP dataset, resulting in identical sample representation across data types, regardless of marker type (e.g., microhaplotype or SNP). Except removal of monomorphic sites, no additional filters were applied on the all-SNPs set to maintain comparability to microhaplotypes. Target-SNPs was obtained from the all-SNPs and underwent the same filters as microhaplotypes. All three marker sets, including microhaplotypes, all-SNPs, and target-SNPs, comprised of 4,087 matched samples, which were then processed through equivalent PCA and DAPC pipelines. Results Creation of an R Shiny interface for standardizing unique microhaplotype ID assignment We developed an R Shiny application, HapApp (v1.0), that provides a no-code option for users to process their MADC files, assign fixed microhaplotype IDs, and output an updated version of the microhaplotype FASTA file. The interface allows the user to customize the pipeline depending on their genotyping panel attributes such as species, microhaplotype length, and panel design (Fig. 1b). The most recent microhaplotype databases for each of the above-mentioned species (alfalfa, blueberry, cucumber, cranberry, potato pecan, strawberry, and sweetpotato) are available with the app, and updated versions retrievable by the user through the “Get Database” function in HapApp. Since the app directly uses the Bash/Python pipeline described above, our testing showed HapApp to be reliable for processing of MADC files in a no-code framework. Overview of microhaplotype database development For the DArTag panels of all eight species, marker loci were broadly distributed across all chromosomes with greater density in genic regions. As described in the Materials and Methods, eight microhaplotype databases were built following the standardized approach for unique ID convention, filtering, and appending new alleles to databases. Each database consists of RefMatch and AltMatch microhaplotypes in addition to Ref and Alt microhaplotypes (54–81 bp) by panel design. The eight species vary in genome complexity, heterozygosity, and ploidy levels (Table 1 ). It should be noted that not all microhaplotypes are orthologous to the assay design; some paralogous microhaplotypes may also be included, especially in species with high genome duplication rates or autopolyploidy. Here, we will first provide a brief description of microhaplotype databases for four crops (cranberry, cucumber, potato, and strawberry) and highlight the annotation, usage, and advantages of using microhaplotypes over SNPs for four other crops (alfalfa, blueberry, pecan, and sweetpotato) (Fig. 2 ). Cranberry microhaplotype database The cranberry microhaplotype database (v5) consists of 14,380 alleles from 3,050 loci (Fig. 2 ). These were genotyped from 4,146 samples. This represents the latest and most comprehensive version (cranberry 3K DArTag v2.0) of the database, with loci evenly spread across the 12 chromosomes. This version was built on improvements made to the initial v1 panel (Chen et al. 2025 ), addressing important issues of marker proximity and integrating new markers that cover the gapped genomic regions in the v1 panel. These refinements make the v2 panel a robust resource for studying cranberry genetics and cranberry breeding. The microhaplotype database for the cranberry v2 marker panel includes alleles from newly added markers and those retained in the v1 panel, ensuring that any previously validated information is not discarded. A total of 12,821 microhaplotypes from the retained v1 markers were integrated into the latest microhaplotype database. Summary statistics show that most loci have low levels of allelic variation (median = 3 alleles), while the high standard deviation arises from a subset of loci with extreme variability (Table 2 ). Our BLAST-based classification showed that these loci were dominated by off-target best hit (mapping outside the intended site) or putative paralogs (multiple equally good hits) rather than true alleles at a single locus (Table S2). For example, the most variable marker, chr11_035249665 (448 microhaplotypes), had 419 off-target best hits plus 16 putative paralogs (435/448; 97% putative non-allelic), and chr03_024644041 showed a similar pattern (113/131; 86%). In contrast, many loci exhibited clean single-locus behavior with all microhaplotypes mapping to the intended site and no putative paralogs (e.g., chr07_017980443; 64/64 design matches; chr01_018734260: 36/36; chr07_001562431: 32/32; chr09_008835521: 28/28). Together, these observations indicated that the large variance in microhaplotype counts is driven by a small number of multi-locus amplifications, while the majority of markers remain locus specific. We flagged these putative paralogous microhaplotypes in the database since they will likely occur in future genotyping results. Table 2 Summary statistics of the number of microhaplotypes per target marker locus for the eight microhaplotype databases Species #Loci Microhaplotype statistics mean Standard deviation minimum 25% 50% 75% maximum Alfalfa 3,000 12 9 2 6 10 15 160 Blueberry 3,000 10 11 2 4 7 11 270 Cranberry 3,050 5 9 2 2 3 5 448 Cucumber 3,059 3 6 2 2 2 4 256 Pecan 3,100 8 8 2 5 7 10 177 Potato 3,913 9 12 2 4 6 10 335 Strawberry 5,000 8 9 2 2 6 10 274 Sweetpotato 3,120 12 13 2 6 9 14 180 Cucumber microhaplotype database The cucumber microhaplotype database (v10) consists of 9,823 microhaplotypes derived from 8,272 genotyped samples (Fig. 2 ). These microhaplotypes originate from 3,059 target loci (manuscript in preparation), which are evenly distributed across all seven cucumber chromosomes. The number of alleles per locus had a mean of 3, with a standard deviation of 6, indicating uneven allele diversity across surveyed loci (Table 2 ). A few DArTag loci exhibited extremely high rates of polymorphism, with a maximum 256 alleles, which likely result from paralogous sequence amplification or hotspot regions of genetic variability. The cucumber microhaplotype database provides an essential resource for dissecting genomic and allelic diversity in cucumber. The genotyped samples were from several biparental F 1 and backcross populations, which resulted in the small number of alleles per locus (mean ~ 3.2). As no diverse collections have been genotyped using the cucumber marker panel, it is not yet feasible to evaluate the genetic diversity across the cucumber genome. Potato microhaplotype database The potato microhaplotype database (v14) is comprised of 35,506 microhaplotypes across 3,913 DArTag loci (Fig. 2 ), distributed across all 12 chromosomes of the potato reference genome DM6.1 (Pham et al. 2020 ) and a few trait markers that are not located in the DM6.1 genome (Endelman et al. 2024 ). These microhaplotypes were constructed from data across 3,102 samples. The number of alleles per locus had a mean of 9, with a standard deviation of 12, indicating a wide range of allele diversity across loci (Table 2 ). Notably, 50% of loci had six alleles (median value). A small number of loci displayed exceptionally high polymorphism rates, with a maximum of 335 alleles, potentially attributable to factors such as paralogous sequence amplification or regions of extraordinarily high genetic variability. The relatively high mean number of alleles per locus reflects substantial genetic diversity, which is consistent with the structural complexities of the potato genome and its propagation history. While clonal propagation suppresses recombination, it is highly effective at preserving existing diversity within an autotetraploid system. The wide range of allele counts emphasizes the mosaic distribution of diversity, highlighting both highly polymorphic regions and those with more limited variability across the genome. These insights will contribute to breeding programs, genetic mapping studies, and other research applications centered on potato population and genome dynamics. Strawberry microhaplotype database The strawberry microhaplotype database (v3) contains 38,227 microhaplotype sequences derived from 5,000 DArTag loci (Fig. 2 ) distributed across all 28 chromosomes (7 chromosomes with four sub-genomes: A, B, C, and D) (Hardigan et al. 2023 ). These microhaplotypes were derived from 1,880 samples. While 50% of loci exhibited six or fewer alleles (median value), 75% of loci had 10 or fewer alleles (Table 2 ). The number of alleles per locus had a mean of 8 and a standard deviation of 9, indicating notable variability in allele diversity among loci. The median value 6 indicated a right-skewed distribution driven by a small number of highly polymorphic loci (maximum allele count at 274). Such high allele counts likely result from genetic hotspots, paralogous amplification, or regions of high genetic variability. Alfalfa microhaplotype database The alfalfa microhaplotype database (v50) has reached a notable milestone, containing 35,259 microhaplotypes across 3,000 DArTag marker loci (Zhao et al. 2023 ), averaging ~ 12 alleles per locus. These microhaplotypes were discovered through 25 genotyping projects (> 15,600 samples) conducted by the alfalfa public breeding community, spanning diverse and structured polycross populations and breeding programs. The number of alleles per marker locus ranges from 2 to 160, with extended allele counts potentially signifying highly diverse genomic regions or paralogous amplification (Table 2 ). Allele diversity per locus is high compared to other species, as indicated by a mean of 12 alleles (SD = 9) and a median of 10 alleles. Approximately 25% of loci had 2 to 6 alleles, while 75% had 15 alleles or less. The trend in microhaplotype discovery shows rapid growth in earlier database versions, followed by a plateau around ~ 35,000 microhaplotypes despite the continued addition of samples (Fig. 2 ). This plateau suggests that the current marker set (3,000 loci) has effectively captured most genetic diversity at the 3K target loci among the U.S. alfalfa breeding populations. This plateau may also reflect database saturation, such that most of the Medicago sativa alleles existing across these loci have been captured. Notably, while microhaplotype discovery has plateaued, the database serves as a powerful resource for genomic analysis within cultivated alfalfa and its close relatives (Zhao et al. 2024c ). As is common in targeted genotyping platforms, not all resulting amplicons originate from the intended genomic locations due to sequence duplications within the genome. In polyploid species, the set of corresponding loci across homologous chromosomes are known as homologous groups (HGs), which share high sequence similarity and derive from the same ancestral sequences. HGs provide a framework for distinguishing true allelic variation from paralogous amplification. To this end, all 35K microhaplotype sequences were aligned to the four sub-genomes of the alfalfa cultivar ‘XinJiangDaYe’ (Chen et al. 2020 ). Syntenic regions across homologous chromosomes were used to determine the most likely syntenic alignment if a microhaplotype could be aligned to multiple loci on one chromosome. A total of 22,159 microhaplotypes from 2,314 DArTag loci (Figures S2 & S3; Table S3) are likely true allelic variants (i.e., match the panel design specification), ranging from 210 in HG6 to 426 in HG4 (Table S3). A total of 1,085 DArTag loci contained potential paralogous amplifications and on average, produced more microhaplotypes per locus than the 1,915 loci containing only true allelic variants (Fig. 3 a). Four DArTag loci with more than 100 microhaplotypes all exhibited paralogous amplifications. For example, locus chr3.1_010956413 was located within a gene ( MS.gene044613 ), which encodes a leucine-rich repeat and nucleotide-binding-ARC domain (NBS-LRR or NLR) disease-resistance protein (Fig. 3 b). Among the 1,915 DArTag loci with no evidence of paralogous amplifications, 12 have more than 30 alleles each (Fig. 3 b; Table S3), indicating these genomic regions possess relatively high genetic diversity and/or are not under strong selection pressure. For the 1,915 DArTag loci without evidence of paralogous amplification, we further examined their copy numbers, defined here as the number of homologous chromosomes in which each locus is present. Among these loci, 412 (21.5%) and 1,403 (73.3%) were found in three and all four homologous chromosomes, respectively, within each of the eight homologous groups, while 73 had two copies and 27 loci had only one copy (Figure S4). Although alfalfa is tetraploid, the observed variation in copy numbers across the genome underscores the complexity of its genomic structure. Additionally, loci with fewer copies (less than 4) are likely due, or at least in part, to incomplete genome assembly rather than true biological absence. Blueberry microhaplotype database The blueberry microhaplotype database (v20) is comprised of a total of 28,653 sequences generated from 95 genotyping plates (~ 8,930 samples) (Fig. 2 ). It represents a comprehensive characterization of genetic diversity across 3,000 target loci (Zhao et al. 2024b ), with a genomic insert size of about 54 bp (Table 2 ), evenly distributed on all 12 chromosomes. The number of microhaplotypes per locus demonstrated a mean of ~ 10 alleles, with a high variability indicated by a standard deviation of 11. Quartile statistics indicated that more than 50% of the loci contained 7 alleles or fewer. The minimum number of microhaplotypes per locus was two, while the maximum recorded was 270 alleles, showcasing a significant range in polymorphism across loci or paralogous amplifications for some loci. To characterize microhaplotypes potentially resulting from allelic vs. paralogous amplifications, all 28K microhaplotype sequences were aligned to the four sets of homologous chromosome sequences of the tetraploid blueberry cultivar ‘Draper’. Syntenic regions across homologous chromosomes were used to identify the most probable syntenic alignment when a microhaplotype matched multiple loci on the same chromosome. In total, 18,031 microhaplotypes from 2,535 DArTag loci (Figures S5 & S6; Table S4) were identified as true allelic variants. Across all 3,000 loci, 1,053 with potential paralogous amplifications tended to produce more haplotypes per locus than the 1,947 loci containing only allelic variants (Figure S7). Among the 1,947 loci without evidence of paralogous amplification, seven had more than 30 alleles each (Figure S7b; Table S4), with Chr03_005753680 and Chr07_033695728 having 82 and 81 alleles, respectively, followed by Chr05_003355218 with 52 allele (Figure S7b). Copy numbers of the 1,947 loci without paralogous amplification were further analyzed. Of these, 609 (31.3%) and 850 (43.7%) had three and four copies, respectively, across the 12 homologous groups, whereas 264 loci had two copies and 224 had only one copy (Figure S8). Notably, of the 224 loci with a single copy in the reference genome, 45 (20.1%) and 67 (29.9%) were located on chromosomes 4 (VaccDscaff6) and 12 (VaccDscaff22), respectively, which agrees with previous discovery that large genomic regions on these two chromosomes lack synteny (Colle et al. 2019 ). In addition, 66 out of 146 DArTag loci designed on chromosome 10 (VaccDscaff20) had three copies, reflecting that the distal region of chromosome 10 (VaccDscaff20) shares synteny with chromosomes 22 (VaccDscaff28) and 34 (VaccDscaff44), but not with chromosome 46 (VaccDscaff48) (Colle et al. 2019 ), while 52 of these 146 DArTag loci had four copies with three of them in HG10 but one copy in HG6 (chromosome 30/VaccDscaff19). Pecan microhaplotype database The pecan microhaplotype database (v9) is comprised of 26,073 alleles from 3,100 target loci (Chen et al.), derived from 6,768 genotyped samples (Fig. 2 ). The loci are evenly distributed across 16 chromosomes, and each of the 3,100 loci shows diverse allele counts. Most loci exhibit moderate levels of allelic diversity, with half of the loci containing seven or fewer alleles (Table 2 ). A small number of loci with extremely high allele counts may arise due to paralogous sequence amplification or hotspots of genetic variation. The complete pecan linkage map (Chen et al.) shows no large-scale structural variation (e.g., homologous sequence rearrangements) between the studied F₁ population and the reference genome. To compare the performance between microhaplotypes and SNPs, linkage maps were constructed for each of the three marker datasets: microhaplotypes, target-SNPs, and all-SNPs (target plus off-target SNPs). Figure 4. Comparison of marker ordering and filtering outcomes for pecan across three marker datasets (microhaplotypes, target SNPs, target plus off-target SNPs). (a) Summary of marker filtering by genotype class across the three datasets. Bars represent the number of markers per genotype class removed by specific filters applied in the presented order (starting with depth to rf_filter). (b) Linkage marker ordering for chromosome 8 using genome position (left panels) and multidimensional scaling (MDS) genetic distance (right panels). Filtering statistics further highlighted key differences among the three marker sets (Fig. 4a, Table S5). The all-SNPs dataset showed a high number of loci removed by the redundancy filter, which is expected given their proximity to target markers (within 81 bp). The microhaplotype dataset retained a larger number of markers capable of tracking recombination events and providing information for both parent’s meiosis. In contrast, the target-SNPs and all-SNPs datasets contained mostly ab × aa markers (single-parent informative) and very few ab × ab markers (informative for both parents) after filtering. These patterns were consistent across all 16 chromosomes, with chromosome 8 being the most prominent (Fig. 4b and S9). Therefore, in the target-SNPs and all-SNPs datasets, the aa × ab markers were frequently ordered independently and placed far from their expected genomic positions using the MDS method since not enough informative markers for both parents (heterozygous × heterozygous) were present to relate single-parent informative markers (aa × ab and ab × aa). The all-SNPs dataset showed a greater number of gaps when linkage maps were ordered according to the reference genome, indicating a higher incidence of genotyping errors among the retained markers. In contrast, the dataset containing only target SNPs demonstrated fewer gaps in the reference-based linkage map, suggesting lower genotyping error rates compared to the all-SNPs set. However, it contained relatively few markers informative for both parents, which impaired ordering performance using MDS algorithm. The microhaplotype dataset retained more loci throughout the filtering pipeline, had fewer gaps, and preserved a higher proportion of markers informative for both parents. These attributes contributed to more robust ordering when applying the MDS algorithm. Sweetpotato microhaplotype database The sweetpotato DArTag panel is a key molecular tool that was developed and applied in various collaborative projects to advance sweetpotato (2n = 6x = 90) genetics and breeding worldwide (Zhao et al. 2024a ). The latest version of the associated microhaplotype database (v19) consists of 37,593 sequences across 3,120 genomic loci, which are evenly distributed across the 15 chromosomes (Fig. 2 ). These microhaplotypes were derived from genotyping data of 9,212 individual samples. The number of alleles per locus ranges from 2 (minimum) to 180 (maximum), with an average of 12 alleles per locus (Table 2 ). The median number of alleles per locus is 9, and the interquartile ranges between 6 (25th percentile) and 14 (75th percentile). The wide distribution of alleles suggests the markers effectively capture broad genetic variations. The versatility and efficacy of the sweetpotato marker panel are evidenced by its extensive adoption across breeding programs from multiple continents. Participating programs include those in the U.S., Mozambique (Africa), Korea and Taiwan (Asia), Peru (South America), Jamaica (Central America), and Costa Rica (Caribbean/Central America) (Fig. 5A). Figure 5. Geographic context, cross validation accuracy and principal component analysis (PCA) of microhaplotypes and SNP markers in sweetpotato. (a) Geographic distribution of genotyping projects using the sweetpotato 3K DArTag panel. (b) Discriminant analysis of principal components (DAPC) cross-validation accuracies using 5–50 PCs for microhaplotypes, target-SNPs, and all-SNPs marker sets. (c and d) PCA plots for PC1 vs. PC2 (c) and PC2 vs. PC3 (d) of the microhaplotype dataset. (e and f) PCA plots for PC1 vs. PC2 (e) and PC2 vs. PC3 (f) of the target-SNPs dataset. (g and h) PCA plots for PC1 vs. PC2 (g) and PC2 vs. PC3 (h) of the all-SNPs dataset. We evaluated the performance of three marker sets on population genetics inference in sweetpotato: microhaplotypes, target-SNPs, and all-SNPs (target + off-target). To ensure high-quality genotype data while retaining informative low-frequency and private alleles, we implemented a population-aware filtering strategy in which a microhaplotype was retained if it met either a global missingness threshold of ≤ 20% across all samples, or a per-population missingness threshold of ≤ 20% in at least one population. After filtering for marker missingness and alternative allele frequency (AAF) of > 0.05, a total of 2,772 (27,220 microhaplotypes) out of 3,119 marker loci (28,338 microhaplotypes), and 4,087 samples with < 20% missing data were kept. The target-SNPs included the target SNPs present within the 2,772 retained loci. The same filters were applied to the target SNPs, resulting in 2,534 SNPs. The all-SNPs are comprised of all SNPs in the 2,772 marker loci, resulting in a total of 21,002 SNPs across 4,087 samples. After removing monomorphic sites in all-SNPs, 20,935 markers remained. Because these SNPs were extracted directly from the retained microhaplotypes, no additional filtering was applied beyond removal of monomorphic sites. Table 3 Project-level private microhaplotypes and marker information in sweetpotato Projects #Samples #Private_microhaplotypes #Markers with private microhaplotypes Costa Rica 165 261 140 Jamaica 72 2 2 Mozambique 155 27 27 Peru 794 274 214 South Korea 191 440 364 Taiwan 1,352 7,340 2,233 USA 1,358 367 308 To evaluate how well the retained panel captures population-specific variation, we enumerated project-specific microhaplotypes among the retained loci (Table 3 ). Across the seven projects, we identified 8,711 project-specific microhaplotypes, spanning 3,288 marker-project instances in which at least one private microhaplotype occurred. Normalized by sample size, private microhaplotypes per sample were 0.03 (2/72) in Jamaica, 0.17 (27/155) in Mozambique, 0.27 (367/1,358) in the USA, 0.35 (274/794) in Peru, 1.58 (261/165) Costa Rica, 2.3 (440/191) in South Korea, and 5.43 (7,340/1,352) in Taiwan. These results indicate substantial heterogeneity in project-private variant across populations, with Taiwan and South Korea showing the highest densities. This pattern is consistent with the stronger separations observed in the PCA (below). Our comparative analysis revealed that microhaplotypes consistently provided a more efficient and stable representation of population structure than either SNP set. Principal component analysis (PCA) revealed marked differences in the resolution of population structure (Fig. 5c-h). In PC1-PC2 space, the microhaplotype PCA displayed visibly tighter clusters for each population with minimal overlaps, with Taiwan strongly differentiated along PC1 and subtler separations among other populations along PC2. In contrast, the SNP-based PCA exhibited less distinct separation with noticeable overlap among several populations, particularly Costa Rica, Mozambique, South Korea, and Taiwan. Similar trends were observed in PC2-PC3 space. The variance explained by the leading PCs further highlighted these differences. For microhaplotypes, PC1, PC2, and PC3 explained 15.72%, 3.45%, and 2.3% of the total variance, respectively, compared to 10.5%, 7.56%, and 2.6% for target-SNPs, and 5%, 4.24%, and 1.76% for all-SNPs. Therefore, PC1 of the microhaplotypes captured more variance compared with the SNP equivalents, suggesting stronger separation along the most informative axis. Table 4 Eigenvalue distribution and proportion of variance explained by the three marker sets and linear discriminant (LD) axis in sweetpotato populations. Axis eigenvalue Fractions Microhaplotypes All_SNPs Target_SNPs Microhaplotypes All_SNPs Target_SNPs LD1 854,002 215,300 112,056 86.26 60.61 60.28 LD2 74,596 89,897 47,587 7.53 25.31 25.6 LD3 41,989 35,015 16,007 4.24 9.86 8.61 LD4 16,736 12,488 8,395 1.69 3.52 4.52 LD5 1,921 1,731 1,237 0.19 0.49 0.67 LD6 805 784 618 0.08 0.22 0.33 Discriminant analysis of principal components (DAPC) further highlighted the discriminatory advantage of microhaplotypes (Table 4 ). Cross validation showed that 86.3% of between-group variance was captured by the first linear discriminant axis (LD1) for microhaplotypes compared with approximately 60% for both SNP sets, an improvement of over 25%. The LD1 eigenvalue was correspondingly larger for microhaplotypes (854,002) than for all-SNPs (215,300), and target-SNPs (112,056), reflecting tighter cluster resolution. Despite these differences in dimensional efficiency, all three marker types achieved very high classification accuracies (mean success rate > 98%; ANOVA: F = 1.495, p = 0.225) (Fig. 5b). At low dimensionality (5 PCs), targeted-SNPs achieved the highest early success rate (84.3%), followed by microhaplotypes (80.2%) and all-SNPs (77.7%), likely reflecting the pre-selected informativeness of target SNPs. As the number of retained PCs increased, target-SNPs accuracy degraded starting around 40 PCs whereas microhaplotypes maintained a stable high-success plateau near 99%, suggesting greater robustness to model complexity. Comparative insights across all eight databases Microhaplotypes are sets of closely linked SNP sites within a small region of the genome (assumed not to recombine), making them a powerful tool for genotyping and genetic diversity studies. The development of microhaplotype databases across multiple crops has enabled the discovery of rich allelic diversity while providing the plant germplasm, breeding, and genetics community with resources to dissect functional diversity, population structure, and trait architecture using cost-effective targeted genotyping. The microhaplotype databases for these eight species were developed with varying numbers of genotyped samples (Table 1 ). Factors such as the scale of genotyping, species-specific genome complexity, marker diversity, and the genomic insert size used during genotyping library preparation impacted the resulting resources. Unsurprisingly, longer genomic insert sizes (e.g., 81 bp) tend to capture more variant sites compared to shorter insert sizes (e.g., 54 bp), allowing for more variants to be detected per amplicon locus. For example, species like sweetpotato and pecan, which utilize 81 bp panels, show substantial allele diversity relative to their genomic complexity, partly because of the longer genomic region available for detecting polymorphisms. All eight microhaplotype databases were developed by following a standardized pipeline that includes stringent ID conventions, filtering, and sequential updates to maintain biological relevance. Notably, the databases include both orthologous (i.e., microhaplotypes originated from conserved loci between divergent species) and paralogous (i.e., related loci within a genome, product of duplication) microhaplotypes, the latter of which are more common in complex genomes with higher levels of duplications. The retention of paralogous alleles may seem counterintuitive at first glance, since they cannot be used in breeding. However, because these alleles are likely to be encountered in future genotyping experiments, it was important to codify them and append a confidence rating that indicates to the users that they are likely paralogous alleles. Among the databases, alfalfa (autotetraploid), sweetpotato (hexaploid), and blueberry (autotetraploid) exhibit the highest genetic diversity, with average allele counts per locus of 12, 12, and 10, respectively. The polyploid nature of these crops likely contributes to their exceptional allelic richness. Furthermore, these three species also represent the largest genotyping scales undertaken on the respective DArTag panels. The alfalfa database, based on 167 plates of genotyped samples (over 15,600 samples), stands out as the most extensive. Alfalfa’s saturation plateau (~ 35,000 microhaplotypes) demonstrates that the high genotyping scale successfully captures most genetic diversity at the 3K target loci within the studied populations. For sweetpotato, the global importance, spanning Africa, Asia, and the Americas, is mirrored in the diversity captured in its database, which encompasses samples from 9,212 samples and aligns with a worldwide population structure. In contrast, cucumber exhibited the narrowest genetic base among the species examined. With a mean of 3 alleles per locus and 75% of loci containing no more than 4 alleles, which likely reflects sampling from a single breeding program and/or inherently limited diversity in cucumber relative to the other species analyzed. Consistent with this, prior work indicates that wild sexually compatible relatives could enhance the genetic diversity of cultivated cucumber (Zhang et al. 2017 ; Abdel-Salam et al. 2020 ; Tirnaz et al. 2022 ). On the other hand, the octoploid strawberry database reveals rich allele diversity with a mean of 8, despite a relatively smaller genotyping scale (20 plates). The high allelic diversity evidenced here traces back to the selection of sub-genome-specific loci for developing the marker panel (Hardigan et al. 2023 ). Some additional allelic richness may also be attributed to the complex genome structure of strawberry, which may compensate for the smaller sample size. Data accessibility and community engagement In alignment with the FAIR principles ( https://www.go-fair.org/fair-principles/ ), all microhaplotype sequences will be made publicly available. The allele data, metadata (including project and principal investigator (PI) information, and associated documentation will be hosted through globally accessible and FAIR-compliant platforms, HapApp ( https:github.com:Breeding-Insight/HapApp_utils ) and HapSearch (under development). Users who are interested in obtaining samples or germplasm linked to specific alleles must follow established access procedures. Interested users will need to contact the relevant PIs and adhere to the AgCommons Framework ( https://agdatacommons.nal.usda.gov/ ), or a similar standard protocol for germplasm access. The PI has the right to share the resources while respecting intellectual property rights, mutually agreed terms, and proper documentation. Discussion The eight curated microhaplotype databases developed here provide a powerful, standardized framework for advancing genetic diversity studies and breeding programs across a diverse set of crops. Through consistent microhaplotype ID conventions, filtering pipelines, and sequential updates, these resources capture both orthologous and documented paralogous microhaplotypes, ensuring broad relevance to future genotyping. Microhaplotypes deliver richer allelic diversity and greater resolution than bi-allelic SNPs, a benefit especially evident in complex, polyploid genomes such as alfalfa, blueberry, and sweetpotato. Our results demonstrate their advantages for retaining informative markers, improving linkage map ordering, and resolving fine-scale population structure. Orthologous and paralogous microhaplotypes While most loci exhibited low to moderate microhaplotype counts, the extremely high allele counts observed at some loci are likely due to paralogous sequence amplification, in addition to natural orthologous genetic diversity. Paralogous sequences arise from duplicated regions within the genome, which can be unintentionally captured and amplified during genotyping due to sequence similarity. These loci can be important for identifying hotspots of variation, but they may also reflect technical artifacts rather than true orthologous diversity. Comparative patterns in alfalfa and blueberry microhaplotypes illustrated how genome structure shapes these metrics. Though both species have a similar number of loci without paralogous amplification (1,915 and 1,947 loci, respectively), their copy number distributions differ (Tables S1 & S2). In alfalfa, 94.8% of the non-paralogous loci occur in three or four copies (Figure S4), matching the expected dosage for a largely autotetraploid species. In contrast, only 74.9% of blueberry loci fall into this range. Instead, blueberry shows substantially higher proportions of single-copy (20.1%) and two-copy (29.9%) loci (Figure S8). This elevated number of reduced-copy loci in blueberry suggests more extensive post-whole genome duplication structural modifications than that observed in alfalfa. These patterns are consistent with current evidence that highbush blueberry originated through a complex polyploidization process involving hybridization among multiple diploid Vaccinium species followed by genome duplication and subsequent structural divergence (Colle et al. 2019 ; Mengist et al. 2023 ). The cultivar ‘Draper’ is predominantly V. corymbosum with minor introgressions (< 5%) from V. tenellum , V. ashei , and V. darrowii , which may help explain part of the observed pattern (Hancock 2004 ). The unique inter-chromosome translocation between chromosomes 6 and 10 could affect pairing and recombination of these two chromosomes in ‘Draper’ but was not observed in a wild diploid relative ‘W85’ (Mengist et al. 2023 ) or reported in other tetraploid highbush blueberries. Most repetitive sequences in the alfalfa ‘XinJiangDaYe’ reference genome are composed of long terminal repeat (LTR) retrotransposons, consistent with previous observations in other legume species (Chen et al. 2020 ). However, because DArTag markers are predominantly designed within genic regions, most cases of paralogous amplification in our dataset are not likely attributed to LTR-derived repeats. Notably, we identified a highly paralogous marker (chr3.1_010956413) with 105 microhaplotypes (rank of 3) located within MS.gene044613 , an NLR disease resistance protein. This observation aligns with earlier findings in M. truncatula , a wild relative of cultivated alfalfa, in which a dense cluster of NLR genes was reported on chromosome 3.1 (Ameline-Torregrosa et al. 2008 ). Together, these results suggest that local gene family expansion could play a role in generating paralogous signals in DArTag genic markers. Microhaplotypes strengthen linkage maps through high information content and reduced phasing errors Comparison of linkage maps created using microhaplotypes vs. SNPs highlighted the advantages of microhaplotypes in providing higher information content and reducing ambiguity in phase estimation (Fig. 4). Because target and off-target SNPs fall within a 54–81 bp read, variants within each amplicon are physically linked and phased directly from reads, making recombination within an amplicon effectively negligible. Thus, part of the advantage of microhaplotypes reflecting collapsing these co-inherited SNPs into a single multi-allelic locus, which reduces redundancy and increases information content. Their performance depends primarily on read depth and phasing quality rather than population LD. Due to their multi-allelic nature, microhaplotypes increased the number of markers heterozygous in both parents, thereby strengthening linkages among single-dose variant classes such as ab × aa and aa × ab. In bi-allelic datasets, the rare ab × ab class was the sole marker type capable of connecting these groups, limiting integration of parental maps and reducing overall connectivity. Consequently, both target-SNPs and all-SNPs (target plus off-target) performed poorly during marker ordering due to the limited number of markers informative for recombination in both parents. Microhaplotypes also reduce ambiguity in phase estimation. For multi-allelic configurations such as ab × cd, the parental origin of progeny alleles is directly observable, decreasing the reliance on Expectation Maximization (EM) inference and lowering the probability of phasing errors. In contrast, bi-allelic datasets contain ambiguous configurations (e.g., ab × ab) requiring EM inference, increasing the risk of incorrect phasing. The all-SNPs set showed the poorest mapping performance, with incorrect ordering for most chromosomes and a higher proportion of erroneous markers retained after filtering. This was likely due to the physical proximity of off-target sites to their corresponding target sites, resulting in their removal as redundant during early filtering. The off-target markers that were not removed tend to be more error-prone, and if they are not eliminated by subsequent filters such as segregation distortion, they contribute to the larger number of gaps observed in the final maps. Problematic markers can be filtered out by increasing the stringency of the rf_filter parameter, but good-performing markers can also be removed, especially in datasets with few multi-allelic or highly informative markers. In contrast, the microhaplotype strategy integrates off-target information into multi-allelic markers, which reduces redundancy and enhances the effectiveness of downstream filtering. This highlights the need for careful parameter tuning, especially in datasets with fewer multi-allelic or highly informative markers. Microhaplotypes provide more efficient and informative separation of populations than SNPs Microhaplotypes inherently capture multiple variants within a small genomic region and retain linkage phase information, which increases their average heterozygosity and allelic count per locus compared to SNPs. These properties underpin their growing adoption not only in human forensic and ancestry inference applications but also in the plant genetics and breeding realm. The comparative analysis of microhaplotype and SNP data analyzed independently in hexaploid sweetpotato revealed notable differences in allelic diversity, variance captured, and population resolution. The resulting extra diversity using microhaplotypes means that patterns differentiating populations are stronger per locus, which is consistent with findings from other systems. In human genetics, highly polymorphic microhaplotype panels have demonstrated strong performance in relationship detection. For example, a set of 54 microhaplotypes demonstrated high reliability for first-degree relationship detection, approaching the performance of an established forensic panel comprising 27 short tandem repeats (STRs) plus 94 SNPs (Wu et al. 2021 ). Complementary evidence from ancestry inference research identified 120,000 microhaplotypes from nearly one million SNPs and across multiple benchmarks, microhaplotype subsets consistently outperformed SNP panels in resolving ancestry, particularly in complex or admixed populations (Turchi et al. 2026 ). Taken together, these cross-disciplinary insights point toward the value of fully utilizing microhaplotypes in plant breeding, particularly highly complexed polyploid species. These advantages are directly relevant for dosage-aware genomic prediction in polyploids. By encoding each microhaplotype as a copy-number dosage (0–2 for diploids; 0–4 for tetraploids, etc.), multi-allelic loci provide richer predictors than biallelic SNPs, capturing local microhaplotype effects and within-locus interactions. Such dosages can be used to build microhaplotype-based relationship matrices for GBLUP or included as regressors in Bayesian whole-genome model to capture more additive variance and reduce phase-related ambiguity compared with individual SNPs. From a methodological perspective, the implementation of a population-aware filtering strategy addresses a challenge in population genetics by ensuring high data quality without discarding private alleles. This way, we preserved genetic signals relevant for local adaptation, breeding history, and germplasm differentiation even when these microhaplotypes were scarce across the global dataset. One advantage of applying this strategy is the preservation of thousands of private microhaplotypes in the sweetpotato Taiwan population, which allowed its clear separation from other populations. The distinct clustering of Taiwanese accessions in the PCA may reflect the long history of sweetpotato breeding in Taiwan. Repeated cycles of crossing and selection within local breeding programs, together with the use of shared founder lines, may have resulted in a relatively distinct breeding pool. In addition, region-specific breeding objectives and relatively limited germplasm exchange with other regions may have further contributed to divergence in allele frequencies. The utility of the microhaplotypes has also been demonstrated in alfalfa by others. In a study of 28 populations, Medina et al. ( 2025 ) revealed significant population structure based on geographical origin, higher genetic diversity values compared to traditional SNP markers, excess heterozygosity (negative FIS values) in 27 out of 28 populations, and minimal inbreeding in founding populations (Medina et al. 2025 ). Another important application of microhaplotypes is for curation and management of plant genetic resource (PGR) collections. They can be used to: (i) resolve population structure and quantify diversity within and among collections; (ii) identify redundancy and gaps to optimize regeneration, safety-duplication, and sampling strategies; (iii) verify accession identity and detect mixtures or mislabeling to improve passport data and taxonomic assignments; and (iv) define representative core subsets for evaluation and pre-breeding. The standardized IDs and cross-project comparability enable genebanks to benchmark collections across repositories and over time, directly supporting trait discovery and the deployment of novel variation into breeding pipelines. Implications of microhaplotype analysis and future directions Looking forward, the integration of microhaplotypes into both basic and applied research has the potential to markedly improve analytical efficiency and resolution. The efficiency of microhaplotype-based analyses in low-dimensional space makes them particularly attractive for targeted genotyping in resource-limited breeding programs, where computational capacity and sequencing resources are constrained. At present, relatively few tools can fully accommodate multi-allelic data, particularly in complex polyploid species (Table 5 ). Expanding the availability and performance of software that can handle multi-allelic datasets in both diploid and polyploid contexts will greatly accelerate the integration of microhaplotypes into breeding decision-making pipelines. Such tool development will enable breeders and researchers to fully leverage the linkage phase information and high allelic diversity inherent to microhaplotype markers. Table 5 List of computational tools available with multi-allelic marker support. Tool Name a Analysis Support Polyploids Citation Polysat Population Diversity Yes (Clark and Jasieniuk 2011 ; Clark and Schreier 2017 ) PLINK 2.0 b HWE, LD, Relationship matrix, GWAS No (Chang et al. 2015 ) polyRAD Genotype Calling Yes (Clark et al. 2019 ) RAINBOWR GWAS No (Hamazaki and Iwata 2020 ) mpQTL GWAS Yes (Thérèse Navarro et al. 2022 ) OneMap Linkage Mapping No (Taniguti et al. 2022 ) ADAM-multi Population Simulation Yes (Chu and Jensen 2025 ) a All described tools also support highly heterozygous species. b As of January 10, 2026 release The current framework is designed to be species-agnostic and expandable. With HapApp, users can create, update, and maintain species-specific microhaplotype databases for taxa not included in this study by processing MADC files, assigning fixed microhaplotype IDs, and producing updated database. To make the microhaplotype databases more amenable, we are actively developing a user-friendly platform, HapSearch, which could serve as an interactive bridge between raw genomic information and practical breeding decisions. With HapSearch, researchers and breeders can rapidly locate, evaluate, and act upon relevant allelic variations captured in the databases. The envisioned features of HapSearch would enhance the value of microhaplotype resources for plant breeding and genetics. First, the system will allow targeted mining of alleles of interest by crop species, sequence similarity, or project-specific keywords. Second, integrated filtering functions will enable users to narrow search results to specific locus. Multiple sequence alignment will reveal all existing variations for a locus and allow users to identify accessions carrying unique microhaplotype alleles. Third, to promote collaboration and resource exchange, the platform could provide access to PI or germplasm collection curator contact information, enabling rapid follow-up for germplasm acquisition or data verification. The suite of platforms from processing and maintaining microhaplotype databases to user-friendly system for mining microhaplotypes will greatly accelerate the translation of genotypic diversity into practical genetic gain. Conclusion We present standardized microhaplotype databases for eight crops, a species‑agnostic, reproducible pipeline, and a no‑code HapApp for assigning fixed IDs and updating records from a targeted amplicon genotyping platform, DArTag. HapApp can also be applied to other targeted genotyping platforms (e.g., GT-seq, AgriSeq, FlexSeq) by pre-processing the raw FASTQ data into MADC-like format. These resources capture extensive allelic diversity, including orthologous and paralogous signals, and provide higher resolution than SNPs for population structure, linkage mapping, and genotype-phenotype analyses. By retaining more informative markers and reducing phasing ambiguity, microhaplotypes are especially powerful in polyploid and highly heterozygous crops and remain efficient for resource‑limited programs. The databases are FAIR and extensible; community updates and the forthcoming HapSearch interface will make allele mining and cross‑project comparisons routine, strengthening PGR curation and accelerating genetic gain across crops. Declarations Funding This research was supported through Breeding Insight (RRID: SCR_026645), a USDA-ARS initiative previously hosted by Cornell University under Cooperative Agreements ( 8062-21000-043-004-A , 8062-21000-052-002-A , and 8062-21000-052-003-A ) and currently hosted at the University of Florida, Gainesville, under a Cooperative Agreement (8062-21000-052-020-A). Additional support was provided by USDA ARS CRIS Project 2072-21000-059-000D and by the NIFA Alfalfa Seed and Alfalfa Forage Systems Research Program (awards 2019-70005-30361 and 2023-70005-41081 to ECB, BI, HR); the CGIAR Trust Fund contributors (https://www.cgiar.org/funders/ to RS, MK, HLK, and MD); the Bill and Melinda Gates Foundation through the SweetGAINs (OPP1213329) and the RTB breeding (INV-0411050) investments (GCY and SPF); the Cooperative Project No. (to be assigned) Rural Development Administration, Republic of Korea. Disclaimer This work was supported by USDA-ARS Projects 5062-21500-001-000D, 2019-70005-30361 and 2023-70005-41081. Mention of any trade names or commercial products in this article is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the U.S. Department of Agriculture. USDA is an equal opportunity provider and employer, and all agency services are available without discrimination. Competing Interests The authors have no relevant financial or non-financial interests to disclose. Author Contributions DZ: Conceptualization; Data curation; Formal Analysis; Investigation; Methodology; Project administration; Software; Supervision; Validation; Visualization; Writing – original draft preparation. ML: Data curation; Formal Analysis; Investigation; Methodology; Visualization; Writing – original draft preparation. CHT: Data curation; Formal Analysis; Investigation; Methodology; Visualization; Writing – original draft preparation. AMS: Software; Visualization; Writing – original draft preparation. SC: Visualization. DAS: Funding acquisition; Investigation; Resources. ZX, ECB, BMI, HR, DAS, DS, NB, SJC, LMH, EB, JP, JHP, JL, MH, SPF: Resources. CAM: Methodology. XW and AH: Investigation; Resources; Validation. WC: Funding acquisition; Investigation; Resources; Supervision; Validation. PAW: Funding acquisition; Resources. GCY: Funding acquisition; Resources; Investigation; Project administration; Supervision; Validation; SAW: Investigation; Resources; Project administration; Supervision; Validation; THK: Funding acquisition; Resources; RS: Resources; Investigation; Methodology; MK: Conceptualization; Resources; Methodology. MD: Conceptualization; Resources; Investigation; Methodology. HLK: Funding acquisition; Resources; Project administration; Supervision. SYC: Resources; Investigation; Project administration; Supervision. PHL: Resources; Investigation; Project administration. CWL: Resources; Investigation. JRC: Funding acquisition; Resources; Project administration; Supervision. FJM: Resources; Investigation. RCV: Resources; Investigation. CTB and MJS: Funding acquisition; Project administration; Supervision. All authors participated in manuscript Writing – editing. All authors read and approved the final version of the manuscript. Acknowledgements We gratefully acknowledge our collaborators for generously sharing germplasm for genotyping and for contributing metadata, coordination, and logistical support essential to the development of the microhaplotype database. Jeff Neyhart, Juan Zalapa, and Eric Weisman for their work with cranberry; Yiqun Weng and Savannah Beyer for their work with cucumber; Max Feldman, Noelle Anglin, Xiaohong Wang, Mercedes Ames, and Dennis Halterman for their work with potato; and the USDA-ARS Corvallis, OR location for their support with strawberry. Their openness in sharing material and data made this work possible. We also sincerely thank them for their years of partnership and collaboration with Breeding Insight, which continues to advance, innovate, and provide impact across our shared breeding communities. Data Availability Scripts and pipeline for building microhaplotype databases are deposited at GitHub (https://github.com/Breeding-Insight/HapApp_utils; https://github.com/Breeding-Insight/HapApp). The microhaplotype database files of the eight crop species can be downloaded from Zenodo (https://doi.org/10.5281/zenodo.19338534). References Abdel-Salam EM, Faisal M, Alatar AA, et al (2020) Comparative analysis between wild and cultivated cucumbers reveals transcriptional changes during domestication process. Plants 9:63. https://doi.org/10.3390/plants9010063 Abed A, Belzile F (2019) Comparing single-SNP, multi-SNP, and haplotype-based approaches in association studies for major traits in barley. Plant Genome 12:1–14. https://doi.org/10.3835/plantgenome2019.05.0036 Ameline-Torregrosa C, Wang B-B, O’Bleness MS, et al (2008) Identification and characterization of nucleotide-binding site-leucine-rich repeat genes in the model plant Medicago truncatula. Plant Physiol 146:5–21. https://doi.org/10.1104/pp.107.104588 Anderson EC, Clemento AJ, Campbell MA, et al (2025) A multipurpose microhaplotype panel for genetic analysis of California Chinook salmon. Evol Appl 18:e70110. https://doi.org/10.1111/eva.70110 Baetscher DS, Clemento AJ, Ng TC, et al (2018) Microhaplotypes provide increased power from short-read DNA sequences for relationship inference. Mol Ecol Resour 18:296–305. https://doi.org/10.1111/1755-0998.12737 Blyton MDJ, Brice KL, Heller-Uszynska K, et al (2023) A new genetic method for diet determination from faeces that provides species level resolution in the koala. https://doi.org/10.1101/2023.02.12.528172 Botstein D, White RL, Skolnick M, Davis RW (1980) Construction of a genetic linkage map in man using restriction fragment length polymorphisms. Am J Hum Genet 32:314–331 Camacho C, Coulouris G, Avagyan V, et al (2009) BLAST+: architecture and applications. BMC Bioinformatics 10:421. https://doi.org/10.1186/1471-2105-10-421 Campbell NR, Harmon SA, Narum SR (2015) Genotyping-in-Thousands by sequencing (GT-seq): A cost effective SNP genotyping method based on custom amplicon sequencing. Mol Ecol Resour 15:855–867. https://doi.org/10.1111/1755-0998.12357 Chang CC, Chow CC, Tellier LC, et al (2015) Second-generation PLINK: rising to the challenge of larger and richer datasets. Gigascience 4:7. https://doi.org/10.1186/s13742-015-0047-8 Chang W, Cheng J, Allaire JJ, et al (2024) shiny: Web Application Framework for R Chen H, Zeng Y, Yang Y, et al (2020) Allele-aware chromosome-level genome assembly and efficient transgene-free genome editing for the autotetraploid cultivated alfalfa. Nat Commun 11:1–11. https://doi.org/10.1038/s41467-020-16338-x Chen S, Lin M, Taniguti CH, et al (2025) A public mid-density genotyping platform for cultivated cranberry (Vaccinium macrocarpon Aiton). Plant Genome 18:e70118. https://doi.org/10.1002/tpg2.70118 Chen S, Lin M, Zhao D, et al A public mid-density genotyping platform for pecan (Carya illinoinensis (wangenh.) Koch). In preparation Chu TT, Jensen J (2025) ADAM-multi: software to simulate complex breeding programs for animals and plants with different ploidy levels and generalized genotypic effect models to account for multiple alleles. Front Genet 16:1513615. https://doi.org/10.3389/fgene.2025.1513615 Clare SJ, Driskill M, Millar TR, et al (2024) Development of a targeted genotyping platform for reproducible results within tetraploid and hexaploid blueberry. Front Hortic 2:1339310. https://doi.org/10.3389/fhort.2023.1339310 Clark LV, Jasieniuk M (2011) POLYSAT: an R package for polyploid microsatellite analysis. Mol Ecol Resour 11:562–566. https://doi.org/10.1111/j.1755-0998.2011.02985.x Clark LV, Lipka AE, Sacks EJ (2019) PolyRAD: Genotype calling with uncertainty from sequencing data in polyploids and diploids. G3 (Bethesda) 9:663–673. https://doi.org/10.1534/g3.118.200913 Clark LV, Schreier AD (2017) Resolving microsatellite genotype ambiguity in populations of allopolyploid and diploidized autopolyploid organisms using negative correlations between allelic variables. Mol Ecol Resour 17:1090–1103. https://doi.org/10.1111/1755-0998.12639 Colle M, Leisner CP, Wai CM, et al (2019) Haplotype-phased genome and evolution of phytonutrient pathways of tetraploid blueberry. Gigascience 8:giz012. https://doi.org/10.1093/gigascience/giz012 Cruz VMV, Kilian A, Dierig DA (2013) Development of DArT marker platforms and genetic diversity assessment of the U.S. collection of the new oilseed crop lesquerella and related species. PLoS One 8:e64062. https://doi.org/10.1371/journal.pone.0064062 de Barros Rodrigues ML, Rodrigues MP, Norton HL, et al (2025) Large-scale selection of highly informative microhaplotypes for ancestry inference and population specific informativeness. Forensic Sci Int Genet 74:103153. https://doi.org/10.1016/j.fsigen.2024.103153 Delomas TA, Hollenbeck CM, Matt JL, Thompson NF (2024) Microhaplotypes generate higher breeding value accuracy compared to SNPs for imputation-based breeding strategies. Aquaculture 586:740779. https://doi.org/10.1016/j.aquaculture.2024.740779 Diaz-Garcia L, Garcia-Ortega LF, González-Rodríguez M, et al (2021) Chromosome-Level Genome Assembly of the American Cranberry (Vaccinium macrocarpon Ait.) and Its Wild Relative Vaccinium microcarpum. Front Plant Sci 12:633310. https://doi.org/10.3389/fpls.2021.633310 Elshire RJ, Glaubitz JC, Sun Q, et al (2011) A robust, simple genotyping-by-sequencing (GBS) approach for high diversity species. PLoS One 6:e19379. https://doi.org/10.1371/journal.pone.0019379 Endelman JB, Kante M, Lindqvist-Kreuze H, et al (2024) Targeted genotyping-by-sequencing of potato and data analysis with R/polyBreedR. Plant Genome e20484. https://doi.org/10.1002/tpg2.20484 Gabriel SB, Schaffner SF, Nguyen H, et al (2002) The structure of haplotype blocks in the human genome. Science 296:2225–2229. https://doi.org/10.1126/science.1069424 Garrison E, Marth G (2012) Haplotype-based variant detection from short-read sequencing. arXiv [q-bio.GN] Gerard D, Ferrão LFV (2020) Priors for genotyping polyploids. Bioinformatics 36:1795–1800. https://doi.org/10.1093/bioinformatics/btz852 Hamazaki K, Iwata H (2020) RAINBOW: Haplotype-based genome-wide association study using a novel SNP-set method. PLoS Comput Biol 16:e1007663. https://doi.org/10.1371/journal.pcbi.1007663 Hancock JF (2004) Blueberry plant denominated ‘Draper.’ US Patent Hardigan MA, Feldmann MJ, Carling J, et al (2023) A medium-density genotyping platform for cultivated strawberry using DArTag technology. Plant Genome 16:e20399. https://doi.org/10.1002/tpg2.20399 Hardigan MA, Feldmann MJ, Pincot DDA, et al (2021) Blueprint for phasing and assembling the genomes of heterozygous polyploids: Application to the octoploid genome of strawberry. bioRxiv 2021.11.03.467115 Jombart T (2008) adegenet: a R package for the multivariate analysis of genetic markers. Bioinformatics 24:1403–1405. https://doi.org/10.1093/bioinformatics/btn129 Kidd KK, Pakstis AJ, Speed WC, et al (2013) Microhaplotype loci are a powerful new type of forensic marker. Forensic Sci Int Genet Suppl Ser 4:e123–e124. https://doi.org/10.1016/j.fsigss.2013.10.063 Kidd KK, Pakstis AJ, Speed WC, et al (2014) Current sequencing technology makes microhaplotypes a powerful new type of genetic marker for forensics. Forensic Sci Int Genet 12:215–224. https://doi.org/10.1016/j.fsigen.2014.06.014 Knaus BJ, Grünwald NJ (2017) vcfr: a package to manipulate and visualize variant call format data in R. Mol Ecol Resour 17:44–53. https://doi.org/10.1111/1755-0998.12549 Langmead B, Salzberg SL (2012) Fast gapped-read alignment with Bowtie 2. Nat Methods 9:357–359. https://doi.org/10.1038/nmeth.1923 Lovell JT, Bentley NB, Bhattarai G, et al (2021) Four chromosome scale genomes and a pan-genome annotation to accelerate pecan tree breeding. Nat Commun 12:4125. https://doi.org/10.1038/s41467-021-24328-w Mammadov J, Aggarwal R, Buyyarapu R, Kumpatla S (2012) SNP markers and their impact on plant breeding. Int J Plant Genomics 2012:728398. https://doi.org/10.1155/2012/728398 Martin M (2011) Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet.journal 17:10–12 Martin M, Patterson M, Garg S, et al (2016) WhatsHap: fast and accurate read-based phasing. bioRxiv Matias FI, Galli G, Correia Granato IS, Fritsche-Neto R (2017) Genomic prediction of autogamous and allogamous plants by SNPs and haplotypes. Crop Sci 57:2951–2958. https://doi.org/10.2135/cropsci2017.01.0022 Medina CA, Zhao D, Lin M, et al (2025) Pre-breeding in alfalfa germplasm develops highly differentiated populations, as revealed by genome-wide microhaplotype markers. Scientific Reports 15:1253 Mengist MF, Bostan H, De Paola D, et al (2023) Autopolyploid inheritance and a heterozygous reciprocal translocation shape chromosome genetic behavior in tetraploid blueberry (Vaccinium corymbosum). New Phytol 237:1024–1039. https://doi.org/10.1111/nph.18428 Mrázek J, Guo X, Shah A (2007) Simple sequence repeats in prokaryotic genomes. Proc Natl Acad Sci U S A 104:8472–8477. https://doi.org/10.1073/pnas.0702412104 Oldoni F, Podini D (2019) Forensic molecular biomarkers for mixture analysis. Forensic Sci Int Genet 41:107–119. https://doi.org/10.1016/j.fsigen.2019.04.003 Pham GM, Hamilton JP, Wood JC, et al (2020) Construction of a chromosome-scale long-read reference genome assembly for potato. Gigascience 9:giaa100. https://doi.org/10.1093/gigascience/giaa100 Preedy KF, Hackett CA (2016) A rapid marker ordering approach for high-density genetic linkage maps in experimental autotetraploid populations using multidimensional scaling. Züchter Genet Breed Res 129:2117–2132. https://doi.org/10.1007/s00122-016-2761-8 Sandercock AM, Peel MD, Taniguti CH, et al (2025) BIGapp: A user-friendly genomic tool kit identified quantitative trait loci for creeping rootedness in alfalfa (Medicago sativa L.). Plant Genome 18:e70067. https://doi.org/10.1002/tpg2.70067 Semalaiyappan J, Selvanayagam S, Rathore A, et al (2022) Development of a new AgriSeq 4K mid-density SNP genotyping panel and its utility in pearl millet breeding. Front Plant Sci 13:1068883. https://doi.org/10.3389/fpls.2022.1068883 Taniguti CH, Taniguti LM, Amadeu RR, et al (2022) Developing best practices for genotyping-by-sequencing analysis in the construction of linkage maps. Gigascience 12:. https://doi.org/10.1093/gigascience/giad092 Thérèse Navarro A, Tumino G, Voorrips RE, et al (2022) Multiallelic models for QTL mapping in diverse polyploid populations. BMC Bioinformatics 23:67. https://doi.org/10.1186/s12859-022-04607-z Thompson NF, Sutherland BJG, Green TJ, Delomas TA (2025) A free lunch: microhaplotype discovery in an existing amplicon panel improves parentage assignment for the highly polymorphic Pacific oyster. G3 (Bethesda) 15:1–11. https://doi.org/10.1093/g3journal/jkae280 Tirnaz S, Zandberg J, Thomas WJW, et al (2022) Application of crop wild relatives in modern breeding: An overview of resources, experimental and computational methodologies. Front Plant Sci 13:1008904. https://doi.org/10.3389/fpls.2022.1008904 Turchi C, Melchionda F, Gentile F, et al (2026) Microhaplotypes in forensic genetics: From exploration to application in degraded DNA specimens. Forensic Sci Int Genet 81:103391. https://doi.org/10.1016/j.fsigen.2025.103391 Vos P, Hogers R, Bleeker M, et al (1995) AFLP: a new technique for DNA fingerprinting. Nucleic Acids Res 23:4407–4414. https://doi.org/10.1093/nar/23.21.4407 Wang N, Akey JM, Zhang K, et al (2002) Distribution of recombination crossovers and the origin of haplotype blocks: the interplay of population history, recombination, and mutation. Am J Hum Genet 71:1227–1234. https://doi.org/10.1086/344398 Welsh J, McClelland M (1990) Fingerprinting genomes using PCR with arbitrary primers. Nucleic Acids Res 18:7213–7218. https://doi.org/10.1093/nar/18.24.7213 Williams JG, Kubelik AR, Livak KJ, et al (1990) DNA polymorphisms amplified by arbitrary primers are useful as genetic markers. Nucleic Acids Res 18:6531–6535. https://doi.org/10.1093/nar/18.22.6531 Wu R, Chen H, Li R, et al (2021) Pairwise kinship testing with microhaplotypes: Can advancements be made in kinship inference with these markers? Forensic Sci Int 325:110875. https://doi.org/10.1016/j.forsciint.2021.110875 Wu S, Lau KH, Cao Q, et al (2018) Genome sequences of two diploid wild relatives of cultivated sweetpotato reveal targets for genetic improvement. Nat Commun 9:. https://doi.org/10.1038/s41467-018-06983-8 Yang L, Koo D-H, Li Y, et al (2012) Chromosome rearrangements during domestication of cucumber as revealed by high-density genetic mapping and draft genome assembly. Plant J 71:895–906. https://doi.org/10.1111/j.1365-313X.2012.05017.x Zhang H, Mittal N, Leamy LJ, et al (2017) Back into the wild-Apply untapped genetic diversity of wild relatives for crop improvement. Evol Appl 10:5–24. https://doi.org/10.1111/eva.12434 Zhao D, Chinchilla-Vargas J, Sandercock AM, et al (2025) A public mid-density genotyping platform for North American Atlantic salmon (Salmo salar L.). GenResJ 6:111–119. https://doi.org/10.46265/genresj.skpz5680 Zhao D, Mejia-Guerra KM, Mollinari M, et al (2023) A public mid-density genotyping platform for alfalfa (Medicago sativa L.). Gen Res J 4:55–63. https://doi.org/10.46265/genresj.EMOR6509 Zhao D, Sandercock AM, Mejia-Guerra MK, et al (2024a) A public mid-density genotyping platform for hexaploid Sweetpotato (Ipomoea batatas [L.] Lam). Genes (Basel) 15:1047. https://doi.org/10.3390/genes15081047 Zhao D, Sapkota M, Glaubitz J, et al (2024b) A public mid-density genotyping platform for cultivated blueberry (Vaccinium spp.). GenResJ 5:36–44. https://doi.org/10.46265/genresj.WQZS1824 Zhao D, Sapkota M, Lin M, et al (2024c) Genetic diversity, population structure, and taxonomic confirmation in annual medic (Medicago spp.) collections from Crimea, Ukraine. Front Plant Sci 15:. https://doi.org/10.3389/fpls.2024.1339298 Additional Declarations No competing interests reported. Supplementary Files SupplementaryMaterials.zip Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 18 Apr, 2026 Reviewers agreed at journal 14 Apr, 2026 Reviewers agreed at journal 12 Apr, 2026 Reviewers invited by journal 09 Apr, 2026 Editor assigned by journal 03 Apr, 2026 Submission checks completed at journal 02 Apr, 2026 First submitted to journal 01 Apr, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9292361","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":623209420,"identity":"800a9678-572f-4a6d-ac0a-27b6aa191209","order_by":0,"name":"Dongyan Zhao","email":"","orcid":"","institution":"University of Florida","correspondingAuthor":false,"prefix":"","firstName":"Dongyan","middleName":"","lastName":"Zhao","suffix":""},{"id":623209422,"identity":"419d6dde-00f3-467f-ac56-0289ca949aee","order_by":1,"name":"Meng Lin","email":"","orcid":"","institution":"University of Florida","correspondingAuthor":false,"prefix":"","firstName":"Meng","middleName":"","lastName":"Lin","suffix":""},{"id":623209426,"identity":"3c24dc74-6733-448d-a250-9e508ec74343","order_by":2,"name":"Cristiane H. Taniguti","email":"","orcid":"","institution":"University of Florida","correspondingAuthor":false,"prefix":"","firstName":"Cristiane","middleName":"H.","lastName":"Taniguti","suffix":""},{"id":623209432,"identity":"19f35ccb-3b2d-429d-8b1c-7c4579952a72","order_by":3,"name":"Alexander M. Sandercock","email":"","orcid":"","institution":"University of Florida","correspondingAuthor":false,"prefix":"","firstName":"Alexander","middleName":"M.","lastName":"Sandercock","suffix":""},{"id":623209438,"identity":"6ca6f85e-7520-4271-beef-52b95e54ada2","order_by":4,"name":"Shufen Chen","email":"","orcid":"","institution":"University of Florida","correspondingAuthor":false,"prefix":"","firstName":"Shufen","middleName":"","lastName":"Chen","suffix":""},{"id":623209443,"identity":"e99f9fe9-f9b0-4a8c-af4f-bfe57bfb2e44","order_by":5,"name":"Ebrahiem Babiker","email":"","orcid":"","institution":"USDA ARS Corn Host Plant Resistance Research Unit","correspondingAuthor":false,"prefix":"","firstName":"Ebrahiem","middleName":"","lastName":"Babiker","suffix":""},{"id":623209448,"identity":"7918ead0-e079-4629-a07e-c64d0b690590","order_by":6,"name":"Nahla Bassil","email":"","orcid":"","institution":"USDA ARS National Clonal Germplasm Repository","correspondingAuthor":false,"prefix":"","firstName":"Nahla","middleName":"","lastName":"Bassil","suffix":""},{"id":623209452,"identity":"27e4ee43-cd5c-4a14-b343-27b92ca69a95","order_by":7,"name":"E. Charles Brummer","email":"","orcid":"","institution":"University of California Davis","correspondingAuthor":false,"prefix":"","firstName":"E.","middleName":"Charles","lastName":"Brummer","suffix":""},{"id":623209456,"identity":"6b29fa47-229a-43bb-87d1-b714bd505c58","order_by":8,"name":"Jose R. Camacho","email":"","orcid":"","institution":"The Instituto Nacional de Innovación y Transferencia en Tecnología Agropecuaria (INTA)","correspondingAuthor":false,"prefix":"","firstName":"Jose","middleName":"R.","lastName":"Camacho","suffix":""},{"id":623209460,"identity":"06565f57-36b1-4b54-9deb-485c3ec22dc4","order_by":9,"name":"Warren Chatwin","email":"","orcid":"","institution":"USDA ARS Pecan Breeding and Genetics Program","correspondingAuthor":false,"prefix":"","firstName":"Warren","middleName":"","lastName":"Chatwin","suffix":""},{"id":623209465,"identity":"b52aebff-8da4-4692-a6e0-0359d448599b","order_by":10,"name":"Shu-Yun Chen","email":"","orcid":"","institution":"National Chung Hsing University","correspondingAuthor":false,"prefix":"","firstName":"Shu-Yun","middleName":"","lastName":"Chen","suffix":""},{"id":623209468,"identity":"0f322589-db86-4ec8-b2f4-96c436c5b376","order_by":11,"name":"Shaun J. Clare","email":"","orcid":"","institution":"Carlsberg Research Laboratory","correspondingAuthor":false,"prefix":"","firstName":"Shaun","middleName":"J.","lastName":"Clare","suffix":""},{"id":623209471,"identity":"7f782e63-92b9-4dad-9f96-72112266a773","order_by":12,"name":"Maria David","email":"","orcid":"","institution":"International Potato Center (CIP)","correspondingAuthor":false,"prefix":"","firstName":"Maria","middleName":"","lastName":"David","suffix":""},{"id":623209474,"identity":"1090ff43-4cff-4444-8130-8d08ad3eced1","order_by":13,"name":"Simon P. Fraher","email":"","orcid":"","institution":"NC State University","correspondingAuthor":false,"prefix":"","firstName":"Simon","middleName":"P.","lastName":"Fraher","suffix":""},{"id":623209477,"identity":"f7f135a9-2817-42e6-9b8a-ada127999a4c","order_by":14,"name":"Michael Hardigan","email":"","orcid":"","institution":"USDA ARS Horticultural Crops Production and Genetic Improvement Research Unit","correspondingAuthor":false,"prefix":"","firstName":"Michael","middleName":"","lastName":"Hardigan","suffix":""},{"id":623209482,"identity":"ef57d8d4-ceef-4b48-936a-d6ed9ae4d73b","order_by":15,"name":"Angelyn Hilton","email":"","orcid":"","institution":"USDA ARS Pecan Breeding and Genetics Program","correspondingAuthor":false,"prefix":"","firstName":"Angelyn","middleName":"","lastName":"Hilton","suffix":""},{"id":623209488,"identity":"92ecf811-6988-4cda-8d38-4fc36d783759","order_by":16,"name":"Lilian M. Hislop","email":"","orcid":"","institution":"Savanna Institute","correspondingAuthor":false,"prefix":"","firstName":"Lilian","middleName":"M.","lastName":"Hislop","suffix":""},{"id":623209494,"identity":"e5287b30-3454-4173-a5d1-c3cfeaa8ad33","order_by":17,"name":"Brian M. Irish","email":"","orcid":"","institution":"USDA ARS Plant Germplasm Introduction and Testing Research Unit","correspondingAuthor":false,"prefix":"","firstName":"Brian","middleName":"M.","lastName":"Irish","suffix":""},{"id":623209499,"identity":"852f5698-941f-4202-a575-698ee90cb47d","order_by":18,"name":"Moctar Kante","email":"","orcid":"","institution":"International Potato Center (CIP)","correspondingAuthor":false,"prefix":"","firstName":"Moctar","middleName":"","lastName":"Kante","suffix":""},{"id":623209501,"identity":"6818751b-8348-412d-805f-138d36383b43","order_by":19,"name":"Tae Hwa Kim","email":"","orcid":"","institution":"National Institute of Crop and Food Science, Rural Development Administration","correspondingAuthor":false,"prefix":"","firstName":"Tae","middleName":"Hwa","lastName":"Kim","suffix":""},{"id":623209505,"identity":"6940b913-b39d-4f79-8c69-c96750ece7c3","order_by":20,"name":"Chong-Wei Lee","email":"","orcid":"","institution":"National Chung Hsing University","correspondingAuthor":false,"prefix":"","firstName":"Chong-Wei","middleName":"","lastName":"Lee","suffix":""},{"id":623209507,"identity":"bbe300ae-5718-4e0f-95e8-63036d7ab6df","order_by":21,"name":"Hannele Lindqvist-Kreuze","email":"","orcid":"","institution":"International Potato Center (CIP)","correspondingAuthor":false,"prefix":"","firstName":"Hannele","middleName":"","lastName":"Lindqvist-Kreuze","suffix":""},{"id":623209508,"identity":"22d0d35f-5ec0-48c5-b05d-ccb35ac7f238","order_by":22,"name":"Jenyne Loarca","email":"","orcid":"","institution":"USDA ARS Horticultural Crops Production and Genetic Improvement Research Unit","correspondingAuthor":false,"prefix":"","firstName":"Jenyne","middleName":"","lastName":"Loarca","suffix":""},{"id":623209509,"identity":"4e1912e8-3187-4a3a-94e5-6a6d9725cffa","order_by":23,"name":"Po-Hsien Lu","email":"","orcid":"","institution":"National Chung Hsing University","correspondingAuthor":false,"prefix":"","firstName":"Po-Hsien","middleName":"","lastName":"Lu","suffix":""},{"id":623209510,"identity":"3a318d5d-5278-498e-bae3-12aad8faaaa3","order_by":24,"name":"Cesar A. Medina","email":"","orcid":"","institution":"University of Minnesota","correspondingAuthor":false,"prefix":"","firstName":"Cesar","middleName":"A.","lastName":"Medina","suffix":""},{"id":623209511,"identity":"09e7a8af-5986-4a4e-b777-b479f7d198a3","order_by":25,"name":"Fabián Jiménez Morales","email":"","orcid":"","institution":"The Instituto Nacional de Innovación y Transferencia en Tecnología Agropecuaria (INTA)","correspondingAuthor":false,"prefix":"","firstName":"Fabián","middleName":"Jiménez","lastName":"Morales","suffix":""},{"id":623209514,"identity":"f6d062d7-f24a-4f74-867e-751473bf1e27","order_by":26,"name":"James Polashock","email":"","orcid":"","institution":"USDA ARS Genetic Improvement for Fruits \u0026 Vegetables Laboratory","correspondingAuthor":false,"prefix":"","firstName":"James","middleName":"","lastName":"Polashock","suffix":""},{"id":623209515,"identity":"710b3798-10ae-4d00-b184-f6dc399b420c","order_by":27,"name":"John H. Price","email":"","orcid":"","institution":"USDA ARS Genetic Improvement for Fruits \u0026 Vegetables Laboratory","correspondingAuthor":false,"prefix":"","firstName":"John","middleName":"H.","lastName":"Price","suffix":""},{"id":623209516,"identity":"5928469a-3326-4a6c-a343-6d01b6d857e7","order_by":28,"name":"Heathcliffe Riday","email":"","orcid":"","institution":"USDA ARS Dairy Forage Research Center","correspondingAuthor":false,"prefix":"","firstName":"Heathcliffe","middleName":"","lastName":"Riday","suffix":""},{"id":623209517,"identity":"4a8cd3b9-84e6-4252-8776-6a900f958cdd","order_by":29,"name":"Deborah A. Samac","email":"","orcid":"","institution":"USDA ARS Plant Science Research Unit","correspondingAuthor":false,"prefix":"","firstName":"Deborah","middleName":"A.","lastName":"Samac","suffix":""},{"id":623209518,"identity":"ca87ed2f-d3ed-4247-9f38-19d23bb6af7f","order_by":30,"name":"Devinder Sandhu","email":"","orcid":"","institution":"USDA ARS Agricultural Water Efficiency and Salinity Research Unit","correspondingAuthor":false,"prefix":"","firstName":"Devinder","middleName":"","lastName":"Sandhu","suffix":""},{"id":623209519,"identity":"1045685e-67ac-41dd-9a46-1b62df2580c6","order_by":31,"name":"Reuben Ssali","email":"","orcid":"","institution":"International Potato Center (CIP)","correspondingAuthor":false,"prefix":"","firstName":"Reuben","middleName":"","lastName":"Ssali","suffix":""},{"id":623209520,"identity":"4a7aa4fb-f1e6-45ad-a4f9-d7f48aeb4bff","order_by":32,"name":"Ruth Castro Vásquez","email":"","orcid":"","institution":"The Instituto Nacional de Innovación y Transferencia en Tecnología Agropecuaria (INTA)","correspondingAuthor":false,"prefix":"","firstName":"Ruth","middleName":"Castro","lastName":"Vásquez","suffix":""},{"id":623209521,"identity":"b5f9b9de-d408-4b78-bb63-fd1477b60433","order_by":33,"name":"Phillip A. Wadl","email":"","orcid":"","institution":"USDA ARS US Vegetable Laboratory","correspondingAuthor":false,"prefix":"","firstName":"Phillip","middleName":"A.","lastName":"Wadl","suffix":""},{"id":623209522,"identity":"9e55d623-31bf-47fe-8d17-5cb67e16bd13","order_by":34,"name":"Xinwang Wang","email":"","orcid":"","institution":"USDA ARS Pecan Breeding and Genetics Program","correspondingAuthor":false,"prefix":"","firstName":"Xinwang","middleName":"","lastName":"Wang","suffix":""},{"id":623209523,"identity":"7b030371-26c9-42c3-a48d-8d860778f53d","order_by":35,"name":"Seymour A. Webster","email":"","orcid":"","institution":"College of Agriculture, Science and Education","correspondingAuthor":false,"prefix":"","firstName":"Seymour","middleName":"A.","lastName":"Webster","suffix":""},{"id":623209524,"identity":"20d37d16-aee1-4fba-b36f-b380309a798f","order_by":36,"name":"Zhanyou Xu","email":"","orcid":"","institution":"SDA ARS Soybean/Maize Germplasm, Pathology, and Genetics Research Unit","correspondingAuthor":false,"prefix":"","firstName":"Zhanyou","middleName":"","lastName":"Xu","suffix":""},{"id":623209525,"identity":"5261414b-0965-44e2-a03d-6098271b8bcc","order_by":37,"name":"G. Craig Yencho","email":"","orcid":"","institution":"NC State University","correspondingAuthor":false,"prefix":"","firstName":"G.","middleName":"Craig","lastName":"Yencho","suffix":""},{"id":623209526,"identity":"0ab058cd-8a07-4686-a1e6-8434795cb784","order_by":38,"name":"Craig. T. Beil","email":"","orcid":"","institution":"University of Florida","correspondingAuthor":false,"prefix":"","firstName":"Craig.","middleName":"T.","lastName":"Beil","suffix":""},{"id":623209527,"identity":"db876ddd-ca22-4dfe-a84a-0e89605c53b0","order_by":39,"name":"Moira J. Sheehan","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA3UlEQVRIiWNgGAWjYFACxgYgcYCBQQJIfWCQkwFSBsRrYZzBYMxDhBYwgGhh5iFGi257c9uHH3/uyPFLNz97bNtmwMPA3rxNAp8WszMHm2f2tj0zlpxzzNw4F6SF51gZfi03EpsZeBsOJ264kWAmndv2h4dBIseMoBbGP38O1++/kf5N2hJki/wbwlqYedgOJxgADZdmBGmR4CGgBegXZtm2w4YzbuSUSfacM+Bh40krtsCr5Xj7Y8Y3fw7L889I3ybxo8xAjp/98MYb+LRgAjbSlI+CUTAKRsEowAYATKVHJZugVXUAAAAASUVORK5CYII=","orcid":"","institution":"University of Florida","correspondingAuthor":true,"prefix":"","firstName":"Moira","middleName":"J.","lastName":"Sheehan","suffix":""}],"badges":[],"createdAt":"2026-04-01 13:10:25","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9292361/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9292361/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":107482914,"identity":"8395e657-ef99-42a4-9110-03c1621dc9cf","added_by":"auto","created_at":"2026-04-22 02:25:33","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":597693,"visible":true,"origin":"","legend":"\u003cp\u003eHapApp workflow and interface for fixed microhaplotype IDs. (a) Schematic of the workflow to filter microhaplotypes, assign unique fixed microhaplotype IDs, and update the microhaplotype database using MADC genotyping reports. (b) Screenshot of the HapApp R Shiny interface for processing and ID assignment of microhaplotypes.\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-9292361/v1/b159d4ddb48708be83c00f1b.png"},{"id":107166531,"identity":"58a76576-d62b-4a68-9368-a8f130f0d9e9","added_by":"auto","created_at":"2026-04-17 14:05:55","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":159149,"visible":true,"origin":"","legend":"\u003cp\u003eCumulative microhaplotype counts per database version across eight crop species.\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-9292361/v1/c9ee7377c14822e5a62b452b.png"},{"id":107166534,"identity":"17bb8bf3-494c-40ad-a487-2c7ff79eaa4e","added_by":"auto","created_at":"2026-04-17 14:05:55","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":444122,"visible":true,"origin":"","legend":"\u003cp\u003eDistribution and genomic position of microhaplotype counts per DArTag locus in alfalfa, grouped by amplification category. (a) Boxplots showing the number of microhaplotypes for DArTag loci containing potential paralogous amplification and those likely containing only allelic amplification. (b) Scatter plots or microhaplotype counts for individual DArTag loci against their physical positions on target chromosomes. The red dashed line is an indicator of 30 haplotypes per locus.\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-9292361/v1/b0de36e64e03bb6024aa7e40.png"},{"id":107166532,"identity":"ff35632c-6b1c-4e0f-ac9b-aaab3b40aff1","added_by":"auto","created_at":"2026-04-17 14:05:55","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":367147,"visible":true,"origin":"","legend":"\u003cp\u003eComparison of marker ordering and filtering outcomes for pecan across three marker datasets (microhaplotypes, target SNPs, target plus off-target SNPs). (a) Summary of marker filtering by genotype class across the three datasets. Bars represent the number of markers per genotype class removed by specific filters applied in the presented order (starting with depth to rf_filter). (b) Linkage marker ordering for chromosome 8 using genome position (left panels) and multidimensional scaling (MDS) genetic distance (right panels).\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-9292361/v1/a7c3b5aebe797c9c3867947c.png"},{"id":107166533,"identity":"98d68624-2b62-455f-aeee-ac3c9b2aa41b","added_by":"auto","created_at":"2026-04-17 14:05:55","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":878371,"visible":true,"origin":"","legend":"\u003cp\u003eGeographic context, cross validation accuracy and principal component analysis (PCA) of microhaplotypes and SNP markers in sweetpotato. (a) Geographic distribution of genotyping projects using the sweetpotato 3K DArTag panel. (b) Discriminant analysis of principal components (DAPC) cross-validation accuracies using 5-50 PCs for microhaplotypes, target-SNPs, and all-SNPs marker sets. (c and d) PCA plots for PC1 vs. PC2 (c) and PC2 vs. PC3 (d) of the microhaplotype dataset. (e and f) PCA plots for PC1 vs. PC2 (e) and PC2 vs. PC3 (f) of the target-SNPs dataset. (g and h) PCA plots for PC1 vs. PC2 (g) and PC2 vs. PC3 (h) of the all-SNPs dataset.\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-9292361/v1/811f41f880499bc11bdaa11b.png"},{"id":107704888,"identity":"050a96cb-4d2f-4ba6-bc05-38a49bf3ca94","added_by":"auto","created_at":"2026-04-24 09:02:44","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2346801,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9292361/v1/1531479a-c847-4414-9db8-a7bea6a5140c.pdf"},{"id":107166529,"identity":"1c30445c-456c-4814-bd43-29e9b57a6663","added_by":"auto","created_at":"2026-04-17 14:05:55","extension":"zip","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":8658257,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMaterials.zip","url":"https://assets-eu.researchsquare.com/files/rs-9292361/v1/82171745db5b71a9147ba9d8.zip"}],"financialInterests":"No competing interests reported.","formattedTitle":"Development and Characterization of Microhaplotype Databases for Diverse Crop Species","fulltext":[{"header":"Key message","content":"\u003cp\u003eStandardized microhaplotype databases for eight diverse crops enable multiallelic analyses, comparative genetics, and breeding decisions.\u003c/p\u003e"},{"header":"Introduction","content":"\u003cp\u003eThe integration of molecular markers into plant breeding programs has dramatically accelerated genetic improvement by enabling precise germplasm characterization and selection. DNA-based markers in plants have taken many forms over the past several decades, starting with restriction fragment length polymorphisms (RFLPs) in 1980 (Botstein et al. \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e1980\u003c/span\u003e), random amplified polymorphic DNA (RAPD) (Williams et al. \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e1990\u003c/span\u003e; Welsh and McClelland \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e1990\u003c/span\u003e), amplified fragment length polymorphisms (AFLPs) (Vos et al. \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e1995\u003c/span\u003e), simple-sequence repeats (SSRs) (Mr\u0026aacute;zek et al. \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2007\u003c/span\u003e), and single-nucleotide polymorphisms (SNPs) (Mammadov et al. \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). Today, the predominant DNA marker type is the bi-allelic SNP marker due to its abundance and wide distribution across genomes and its suitability for high-throughput genotyping. SNPs can be identified cheaply and quickly using a wide range of detection technologies. For researchers, SNP datasets require minimal storage space and are easier to interpret than many other marker systems. There are several software packages available for researchers to use that aid allele calling from raw data, including proprietary software such as KlusterCaller and SNPviewer (LGC Ltd, Huntingdon, UK) and GenomeStudio (Illumina, San Diego, CA, USA), and free R packages such as polyRAD and updog (Clark et al. \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Gerard and Ferr\u0026atilde;o \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). SNPs are often preferred for time-sensitive, data-driven decisions on germplasm in breeding programs. As such, SNPs have become indispensable for assessing genetic diversity and population structure, mapping traits of interest, and implementing marker-assisted selection/genomic selection in plant and animal breeding programs. While bi-allelic SNP calling is often sufficient and accurate for diploids or species with low levels of heterozygosity, it lacks sufficient information to accurately resolve genotypes in highly heterozygous species and species with complex genomes, especially polyploids. Recent advances in DNA sequencing technology have led to the development of various approaches for genetic architecture profiling. Whole-genome sequencing (WGS) provides the most comprehensive view of genetic variation but remains cost-prohibitive to be applied yearly for routine breeding applications. To address this limitation, reduced representation approaches have emerged as cost-effective alternatives. These methods can be broadly categorized into two strategies: random sampling and targeted amplification of genome sequences.\u003c/p\u003e \u003cp\u003eRandom reduced-representation methods, such as genotyping-by-sequencing (GBS) and Diversity Arrays Technology sequencing (DArTseq), employ restriction enzymes to sample a subset of the genome (Elshire et al. \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Cruz et al. \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). GBS typically utilizes methylation-sensitive restriction enzymes to reduce genome complexity, thereby enriching for low-copy and genic regions and increasing the likelihood of detecting functional variants. In DArTseq, a species-specific combination of restriction enzymes is used to achieve genome complexity reduction and preferentially amplify low-copy fragments. Both approaches generate reduced-representation libraries and have proven particularly valuable for species with limited or no reference genome resources, enabling rapid and cost-effective genome-wide variant discovery. The downside of both methods is the low reproducibility from one project to another, since a random subset of a reduced representation library is sequenced each time, leading to high missing data rates between project runs. Additionally, the data analysis is computationally intensive, requiring skilled bioinformatic support and ample time to complete.\u003c/p\u003e \u003cp\u003eIn contrast, targeted genotyping approaches such as DArTag, AgriSeq, FlexSeq, GT-seq, and related platforms are typically designed with primers flanking an anchor SNP at each locus (Campbell et al. \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Semalaiyappan et al. \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Blyton et al. \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Clare et al. \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). These methods offer several advantages over GBS and DArTseq, including higher coverage of target loci, low missing data rates, read depths that are consistent across samples and projects ensures data are comparable across projects, and reduced computational requirements for data analysis (Zhao et al. \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2023\u003c/span\u003e, \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2024b\u003c/span\u003e, \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003ea\u003c/span\u003e; Hardigan et al. \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Sandercock et al. \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). A further key advantage is the ability to generate detailed sequence information surrounding variant sites, as these approaches employ next-generation sequencing of short reads (typically 50\u0026ndash;250 bp). Rather than providing only single nucleotide calls at target variant sites, these platforms return data in two formats: raw FASTQ files containing individual sequencing reads, and semi-processed marker reports presenting multi-allelic sequences for each locus. These sequences contain not only the target variant site but also capture adjacent polymorphisms within the amplified regions. The co-occurrence of closely linked variants within these short haplotype blocks can yield three or more allelic combinations, characterizing a multi-allelic microhaplotype (Kidd et al. \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). Due to their short length, variants within a microhaplotype are assumed to remain in strong linkage and be inherited as a unit, and the combination of variants within have known phasing. Multiallelic genetic data can provide more detailed and nuanced insights into genetic diversity, allele dosage, trait expression, and population structure. The comprehensive sequence information also enables more accurate genotype calling through sequence-context evaluation, enhanced discrimination power through microhaplotype-based analysis, and the identification of novel alleles within the amplified regions.\u003c/p\u003e \u003cp\u003eThe concept of microhaplotypes was first introduced in the forensics field by Kidd and colleagues (Kidd et al. \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2013\u003c/span\u003e) and rapidly gained popularity due to their high information content and analytical efficiency. Subsequent studies in forensic genetics have demonstrated the advantages of using multi-allelic microhaplotypes over individual SNPs in diverse applications, including kinship investigations, ancestry inference, resolution of complex DNA mixture, and recovery of allelic information from degraded DNA specimens (Kidd et al. \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Oldoni and Podini \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; de Barros Rodrigues et al. \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Turchi et al. \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2026\u003c/span\u003e). Beyond forensics, the use of microhaplotypes derived from targeted genotyping panels has demonstrated notable advantages in aquaculture. GT-seq microhaplotypes have been successfully applied to improve parentage assignment in oysters, salmon, and rockfishes (Baetscher et al. \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Thompson et al. \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Anderson et al. \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). For example, simulation studies in oyster and salmon breeding programs have indicated that incorporating microhaplotypes can increase the accuracy of estimated breeding values (Delomas et al. \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn plant breeding, the advantages of using multi-allelic markers have been highlighted across diverse applications, including genomic prediction (GP) and genomic selection (GS) as well as genome-wide association studies (GWAS) (Matias et al. \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Abed and Belzile \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Traditionally, most multi-allelic markers in crop studies have been derived from GBS or WGS datasets by estimating combinations of SNPs in complete linkage disequilibrium, referred to as haplotype blocks. Such blocks are typically inferred through methods such as confidence interval-based approaches using normalized measures of allelic association, identification of genomic regions with limited recombination events, or read-based phasing approaches (Gabriel et al. \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2002\u003c/span\u003e; Wang et al. \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2002\u003c/span\u003e; Garrison and Marth \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Martin et al. \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). Simulation studies have shown that GWAS models accommodating multiallelic markers and haplotype blocks in polyploid species achieve higher accuracy than models based on bi-allelic SNPs (Th\u0026eacute;r\u0026egrave;se Navarro et al. \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). More recently, microhaplotypes derived from DArTag panels have been applied to genomic studies in alfalfa, demonstrating higher values of both intra- and inter-population diversity compared to SNP markers (Medina et al. \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eOver the past five years, Breeding Insight (BI; RRID: SCR_026645) and collaborators have developed a series of medium-density DArTag genotyping panels (Zhao et al. \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2023\u003c/span\u003e, \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2024b\u003c/span\u003e, \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003ea\u003c/span\u003e, \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Hardigan et al. \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Endelman et al. \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Chen et al. \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). These panels were developed using a diverse breeding germplasm and validated across F\u003csub\u003e1\u003c/sub\u003e and/or backcross mapping populations as well as core diversity panels, ensuring robustness across breeding materials. The DArTag genotyping platform produces several data outputs including traditional FASTQ files and a proprietary format called Missing Allele Discovery Counts (MADC) file. The MADC file contains read counts for all microhaplotypes at targeted loci and captures four distinct classes of microhaplotypes. Reference (Ref) and Alternative (Alt) alleles represent the target variants known \u003cem\u003ea priori\u003c/em\u003e from panel design and are analogous to SNP marker definitions of the same designation. In contrast to Ref and Alt, RefMatch and AltMatch sequences match either the reference or alternative state at the target site while containing additional polymorphisms in the flanking sequences that were not known at the time of design, thus newly \u0026ldquo;discovered\u0026rdquo;. One challenge in utilizing MADC data across projects stems from the lack of unique identifiers for RefMatch and AltMatch alleles from all genotyping projects. To systematically organize the RefMatch and AltMatch information, we have established a species-agnostic pipeline to assign standardized and unique microhaplotype identities (IDs) to serve three essential purposes: 1) enable efficient tracking of marker genetic information, 2) facilitate cross-project and cross-institution comparison of genotyping results, and 3) abide by Findable, Accessible, Interoperable, and Reusable (FAIR) data principles (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.go-fair.org/fair-principles/\u003c/span\u003e\u003cspan address=\"https://www.go-fair.org/fair-principles/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Here, we present the individual microhaplotype databases for eight plant species (alfalfa, blueberry, cranberry, cucumber, pecan, potato, strawberry, and sweetpotato) and open them to the public such that breeders and researchers can fully leverage the rich information generated via DArTag genotyping through community efforts.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\n\u003ch2\u003eDatasets used to generate the microhaplotype databases\u003c/h2\u003e\n\u003cp\u003eDiverse sample sets, including F\u003csub\u003e1\u003c/sub\u003e and backcross populations for each species, were collected by breeders and researchers. These samples were genotyped at DArT (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ewww.diversityarrays.com\u003c/span\u003e\u003c/span\u003e), and the resulting MADC reports were shared with BI to facilitate the generation of microhaplotype databases.\u003c/p\u003e\n\u003c/div\u003e\n\u003ch3\u003eNaming convention\u003c/h3\u003e\n\u003cp\u003eTo ensure naming consistency across DArTag panels developed at different times and by different research groups, we implemented a systematic marker ID standardization process. The original marker IDs, which varied in format and naming conventions, were converted to a uniform structure based on chromosome number and genomic position from the reference genome used for panel design. Genomic positions were padded to nine digits to maintain consistent formatting (e.g., chr1_000589632). This standardization step was performed prior to microhaplotype ID assignment to ensure marker tracking across all panels and downstream analyses.\u003c/p\u003e\n\u003ch3\u003eCore database structure and panel-specific considerations\u003c/h3\u003e\n\u003cp\u003eThe core database forms the foundation of the microhaplotype tracking system, offering consistent reference points for allele comparison and standard templates for sequence alignment. This core database contains both Ref and Alt alleles, with sequence lengths varying according to panel design specifications. The genomic insert size in DArTag panels has lengthened over time as sequencing costs have decreased, resulting in two main design categories. The 54-bp design (~\u0026thinsp;75 bp with primer sequences before trimming) was implemented in alfalfa, blueberry, and Phase I markers in potato and strawberry panels. The 81-bp design (~\u0026thinsp;100 bp with primer sequences before trimming) was implemented in cranberry, cucumber, pecan, sweetpotato panels, and Phase II markers in potato and strawberry panels.\u003c/p\u003e\n\u003ch3\u003eConstruction of reference-based core database\u003c/h3\u003e\n\u003cp\u003eThe construction of the core microhaplotype database for reference-based markers involves an iterative process using probe design files, reference genomes, and analytical scripts. This workflow addresses two key challenges: 1) in the genotyping reports, Ref and Alt amplicons often contain International Union of Pure and Applied Chemistry (IUPAC) codes representing ambiguous base calls, and 2) the 3\u0026rsquo; ends of some amplicons may contain sequencing errors. The primary goal is to create a core database of Ref and Alt microhaplotypes that are free of IUPAC codes and mismatches (except for the target variant in Alt) with the reference genome. Another critical consideration is that a DArTag panel, like other amplicon-sequencing technologies, is composed of amplicons derived from both the top and bottom strands of the reference genome. Accurately establishing the strand orientation is crucial for determining the precise positions and nucleotides of the variant sites. To maintain consistency, all microhaplotype databases preserve strand orientation as implemented during panel creation by DArT.\u003c/p\u003e\n\u003cp\u003eThe workflow begins by extracting 180\u0026ndash;300 bp flanking sequences encompassing the target variants from the corresponding reference genome (Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e). Ref and Alt amplicon sequences are then retrieved from the DArTag MADC report and aligned to the flanking sequences via BLAST (Camacho et al. \u003cspan class=\"CitationRef\"\u003e2009\u003c/span\u003e). This alignment step helps determine the correct strand orientation of the amplicons, and those amplicons originating from the bottom strand are exported as a key file for downstream analysis (\u003cstrong\u003eTable \u003cspan class=\"InternalRef\"\u003eS1\u003c/span\u003e\u003c/strong\u003e). Meanwhile, the coordinates of the Ref and Alt microhaplotypes are determined based on the reference genome, and sequences are fetched to populate the core database. By design, the core database contains twice the number of marker loci in a DArTag panel, as it incorporates both Ref and Alt microhaplotype sequences for each locus.\u003c/p\u003e\n\u003cdiv class=\"gridtable\"\u003e\n\u003cdiv class=\"colspec\" align=\"left\"\u003e\u0026nbsp;\u003c/div\u003e\n\u003cdiv class=\"colspec\" align=\"char\"\u003e\u0026nbsp;\u003c/div\u003e\n\u003ctable id=\"Tab1\" border=\"1\"\u003e\u003ccaption\u003e\n\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\n\u003cdiv class=\"CaptionContent\"\u003e\n\u003cp\u003eSummary of DArTag panels and microhaplotype databases across eight species.\u003c/p\u003e\n\u003c/div\u003e\n\u003c/caption\u003e\n\u003cthead\u003e\n\u003ctr\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eSpecies\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003ePloidy\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eReference genome\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eDeveloped by\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eDesign length\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003e#Markers\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003eDatabase version\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003e#Microhaplotypes\u003c/p\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cp\u003e#Samples\u003c/p\u003e\n\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eAlfalfa\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e4x\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(Chen et al. \u003cspan class=\"CitationRef\"\u003e2020\u003c/span\u003e)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eBI\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e54 bp\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e3K\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e50\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e35,259\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e15,600\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eBlueberry\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e4x\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(Colle et al. \u003cspan class=\"CitationRef\"\u003e2019\u003c/span\u003e)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eBI\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e54 bp\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e3K\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e20\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e28,653\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e8,930\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eCranberry\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e2x\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(Diaz-Garcia et al. \u003cspan class=\"CitationRef\"\u003e2021\u003c/span\u003e)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eBI\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e81 bp\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e3K\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e5\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e14,380\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e4,146\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eCucumber\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e2x\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(Yang et al. \u003cspan class=\"CitationRef\"\u003e2012\u003c/span\u003e)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eBI\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e81 bp\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e3K\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e10\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e9,823\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e8,272\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003ePecan\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e2x\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(Lovell et al. \u003cspan class=\"CitationRef\"\u003e2021\u003c/span\u003e)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eBI\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e81 bp\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e3K\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e9\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e26,073\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e6,768\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eSweetpotato\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e6x\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(Wu et al. \u003cspan class=\"CitationRef\"\u003e2018\u003c/span\u003e)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eBI\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e81 bp\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e3K\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e19\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e37,593\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e9,212\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003ePotato\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e4x\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(Pham et al. \u003cspan class=\"CitationRef\"\u003e2020\u003c/span\u003e)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eCIP-WU\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e54\u0026ndash;81 bp\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e4K\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e15\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e35,741\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e3,102\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eStrawberry\u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e8x\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e(Hardigan et al. \u003cspan class=\"CitationRef\"\u003e2021\u003c/span\u003e)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003eCollaborator\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e54\u0026ndash;81 bp\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cp\u003e5K\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e3\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e38,227\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cp\u003e1,880\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003e\u003csup\u003ea\u003c/sup\u003e Panel designed for cultivated strawberry (octoploid, 8x) where markers were sub-genome specific, enabling functional diploid analysis.\u003c/p\u003e\n\u003ch3\u003eSystematic processing and database integration of novel microhaplotypes\u003c/h3\u003e\n\u003cp\u003eWe developed a systematic workflow to process MADC reports and assign standardized microhaplotype IDs. The process begins with updating marker IDs to the standard \u0026ldquo;chromosomeN_000000000\u0026rdquo; format from the initial MADC report. As an initial quality control, we filter RefMatch and AltMatch sequences to require a minimum presence of either 10 samples or 5% of the total samples in a project (whichever is smaller), with each sample having at least two reads (Fig.\u0026nbsp;1a).\u003c/p\u003e\n\u003cp\u003eThe workflow diverges from here, based on amplicon size. For reports containing sequencing reads longer than genomic insert size by panel design, the reads often contain adapter sequences. In this case, microhaplotypes were first cleaned by removing potential adapter sequences with Cutadapt (Martin \u003cspan class=\"CitationRef\"\u003e2011\u003c/span\u003e), and any duplicates identified among RefMatch and AltMatch sequences are marked for exclusion. For reports with reads matching the panel design length, no adapter removal is needed. All sequences underwent BLAST analysis against the existing microhaplotype database. Any previously cataloged sequences in the database were renamed with their previously assigned unique database IDs to produce an updated MADC report with fixed microhaplotype IDs. Any novel sequences that met our inclusion criteria (\u0026ge;\u0026thinsp;90% identity over \u0026ge;\u0026thinsp;90% coverage of subject microhaplotypes) were incorporated into the database and then assigned new unique identifiers, following the RefMatch and AltMatch classification scheme. These new unique IDs were then written back to the updated MADC report, which then had unique standardized IDs for all observed microhaplotypes. The process yielded two key outputs: (1) MADC report with standardized universal microhaplotypes IDs (MADC_fixedID) and filtered for missing data, and (2) a new version of the microhaplotype database incorporating newly identified alleles with unique IDs. If no new alleles were found, the database remained unchanged, but the MADC report is still generated with universal standardized microhaplotype IDs.\u003c/p\u003e\n\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\n\u003ch2\u003eCreation of an R Shiny interface for processing standardized microhaplotype ID assignment\u003c/h2\u003e\n\u003cp\u003eThe workflow described above provides a flexible pipeline for processing MADC files, identifying new unique microhaplotypes, and producing an updated MADC with fixed microhaplotype IDs based on the microhaplotype database. However, its Bash and Python scripting format makes it inaccessible to users without coding experience or who are uncomfortable with their skill level. To improve user accessibility, we created a no-code R Shiny interface that provides a point-and-click method for using the workflow (Chang et al. \u003cspan class=\"CitationRef\"\u003e2024\u003c/span\u003e), which has been shown to provide complex plant and animal breeding workflows with a more approachable framework (Sandercock et al. \u003cspan class=\"CitationRef\"\u003e2025\u003c/span\u003e). When the original MADC file is uploaded, it is first checked to confirm it is in the correct format. Then the application submits the MADC to the pipeline described above, which performs the computation steps. The updated information is captured and displayed to the user along with any warnings or notifications. The output files include 1) a new version of the microhaplotype FASTA file, and 2) a MADC file with microhaplotypes that passed initial quality control and were assigned a unique standardized ID.\u003c/p\u003e\n\u003c/div\u003e\n\u003ch3\u003eSequence alignment for the alfalfa and blueberry microhaplotype databases\u003c/h3\u003e\n\u003cp\u003eThe alfalfa microhaplotype database, as of November 2025, is at v50. It was aligned to the haplotype-based \u0026lsquo;XinJiangDaYe\u0026rsquo; reference genome (Chen et al. \u003cspan class=\"CitationRef\"\u003e2020\u003c/span\u003e) to understand the distribution of potential paralogous amplifications. The alignment was performed using Bowtie2 v2.5.1 (Langmead and Salzberg \u003cspan class=\"CitationRef\"\u003e2012\u003c/span\u003e) with default parameters. All possible alignments were returned followed by a preliminary filter to remove any alignment with a total number of insertions/deletions \u0026gt;5bp. A total of 34,891 microhaplotypes were successfully aligned to at least one locus in the reference genome. The syntenic genomic regions (Chen et al. \u003cspan class=\"CitationRef\"\u003e2020\u003c/span\u003e) across four haplotypes in each of the eight homologous groups were used to determine the most likely syntenic alignment when a microhaplotype aligned to multiple loci on a single haplotype. A microhaplotype was considered a potential paralogous amplification if it had 1) one or more alignments outside its target homologous group or 2) two or more copies on any of the four haplotypes within the homologous group but outside the target genomic region (Figure \u003cspan class=\"InternalRef\"\u003eS1\u003c/span\u003e C \u0026amp; D). For DArTag loci containing only non-paralogous microhaplotypes, the copy number (ranging from one to four) of each target locus was determined based on the number of aligned positions within the corresponding syntenic regions (Figure \u003cspan class=\"InternalRef\"\u003eS1\u003c/span\u003e a \u0026amp; b).\u003c/p\u003e\n\u003cp\u003eThe most up-to-date blueberry microhaplotype database, v20, was characterized for potential paralogous amplification using the same strategy as in alfalfa. The blueberry database was aligned to \u0026lsquo;Draper\u0026rsquo; reference genome (Colle et al. \u003cspan class=\"CitationRef\"\u003e2019\u003c/span\u003e) and 28,448 microhaplotypes were successfully aligned to at least one locus in the reference genome. Potential paralogous amplifications and copy number of non-paralogous loci in the blueberry microhaplotype database were determined as described above.\u003c/p\u003e\n\u003ch3\u003eComparison of microhaplotypes, target, and off-target markers through linkage map construction in pecan\u003c/h3\u003e\n\u003cp\u003eLinkage maps were built using OneMap v3.2.2 (Taniguti et al. \u003cspan class=\"CitationRef\"\u003e2022\u003c/span\u003e) based on a previously published pecan (\u003cem\u003eCarya illinoinensis\u003c/em\u003e) F\u003csub\u003e1\u003c/sub\u003e dataset comprising 188 progeny (Chen et al.). The mapping procedure was applied using three different data set marker types: (i) microhaplotypes, (ii) target SNPs only, and (iii) combined target and off-target SNPs.\u003c/p\u003e\n\u003cp\u003eFor the microhaplotype dataset (i), the MADC file was processed using our standard workflow, which assigns fixed microhaplotype IDs based on the pecan microhaplotype database v9. The processed file was converted to a polyRAD (v2.0.0) input object using the \u003cem\u003ereadDArTag\u003c/em\u003e function, and genotypes were called using the \u003cem\u003eIterateHWE\u003c/em\u003e function, which assumes Hardy-Weinberg equilibrium without population structure (Clark et al. \u003cspan class=\"CitationRef\"\u003e2019\u003c/span\u003e). We used this approach instead of \u003cem\u003ePipelineMapping2Parents\u003c/em\u003e to avoid errors that can arise when parental genotypes are altered or imputed from progeny segregation patterns (Taniguti et al. \u003cspan class=\"CitationRef\"\u003e2022\u003c/span\u003e). The resulting genotypes were exported as a Variant Call Format (VCF) file using the \u003cem\u003eRADdata2VCF\u003c/em\u003e function with asSNP parameter set to FALSE. For datasets (ii) and (iii), VCF files were generated from the MADC file with the \u003cem\u003emadc2vcf_targets\u003c/em\u003e and \u003cem\u003emadc2vcf_all\u003c/em\u003e functions from the BIGr (v0.6.2) package (Sandercock et al. \u003cspan class=\"CitationRef\"\u003e2025\u003c/span\u003e), respectively. These VCFs contained allele count data and were processed in polyRAD under the same Hardy-Weinberg equilibrium (HWE) genotype calling model used for dataset (i). All datasets were subjected to the same filtering criteria: a minimum genotype depth of six reads, a mean marker depth greater than 30, a maximum of 25% missing genotypes per marker and per individual, removal of redundant markers, and exclusion of markers deviating from expected Mendelian segregation (\u0026alpha;\u0026thinsp;=\u0026thinsp;0.05, Bonferroni-corrected for multiple tests).\u003c/p\u003e\n\u003cp\u003eAdditional filtering was performed using the \u003cem\u003erf_snp_filter_onemap\u003c/em\u003e function, which assesses pairwise recombination fractions (rf) and LOD scores to identify informative markers. For each pair of markers, the function retained only comparisons with strong linkage support (LOD\u0026thinsp;\u0026gt;\u0026thinsp;5) and low recombination (rf\u0026thinsp;\u0026lt;\u0026thinsp;0.15). Pairs that did not meet these criteria were temporarily masked (set to missing), ensuring that only informative pairwise combinations contribute to the count. The number of remaining informative (non-missing) pairwise comparisons was then counted for each marker, and markers falling below the 5th percentile or above the 95th percentile of the resulted counts distribution were removed.\u003c/p\u003e\n\u003cp\u003eMarkers were assigned to linkage groups according to their chromosome in the reference genome. Linkage distances were estimated considering the order of markers using either their physical positions or the multidimensional scaling (MDS) ordering algorithm (Preedy and Hackett \u003cspan class=\"CitationRef\"\u003e2016\u003c/span\u003e). When MDS ordering failed due to collinearity, the lower quantile threshold in \u003cem\u003erf_snp_filter_onemap\u003c/em\u003e was relaxed. In dataset (ii), the threshold was raised to 0.40 for linkage group (LG) 2 and to 0.35 for LG 10 and LG 11. In dataset (iii), the threshold was raised to 0.35 for LG 2 and LG 11. The performance of MDS ordering algorithm in each dataset was evaluated by computing the absolute value of Spearman's rank correlation coefficient \u0026rho; between the algorithm-derived marker order and the reference genome order.\u003c/p\u003e\n\u003cp\u003eThe correlation coefficient was calculated according to Spearman (1904) using the formula\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\rho\\:=\\:1\\:-\\:\\backslash\\:frac\\left\\{6{\\sum\\:}_{\\left\\{i=1\\right\\}}^{\\left\\{m\\right\\}{d}_{i}^{2}}\\right\\}\\left\\{m\\right(m^2-1\\left)\\right\\}\\)\u003c/span\u003e\u003c/span\u003e,\u003c/p\u003e\n\u003cp\u003ewhere \u003cem\u003ed\u003c/em\u003e\u003csub\u003e\u003cem\u003ei\u003c/em\u003e\u003c/sub\u003e denotes the rank difference for marker \u003cem\u003emi\u003c/em\u003e between the estimated and true orders.\u003c/p\u003e\n\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\n\u003ch2\u003eComparative analysis of microhaplotype and SNP genotyping data in sweetpotato breeding germplasm\u003c/h2\u003e\n\u003cp\u003eMicrohaplotype variant data were generated from MADC reports in which alleles had been assigned unique standardized IDs across seven independent sweetpotato DArTag genotyping projects. For each MADC dataset, only Ref and Alt microhaplotypes were retained for marker loci with greater than 15 microhaplotypes since they likely contained some paralogous amplification. Microhaplotype genotype dosage calls were produced as VCF using the \u003cem\u003ereadDArTag\u003c/em\u003e function in the polyRAD (v2.0.0) package (Clark et al. \u003cspan class=\"CitationRef\"\u003e2019\u003c/span\u003e). Individual project VCF files were merged to form a single consolidated VCF and subjected to sequential quality filtering. Variants with low coverage (mean read depth\u0026thinsp;\u0026lt;\u0026thinsp;10), excessively high coverage (mean read depth\u0026thinsp;\u0026gt;\u0026thinsp;1000), or failing missingness criteria were filtered using a population-aware strategy. A microhaplotype was retained if it met either the global threshold of \u0026le;\u0026thinsp;20% missing data across all samples or the per-population threshold of \u0026le;\u0026thinsp;20% missing data in at least one population. This approach ensures that markers informative for specific populations, including low-frequency or private alleles, are not discarded due to varying population size or poor performance in other populations. After dosage calling and marker filtering, samples with \u0026le;\u0026thinsp;20% missing data were removed. Given the multi-allelic nature of microhaplotype genotypes, conventional minor allele frequency (MAF) was not an appropriate measure. Instead, a combined alternative allele frequency (AAF) was calculated and used to filter loci, with those having AAF below 0.05 excluded from further analysis. All filtering steps were performed under a hexaploid ploidy assumption, consistent with the known sweetpotato genome structure. The resulting microhaplotype dataset represented the union of high-quality loci across all projects and was used for principal component analysis (PCA) and discriminant analysis of principal components (DAPC) to evaluate population structure and assignment accuracy using the R packages vcfR (v1.15.0) and adegenet (v2.1.11) (Jombart \u003cspan class=\"CitationRef\"\u003e2008\u003c/span\u003e; Knaus and Gr\u0026uuml;nwald \u003cspan class=\"CitationRef\"\u003e2017\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003eTwo SNP datasets were generated, target-SNPs and all-SNPs (target\u0026thinsp;+\u0026thinsp;off-target SNPs). For SNP analysis, loci absent from the final filtered microhaplotype dataset were first removed from all per-project MADC files to ensure that downstream SNP calls were drawn from the same set of genomic regions. Individual project SNP genotypes were generated using the \u003cem\u003ereadDArTag\u003c/em\u003e function with the parameter asSNPs set to TRUE in polyRAD (v2.0.0) package. The resulting SNP datasets were merged into a single SNP VCF under the same hexaploid model. Samples not present in the final filtered microhaplotype dataset were excluded from the concatenated SNP dataset, resulting in identical sample representation across data types, regardless of marker type (e.g., microhaplotype or SNP). Except removal of monomorphic sites, no additional filters were applied on the all-SNPs set to maintain comparability to microhaplotypes. Target-SNPs was obtained from the all-SNPs and underwent the same filters as microhaplotypes. All three marker sets, including microhaplotypes, all-SNPs, and target-SNPs, comprised of 4,087 matched samples, which were then processed through equivalent PCA and DAPC pipelines.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eCreation of an R Shiny interface for standardizing unique microhaplotype ID assignment\u003c/h2\u003e \u003cp\u003eWe developed an R Shiny application, HapApp (v1.0), that provides a no-code option for users to process their MADC files, assign fixed microhaplotype IDs, and output an updated version of the microhaplotype FASTA file. The interface allows the user to customize the pipeline depending on their genotyping panel attributes such as species, microhaplotype length, and panel design (Fig.\u0026nbsp;1b). The most recent microhaplotype databases for each of the above-mentioned species (alfalfa, blueberry, cucumber, cranberry, potato pecan, strawberry, and sweetpotato) are available with the app, and updated versions retrievable by the user through the \u0026ldquo;Get Database\u0026rdquo; function in HapApp. Since the app directly uses the Bash/Python pipeline described above, our testing showed HapApp to be reliable for processing of MADC files in a no-code framework.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eOverview of microhaplotype database development\u003c/h2\u003e \u003cp\u003eFor the DArTag panels of all eight species, marker loci were broadly distributed across all chromosomes with greater density in genic regions. As described in the Materials and Methods, eight microhaplotype databases were built following the standardized approach for unique ID convention, filtering, and appending new alleles to databases. Each database consists of RefMatch and AltMatch microhaplotypes in addition to Ref and Alt microhaplotypes (54\u0026ndash;81 bp) by panel design. The eight species vary in genome complexity, heterozygosity, and ploidy levels (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). It should be noted that not all microhaplotypes are orthologous to the assay design; some paralogous microhaplotypes may also be included, especially in species with high genome duplication rates or autopolyploidy. Here, we will first provide a brief description of microhaplotype databases for four crops (cranberry, cucumber, potato, and strawberry) and highlight the annotation, usage, and advantages of using microhaplotypes over SNPs for four other crops (alfalfa, blueberry, pecan, and sweetpotato) (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eCranberry microhaplotype database\u003c/h2\u003e \u003cp\u003eThe cranberry microhaplotype database (v5) consists of 14,380 alleles from 3,050 loci (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e2\u003c/span\u003e). These were genotyped from 4,146 samples. This represents the latest and most comprehensive version (cranberry 3K DArTag v2.0) of the database, with loci evenly spread across the 12 chromosomes. This version was built on improvements made to the initial v1 panel (Chen et al. \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), addressing important issues of marker proximity and integrating new markers that cover the gapped genomic regions in the v1 panel. These refinements make the v2 panel a robust resource for studying cranberry genetics and cranberry breeding.\u003c/p\u003e \u003cp\u003eThe microhaplotype database for the cranberry v2 marker panel includes alleles from newly added markers and those retained in the v1 panel, ensuring that any previously validated information is not discarded. A total of 12,821 microhaplotypes from the retained v1 markers were integrated into the latest microhaplotype database. Summary statistics show that most loci have low levels of allelic variation (median\u0026thinsp;=\u0026thinsp;3 alleles), while the high standard deviation arises from a subset of loci with extreme variability (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Our BLAST-based classification showed that these loci were dominated by off-target best hit (mapping outside the intended site) or putative paralogs (multiple equally good hits) rather than true alleles at a single locus (Table S2). For example, the most variable marker, chr11_035249665 (448 microhaplotypes), had 419 off-target best hits plus 16 putative paralogs (435/448; 97% putative non-allelic), and chr03_024644041 showed a similar pattern (113/131; 86%). In contrast, many loci exhibited clean single-locus behavior with all microhaplotypes mapping to the intended site and no putative paralogs (e.g., chr07_017980443; 64/64 design matches; chr01_018734260: 36/36; chr07_001562431: 32/32; chr09_008835521: 28/28). Together, these observations indicated that the large variance in microhaplotype counts is driven by a small number of multi-locus amplifications, while the majority of markers remain locus specific. We flagged these putative paralogous microhaplotypes in the database since they will likely occur in future genotyping results.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eSummary statistics of the number of microhaplotypes per target marker locus for the eight microhaplotype databases\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"9\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eSpecies\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e#Loci\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"7\" nameend=\"c9\" namest=\"c3\"\u003e \u003cp\u003eMicrohaplotype statistics\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003emean\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eStandard deviation\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eminimum\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003e25%\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003e50%\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003e75%\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003emaximum\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAlfalfa\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e3,000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e160\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBlueberry\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e3,000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e270\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCranberry\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e3,050\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e448\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCucumber\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e3,059\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e256\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePecan\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e3,100\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e177\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePotato\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e3,913\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e335\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eStrawberry\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e5,000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e274\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSweetpotato\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e3,120\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e180\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eCucumber microhaplotype database\u003c/h2\u003e \u003cp\u003eThe cucumber microhaplotype database (v10) consists of 9,823 microhaplotypes derived from 8,272 genotyped samples (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e2\u003c/span\u003e). These microhaplotypes originate from 3,059 target loci (manuscript in preparation), which are evenly distributed across all seven cucumber chromosomes. The number of alleles per locus had a mean of 3, with a standard deviation of 6, indicating uneven allele diversity across surveyed loci (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). A few DArTag loci exhibited extremely high rates of polymorphism, with a maximum 256 alleles, which likely result from paralogous sequence amplification or hotspot regions of genetic variability.\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eThe cucumber microhaplotype database provides an essential resource for dissecting genomic and allelic diversity in cucumber. The genotyped samples were from several biparental F\u003csub\u003e1\u003c/sub\u003e and backcross populations, which resulted in the small number of alleles per locus (mean\u0026thinsp;~\u0026thinsp;3.2). As no diverse collections have been genotyped using the cucumber marker panel, it is not yet feasible to evaluate the genetic diversity across the cucumber genome.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003ePotato microhaplotype database\u003c/h2\u003e \u003cp\u003eThe potato microhaplotype database (v14) is comprised of 35,506 microhaplotypes across 3,913 DArTag loci (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e2\u003c/span\u003e), distributed across all 12 chromosomes of the potato reference genome DM6.1 (Pham et al. \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) and a few trait markers that are not located in the DM6.1 genome (Endelman et al. \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). These microhaplotypes were constructed from data across 3,102 samples. The number of alleles per locus had a mean of 9, with a standard deviation of 12, indicating a wide range of allele diversity across loci (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Notably, 50% of loci had six alleles (median value). A small number of loci displayed exceptionally high polymorphism rates, with a maximum of 335 alleles, potentially attributable to factors such as paralogous sequence amplification or regions of extraordinarily high genetic variability.\u003c/p\u003e \u003cp\u003eThe relatively high mean number of alleles per locus reflects substantial genetic diversity, which is consistent with the structural complexities of the potato genome and its propagation history. While clonal propagation suppresses recombination, it is highly effective at preserving existing diversity within an autotetraploid system. The wide range of allele counts emphasizes the mosaic distribution of diversity, highlighting both highly polymorphic regions and those with more limited variability across the genome. These insights will contribute to breeding programs, genetic mapping studies, and other research applications centered on potato population and genome dynamics.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eStrawberry microhaplotype database\u003c/h2\u003e \u003cp\u003eThe strawberry microhaplotype database (v3) contains 38,227 microhaplotype sequences derived from 5,000 DArTag loci (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e2\u003c/span\u003e) distributed across all 28 chromosomes (7 chromosomes with four sub-genomes: A, B, C, and D) (Hardigan et al. \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). These microhaplotypes were derived from 1,880 samples. While 50% of loci exhibited six or fewer alleles (median value), 75% of loci had 10 or fewer alleles (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). The number of alleles per locus had a mean of 8 and a standard deviation of 9, indicating notable variability in allele diversity among loci. The median value 6 indicated a right-skewed distribution driven by a small number of highly polymorphic loci (maximum allele count at 274). Such high allele counts likely result from genetic hotspots, paralogous amplification, or regions of high genetic variability.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003eAlfalfa microhaplotype database\u003c/h2\u003e \u003cp\u003eThe alfalfa microhaplotype database (v50) has reached a notable milestone, containing 35,259 microhaplotypes across 3,000 DArTag marker loci (Zhao et al. \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), averaging\u0026thinsp;~\u0026thinsp;12 alleles per locus. These microhaplotypes were discovered through 25 genotyping projects (\u0026gt;\u0026thinsp;15,600 samples) conducted by the alfalfa public breeding community, spanning diverse and structured polycross populations and breeding programs. The number of alleles per marker locus ranges from 2 to 160, with extended allele counts potentially signifying highly diverse genomic regions or paralogous amplification (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Allele diversity per locus is high compared to other species, as indicated by a mean of 12 alleles (SD\u0026thinsp;=\u0026thinsp;9) and a median of 10 alleles. Approximately 25% of loci had 2 to 6 alleles, while 75% had 15 alleles or less.\u003c/p\u003e \u003cp\u003eThe trend in microhaplotype discovery shows rapid growth in earlier database versions, followed by a plateau around ~\u0026thinsp;35,000 microhaplotypes despite the continued addition of samples (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e2\u003c/span\u003e). This plateau suggests that the current marker set (3,000 loci) has effectively captured most genetic diversity at the 3K target loci among the U.S. alfalfa breeding populations. This plateau may also reflect database saturation, such that most of the \u003cem\u003eMedicago sativa\u003c/em\u003e alleles existing across these loci have been captured. Notably, while microhaplotype discovery has plateaued, the database serves as a powerful resource for genomic analysis within cultivated alfalfa and its close relatives (Zhao et al. \u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e2024c\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAs is common in targeted genotyping platforms, not all resulting amplicons originate from the intended genomic locations due to sequence duplications within the genome. In polyploid species, the set of corresponding loci across homologous chromosomes are known as homologous groups (HGs), which share high sequence similarity and derive from the same ancestral sequences. HGs provide a framework for distinguishing true allelic variation from paralogous amplification. To this end, all 35K microhaplotype sequences were aligned to the four sub-genomes of the alfalfa cultivar \u0026lsquo;XinJiangDaYe\u0026rsquo; (Chen et al. \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Syntenic regions across homologous chromosomes were used to determine the most likely syntenic alignment if a microhaplotype could be aligned to multiple loci on one chromosome. A total of 22,159 microhaplotypes from 2,314 DArTag loci (Figures S2 \u0026amp; S3; Table S3) are likely true allelic variants (i.e., match the panel design specification), ranging from 210 in HG6 to 426 in HG4 (Table S3). A total of 1,085 DArTag loci contained potential paralogous amplifications and on average, produced more microhaplotypes per locus than the 1,915 loci containing only true allelic variants (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e3\u003c/span\u003ea). Four DArTag loci with more than 100 microhaplotypes all exhibited paralogous amplifications. For example, locus chr3.1_010956413 was located within a gene (\u003cem\u003eMS.gene044613\u003c/em\u003e), which encodes a leucine-rich repeat and nucleotide-binding-ARC domain (NBS-LRR or NLR) disease-resistance protein (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e3\u003c/span\u003eb). Among the 1,915 DArTag loci with no evidence of paralogous amplifications, 12 have more than 30 alleles each (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e3\u003c/span\u003eb; Table S3), indicating these genomic regions possess relatively high genetic diversity and/or are not under strong selection pressure.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFor the 1,915 DArTag loci without evidence of paralogous amplification, we further examined their copy numbers, defined here as the number of homologous chromosomes in which each locus is present. Among these loci, 412 (21.5%) and 1,403 (73.3%) were found in three and all four homologous chromosomes, respectively, within each of the eight homologous groups, while 73 had two copies and 27 loci had only one copy (Figure S4). Although alfalfa is tetraploid, the observed variation in copy numbers across the genome underscores the complexity of its genomic structure. Additionally, loci with fewer copies (less than 4) are likely due, or at least in part, to incomplete genome assembly rather than true biological absence.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003eBlueberry microhaplotype database\u003c/h2\u003e \u003cp\u003eThe blueberry microhaplotype database (v20) is comprised of a total of 28,653 sequences generated from 95 genotyping plates (~\u0026thinsp;8,930 samples) (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e2\u003c/span\u003e). It represents a comprehensive characterization of genetic diversity across 3,000 target loci (Zhao et al. \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2024b\u003c/span\u003e), with a genomic insert size of about 54 bp (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e), evenly distributed on all 12 chromosomes. The number of microhaplotypes per locus demonstrated a mean of ~\u0026thinsp;10 alleles, with a high variability indicated by a standard deviation of 11. Quartile statistics indicated that more than 50% of the loci contained 7 alleles or fewer. The minimum number of microhaplotypes per locus was two, while the maximum recorded was 270 alleles, showcasing a significant range in polymorphism across loci or paralogous amplifications for some loci.\u003c/p\u003e \u003cp\u003eTo characterize microhaplotypes potentially resulting from allelic vs. paralogous amplifications, all 28K microhaplotype sequences were aligned to the four sets of homologous chromosome sequences of the tetraploid blueberry cultivar \u0026lsquo;Draper\u0026rsquo;. Syntenic regions across homologous chromosomes were used to identify the most probable syntenic alignment when a microhaplotype matched multiple loci on the same chromosome. In total, 18,031 microhaplotypes from 2,535 DArTag loci (Figures S5 \u0026amp; S6; Table S4) were identified as true allelic variants. Across all 3,000 loci, 1,053 with potential paralogous amplifications tended to produce more haplotypes per locus than the 1,947 loci containing only allelic variants (Figure S7). Among the 1,947 loci without evidence of paralogous amplification, seven had more than 30 alleles each (Figure S7b; Table S4), with Chr03_005753680 and Chr07_033695728 having 82 and 81 alleles, respectively, followed by Chr05_003355218 with 52 allele (Figure S7b).\u003c/p\u003e \u003cp\u003eCopy numbers of the 1,947 loci without paralogous amplification were further analyzed. Of these, 609 (31.3%) and 850 (43.7%) had three and four copies, respectively, across the 12 homologous groups, whereas 264 loci had two copies and 224 had only one copy (Figure S8). Notably, of the 224 loci with a single copy in the reference genome, 45 (20.1%) and 67 (29.9%) were located on chromosomes 4 (VaccDscaff6) and 12 (VaccDscaff22), respectively, which agrees with previous discovery that large genomic regions on these two chromosomes lack synteny (Colle et al. \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). In addition, 66 out of 146 DArTag loci designed on chromosome 10 (VaccDscaff20) had three copies, reflecting that the distal region of chromosome 10 (VaccDscaff20) shares synteny with chromosomes 22 (VaccDscaff28) and 34 (VaccDscaff44), but not with chromosome 46 (VaccDscaff48) (Colle et al. \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), while 52 of these 146 DArTag loci had four copies with three of them in HG10 but one copy in HG6 (chromosome 30/VaccDscaff19).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003ePecan microhaplotype database\u003c/h2\u003e \u003cp\u003eThe pecan microhaplotype database (v9) is comprised of 26,073 alleles from 3,100 target loci (Chen et al.), derived from 6,768 genotyped samples (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e2\u003c/span\u003e). The loci are evenly distributed across 16 chromosomes, and each of the 3,100 loci shows diverse allele counts. Most loci exhibit moderate levels of allelic diversity, with half of the loci containing seven or fewer alleles (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). A small number of loci with extremely high allele counts may arise due to paralogous sequence amplification or hotspots of genetic variation.\u003c/p\u003e \u003cp\u003eThe complete pecan linkage map (Chen et al.) shows no large-scale structural variation (e.g., homologous sequence rearrangements) between the studied F₁ population and the reference genome. To compare the performance between microhaplotypes and SNPs, linkage maps were constructed for each of the three marker datasets: microhaplotypes, target-SNPs, and all-SNPs (target plus off-target SNPs).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eFigure 4.\u003c/b\u003e Comparison of marker ordering and filtering outcomes for pecan across three marker datasets (microhaplotypes, target SNPs, target plus off-target SNPs). (a) Summary of marker filtering by genotype class across the three datasets. Bars represent the number of markers per genotype class removed by specific filters applied in the presented order (starting with depth to rf_filter). (b) Linkage marker ordering for chromosome 8 using genome position (left panels) and multidimensional scaling (MDS) genetic distance (right panels).\u003c/p\u003e \u003cp\u003eFiltering statistics further highlighted key differences among the three marker sets (Fig.\u0026nbsp;4a, Table S5). The all-SNPs dataset showed a high number of loci removed by the redundancy filter, which is expected given their proximity to target markers (within 81 bp). The microhaplotype dataset retained a larger number of markers capable of tracking recombination events and providing information for both parent\u0026rsquo;s meiosis. In contrast, the target-SNPs and all-SNPs datasets contained mostly ab \u0026times; aa markers (single-parent informative) and very few ab \u0026times; ab markers (informative for both parents) after filtering. These patterns were consistent across all 16 chromosomes, with chromosome 8 being the most prominent (Fig.\u0026nbsp;4b and S9). Therefore, in the target-SNPs and all-SNPs datasets, the aa \u0026times; ab markers were frequently ordered independently and placed far from their expected genomic positions using the MDS method since not enough informative markers for both parents (heterozygous \u0026times; heterozygous) were present to relate single-parent informative markers (aa \u0026times; ab and ab \u0026times; aa).\u003c/p\u003e \u003cp\u003eThe all-SNPs dataset showed a greater number of gaps when linkage maps were ordered according to the reference genome, indicating a higher incidence of genotyping errors among the retained markers. In contrast, the dataset containing only target SNPs demonstrated fewer gaps in the reference-based linkage map, suggesting lower genotyping error rates compared to the all-SNPs set. However, it contained relatively few markers informative for both parents, which impaired ordering performance using MDS algorithm. The microhaplotype dataset retained more loci throughout the filtering pipeline, had fewer gaps, and preserved a higher proportion of markers informative for both parents. These attributes contributed to more robust ordering when applying the MDS algorithm.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003eSweetpotato microhaplotype database\u003c/h2\u003e \u003cp\u003eThe sweetpotato DArTag panel is a key molecular tool that was developed and applied in various collaborative projects to advance sweetpotato (2n\u0026thinsp;=\u0026thinsp;6x\u0026thinsp;=\u0026thinsp;90) genetics and breeding worldwide (Zhao et al. \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2024a\u003c/span\u003e). The latest version of the associated microhaplotype database (v19) consists of 37,593 sequences across 3,120 genomic loci, which are evenly distributed across the 15 chromosomes (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e2\u003c/span\u003e). These microhaplotypes were derived from genotyping data of 9,212 individual samples. The number of alleles per locus ranges from 2 (minimum) to 180 (maximum), with an average of 12 alleles per locus (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). The median number of alleles per locus is 9, and the interquartile ranges between 6 (25th percentile) and 14 (75th percentile). The wide distribution of alleles suggests the markers effectively capture broad genetic variations.\u003c/p\u003e \u003cp\u003eThe versatility and efficacy of the sweetpotato marker panel are evidenced by its extensive adoption across breeding programs from multiple continents. Participating programs include those in the U.S., Mozambique (Africa), Korea and Taiwan (Asia), Peru (South America), Jamaica (Central America), and Costa Rica (Caribbean/Central America) (Fig.\u0026nbsp;5A).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eFigure 5.\u003c/b\u003e Geographic context, cross validation accuracy and principal component analysis (PCA) of microhaplotypes and SNP markers in sweetpotato. (a) Geographic distribution of genotyping projects using the sweetpotato 3K DArTag panel. (b) Discriminant analysis of principal components (DAPC) cross-validation accuracies using 5\u0026ndash;50 PCs for microhaplotypes, target-SNPs, and all-SNPs marker sets. (c and d) PCA plots for PC1 vs. PC2 (c) and PC2 vs. PC3 (d) of the microhaplotype dataset. (e and f) PCA plots for PC1 vs. PC2 (e) and PC2 vs. PC3 (f) of the target-SNPs dataset. (g and h) PCA plots for PC1 vs. PC2 (g) and PC2 vs. PC3 (h) of the all-SNPs dataset.\u003c/p\u003e \u003cp\u003eWe evaluated the performance of three marker sets on population genetics inference in sweetpotato: microhaplotypes, target-SNPs, and all-SNPs (target\u0026thinsp;+\u0026thinsp;off-target). To ensure high-quality genotype data while retaining informative low-frequency and private alleles, we implemented a population-aware filtering strategy in which a microhaplotype was retained if it met either a global missingness threshold of \u0026le;\u0026thinsp;20% across all samples, or a per-population missingness threshold of \u0026le;\u0026thinsp;20% in at least one population. After filtering for marker missingness and alternative allele frequency (AAF) of \u0026gt;\u0026thinsp;0.05, a total of 2,772 (27,220 microhaplotypes) out of 3,119 marker loci (28,338 microhaplotypes), and 4,087 samples with \u0026lt;\u0026thinsp;20% missing data were kept. The target-SNPs included the target SNPs present within the 2,772 retained loci. The same filters were applied to the target SNPs, resulting in 2,534 SNPs. The all-SNPs are comprised of all SNPs in the 2,772 marker loci, resulting in a total of 21,002 SNPs across 4,087 samples. After removing monomorphic sites in all-SNPs, 20,935 markers remained. Because these SNPs were extracted directly from the retained microhaplotypes, no additional filtering was applied beyond removal of monomorphic sites.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eProject-level private microhaplotypes and marker information in sweetpotato\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eProjects\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e#Samples\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e#Private_microhaplotypes\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e#Markers with private microhaplotypes\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCosta Rica\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e165\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e261\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e140\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eJamaica\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e72\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMozambique\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e155\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e27\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e27\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePeru\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e794\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e274\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e214\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSouth Korea\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e191\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e440\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e364\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTaiwan\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1,352\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e7,340\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2,233\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUSA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1,358\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e367\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e308\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eTo evaluate how well the retained panel captures population-specific variation, we enumerated project-specific microhaplotypes among the retained loci (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Across the seven projects, we identified 8,711 project-specific microhaplotypes, spanning 3,288 marker-project instances in which at least one private microhaplotype occurred. Normalized by sample size, private microhaplotypes per sample were 0.03 (2/72) in Jamaica, 0.17 (27/155) in Mozambique, 0.27 (367/1,358) in the USA, 0.35 (274/794) in Peru, 1.58 (261/165) Costa Rica, 2.3 (440/191) in South Korea, and 5.43 (7,340/1,352) in Taiwan. These results indicate substantial heterogeneity in project-private variant across populations, with Taiwan and South Korea showing the highest densities. This pattern is consistent with the stronger separations observed in the PCA (below).\u003c/p\u003e \u003cp\u003eOur comparative analysis revealed that microhaplotypes consistently provided a more efficient and stable representation of population structure than either SNP set. Principal component analysis (PCA) revealed marked differences in the resolution of population structure (Fig.\u0026nbsp;5c-h). In PC1-PC2 space, the microhaplotype PCA displayed visibly tighter clusters for each population with minimal overlaps, with Taiwan strongly differentiated along PC1 and subtler separations among other populations along PC2. In contrast, the SNP-based PCA exhibited less distinct separation with noticeable overlap among several populations, particularly Costa Rica, Mozambique, South Korea, and Taiwan. Similar trends were observed in PC2-PC3 space. The variance explained by the leading PCs further highlighted these differences. For microhaplotypes, PC1, PC2, and PC3 explained 15.72%, 3.45%, and 2.3% of the total variance, respectively, compared to 10.5%, 7.56%, and 2.6% for target-SNPs, and 5%, 4.24%, and 1.76% for all-SNPs. Therefore, PC1 of the microhaplotypes captured more variance compared with the SNP equivalents, suggesting stronger separation along the most informative axis.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eEigenvalue distribution and proportion of variance explained by the three marker sets and linear discriminant (LD) axis in sweetpotato populations.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eAxis\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e \u003cp\u003eeigenvalue\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c7\" namest=\"c5\"\u003e \u003cp\u003eFractions\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMicrohaplotypes\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAll_SNPs\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTarget_SNPs\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eMicrohaplotypes\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAll_SNPs\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eTarget_SNPs\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eLD1\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e854,002\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e215,300\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e112,056\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e86.26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e60.61\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e60.28\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eLD2\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e74,596\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e89,897\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e47,587\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e7.53\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e25.31\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e25.6\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eLD3\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e41,989\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e35,015\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e16,007\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e4.24\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e9.86\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e8.61\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eLD4\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e16,736\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e12,488\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e8,395\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.69\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e3.52\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e4.52\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eLD5\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1,921\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1,731\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1,237\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.49\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.67\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eLD6\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e805\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e784\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e618\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.33\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eDiscriminant analysis of principal components (DAPC) further highlighted the discriminatory advantage of microhaplotypes (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). Cross validation showed that 86.3% of between-group variance was captured by the first linear discriminant axis (LD1) for microhaplotypes compared with approximately 60% for both SNP sets, an improvement of over 25%. The LD1 eigenvalue was correspondingly larger for microhaplotypes (854,002) than for all-SNPs (215,300), and target-SNPs (112,056), reflecting tighter cluster resolution. Despite these differences in dimensional efficiency, all three marker types achieved very high classification accuracies (mean success rate\u0026thinsp;\u0026gt;\u0026thinsp;98%; ANOVA: F\u0026thinsp;=\u0026thinsp;1.495, p\u0026thinsp;=\u0026thinsp;0.225) (Fig.\u0026nbsp;5b). At low dimensionality (5 PCs), targeted-SNPs achieved the highest early success rate (84.3%), followed by microhaplotypes (80.2%) and all-SNPs (77.7%), likely reflecting the pre-selected informativeness of target SNPs. As the number of retained PCs increased, target-SNPs accuracy degraded starting around 40 PCs whereas microhaplotypes maintained a stable high-success plateau near 99%, suggesting greater robustness to model complexity.\u003c/p\u003e \u003cdiv id=\"Sec23\" class=\"Section3\"\u003e \u003ch2\u003eComparative insights across all eight databases\u003c/h2\u003e \u003cp\u003eMicrohaplotypes are sets of closely linked SNP sites within a small region of the genome (assumed not to recombine), making them a powerful tool for genotyping and genetic diversity studies. The development of microhaplotype databases across multiple crops has enabled the discovery of rich allelic diversity while providing the plant germplasm, breeding, and genetics community with resources to dissect functional diversity, population structure, and trait architecture using cost-effective targeted genotyping.\u003c/p\u003e \u003cp\u003eThe microhaplotype databases for these eight species were developed with varying numbers of genotyped samples (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Factors such as the scale of genotyping, species-specific genome complexity, marker diversity, and the genomic insert size used during genotyping library preparation impacted the resulting resources. Unsurprisingly, longer genomic insert sizes (e.g., 81 bp) tend to capture more variant sites compared to shorter insert sizes (e.g., 54 bp), allowing for more variants to be detected per amplicon locus. For example, species like sweetpotato and pecan, which utilize 81 bp panels, show substantial allele diversity relative to their genomic complexity, partly because of the longer genomic region available for detecting polymorphisms.\u003c/p\u003e \u003cp\u003eAll eight microhaplotype databases were developed by following a standardized pipeline that includes stringent ID conventions, filtering, and sequential updates to maintain biological relevance. Notably, the databases include both orthologous (i.e., microhaplotypes originated from conserved loci between divergent species) and paralogous (i.e., related loci within a genome, product of duplication) microhaplotypes, the latter of which are more common in complex genomes with higher levels of duplications. The retention of paralogous alleles may seem counterintuitive at first glance, since they cannot be used in breeding. However, because these alleles are likely to be encountered in future genotyping experiments, it was important to codify them and append a confidence rating that indicates to the users that they are likely paralogous alleles.\u003c/p\u003e \u003cp\u003eAmong the databases, alfalfa (autotetraploid), sweetpotato (hexaploid), and blueberry (autotetraploid) exhibit the highest genetic diversity, with average allele counts per locus of 12, 12, and 10, respectively. The polyploid nature of these crops likely contributes to their exceptional allelic richness. Furthermore, these three species also represent the largest genotyping scales undertaken on the respective DArTag panels. The alfalfa database, based on 167 plates of genotyped samples (over 15,600 samples), stands out as the most extensive. Alfalfa\u0026rsquo;s saturation plateau (~\u0026thinsp;35,000 microhaplotypes) demonstrates that the high genotyping scale successfully captures most genetic diversity at the 3K target loci within the studied populations. For sweetpotato, the global importance, spanning Africa, Asia, and the Americas, is mirrored in the diversity captured in its database, which encompasses samples from 9,212 samples and aligns with a worldwide population structure.\u003c/p\u003e \u003cp\u003eIn contrast, cucumber exhibited the narrowest genetic base among the species examined. With a mean of 3 alleles per locus and 75% of loci containing no more than 4 alleles, which likely reflects sampling from a single breeding program and/or inherently limited diversity in cucumber relative to the other species analyzed. Consistent with this, prior work indicates that wild sexually compatible relatives could enhance the genetic diversity of cultivated cucumber (Zhang et al. \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Abdel-Salam et al. \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Tirnaz et al. \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). On the other hand, the octoploid strawberry database reveals rich allele diversity with a mean of 8, despite a relatively smaller genotyping scale (20 plates). The high allelic diversity evidenced here traces back to the selection of sub-genome-specific loci for developing the marker panel (Hardigan et al. \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Some additional allelic richness may also be attributed to the complex genome structure of strawberry, which may compensate for the smaller sample size.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec24\" class=\"Section2\"\u003e \u003ch2\u003eData accessibility and community engagement\u003c/h2\u003e \u003cp\u003eIn alignment with the FAIR principles (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.go-fair.org/fair-principles/\u003c/span\u003e\u003cspan address=\"https://www.go-fair.org/fair-principles/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), all microhaplotype sequences will be made publicly available. The allele data, metadata (including project and principal investigator (PI) information, and associated documentation will be hosted through globally accessible and FAIR-compliant platforms, HapApp (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps:github.com:Breeding-Insight/HapApp_utils\u003c/span\u003e\u003cspan address=\"https:github.com:Breeding-Insight/HapApp_utils\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) and HapSearch (under development). Users who are interested in obtaining samples or germplasm linked to specific alleles must follow established access procedures. Interested users will need to contact the relevant PIs and adhere to the AgCommons Framework (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://agdatacommons.nal.usda.gov/\u003c/span\u003e\u003cspan address=\"https://agdatacommons.nal.usda.gov/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), or a similar standard protocol for germplasm access. The PI has the right to share the resources while respecting intellectual property rights, mutually agreed terms, and proper documentation.\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe eight curated microhaplotype databases developed here provide a powerful, standardized framework for advancing genetic diversity studies and breeding programs across a diverse set of crops. Through consistent microhaplotype ID conventions, filtering pipelines, and sequential updates, these resources capture both orthologous and documented paralogous microhaplotypes, ensuring broad relevance to future genotyping. Microhaplotypes deliver richer allelic diversity and greater resolution than bi-allelic SNPs, a benefit especially evident in complex, polyploid genomes such as alfalfa, blueberry, and sweetpotato. Our results demonstrate their advantages for retaining informative markers, improving linkage map ordering, and resolving fine-scale population structure.\u003c/p\u003e \u003cdiv id=\"Sec26\" class=\"Section2\"\u003e \u003ch2\u003eOrthologous and paralogous microhaplotypes\u003c/h2\u003e \u003cp\u003eWhile most loci exhibited low to moderate microhaplotype counts, the extremely high allele counts observed at some loci are likely due to paralogous sequence amplification, in addition to natural orthologous genetic diversity. Paralogous sequences arise from duplicated regions within the genome, which can be unintentionally captured and amplified during genotyping due to sequence similarity. These loci can be important for identifying hotspots of variation, but they may also reflect technical artifacts rather than true orthologous diversity.\u003c/p\u003e \u003cp\u003eComparative patterns in alfalfa and blueberry microhaplotypes illustrated how genome structure shapes these metrics. Though both species have a similar number of loci without paralogous amplification (1,915 and 1,947 loci, respectively), their copy number distributions differ (Tables S1 \u0026amp; S2). In alfalfa, 94.8% of the non-paralogous loci occur in three or four copies (Figure S4), matching the expected dosage for a largely autotetraploid species. In contrast, only 74.9% of blueberry loci fall into this range. Instead, blueberry shows substantially higher proportions of single-copy (20.1%) and two-copy (29.9%) loci (Figure S8). This elevated number of reduced-copy loci in blueberry suggests more extensive post-whole genome duplication structural modifications than that observed in alfalfa. These patterns are consistent with current evidence that highbush blueberry originated through a complex polyploidization process involving hybridization among multiple diploid \u003cem\u003eVaccinium\u003c/em\u003e species followed by genome duplication and subsequent structural divergence (Colle et al. \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Mengist et al. \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). The cultivar \u0026lsquo;Draper\u0026rsquo; is predominantly \u003cem\u003eV. corymbosum\u003c/em\u003e with minor introgressions (\u0026lt;\u0026thinsp;5%) from \u003cem\u003eV. tenellum\u003c/em\u003e, \u003cem\u003eV. ashei\u003c/em\u003e, and \u003cem\u003eV. darrowii\u003c/em\u003e, which may help explain part of the observed pattern (Hancock \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2004\u003c/span\u003e). The unique inter-chromosome translocation between chromosomes 6 and 10 could affect pairing and recombination of these two chromosomes in \u0026lsquo;Draper\u0026rsquo; but was not observed in a wild diploid relative \u0026lsquo;W85\u0026rsquo; (Mengist et al. \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) or reported in other tetraploid highbush blueberries.\u003c/p\u003e \u003cp\u003eMost repetitive sequences in the alfalfa \u0026lsquo;XinJiangDaYe\u0026rsquo; reference genome are composed of long terminal repeat (LTR) retrotransposons, consistent with previous observations in other legume species (Chen et al. \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). However, because DArTag markers are predominantly designed within genic regions, most cases of paralogous amplification in our dataset are not likely attributed to LTR-derived repeats. Notably, we identified a highly paralogous marker (chr3.1_010956413) with 105 microhaplotypes (rank of 3) located within \u003cem\u003eMS.gene044613\u003c/em\u003e, an NLR disease resistance protein. This observation aligns with earlier findings in \u003cem\u003eM. truncatula\u003c/em\u003e, a wild relative of cultivated alfalfa, in which a dense cluster of NLR genes was reported on chromosome 3.1 (Ameline-Torregrosa et al. \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2008\u003c/span\u003e). Together, these results suggest that local gene family expansion could play a role in generating paralogous signals in DArTag genic markers.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec27\" class=\"Section2\"\u003e \u003ch2\u003eMicrohaplotypes strengthen linkage maps through high information content and reduced phasing errors\u003c/h2\u003e \u003cp\u003eComparison of linkage maps created using microhaplotypes vs. SNPs highlighted the advantages of microhaplotypes in providing higher information content and reducing ambiguity in phase estimation (Fig.\u0026nbsp;4). Because target and off-target SNPs fall within a 54\u0026ndash;81 bp read, variants within each amplicon are physically linked and phased directly from reads, making recombination within an amplicon effectively negligible. Thus, part of the advantage of microhaplotypes reflecting collapsing these co-inherited SNPs into a single multi-allelic locus, which reduces redundancy and increases information content. Their performance depends primarily on read depth and phasing quality rather than population LD. Due to their multi-allelic nature, microhaplotypes increased the number of markers heterozygous in both parents, thereby strengthening linkages among single-dose variant classes such as ab \u0026times; aa and aa \u0026times; ab. In bi-allelic datasets, the rare ab \u0026times; ab class was the sole marker type capable of connecting these groups, limiting integration of parental maps and reducing overall connectivity. Consequently, both target-SNPs and all-SNPs (target plus off-target) performed poorly during marker ordering due to the limited number of markers informative for recombination in both parents.\u003c/p\u003e \u003cp\u003eMicrohaplotypes also reduce ambiguity in phase estimation. For multi-allelic configurations such as ab \u0026times; cd, the parental origin of progeny alleles is directly observable, decreasing the reliance on Expectation Maximization (EM) inference and lowering the probability of phasing errors. In contrast, bi-allelic datasets contain ambiguous configurations (e.g., ab \u0026times; ab) requiring EM inference, increasing the risk of incorrect phasing.\u003c/p\u003e \u003cp\u003eThe all-SNPs set showed the poorest mapping performance, with incorrect ordering for most chromosomes and a higher proportion of erroneous markers retained after filtering. This was likely due to the physical proximity of off-target sites to their corresponding target sites, resulting in their removal as redundant during early filtering. The off-target markers that were not removed tend to be more error-prone, and if they are not eliminated by subsequent filters such as segregation distortion, they contribute to the larger number of gaps observed in the final maps. Problematic markers can be filtered out by increasing the stringency of the rf_filter parameter, but good-performing markers can also be removed, especially in datasets with few multi-allelic or highly informative markers. In contrast, the microhaplotype strategy integrates off-target information into multi-allelic markers, which reduces redundancy and enhances the effectiveness of downstream filtering. This highlights the need for careful parameter tuning, especially in datasets with fewer multi-allelic or highly informative markers.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec28\" class=\"Section2\"\u003e \u003ch2\u003eMicrohaplotypes provide more efficient and informative separation of populations than SNPs\u003c/h2\u003e \u003cp\u003eMicrohaplotypes inherently capture multiple variants within a small genomic region and retain linkage phase information, which increases their average heterozygosity and allelic count per locus compared to SNPs. These properties underpin their growing adoption not only in human forensic and ancestry inference applications but also in the plant genetics and breeding realm. The comparative analysis of microhaplotype and SNP data analyzed independently in hexaploid sweetpotato revealed notable differences in allelic diversity, variance captured, and population resolution. The resulting extra diversity using microhaplotypes means that patterns differentiating populations are stronger per locus, which is consistent with findings from other systems.\u003c/p\u003e \u003cp\u003eIn human genetics, highly polymorphic microhaplotype panels have demonstrated strong performance in relationship detection. For example, a set of 54 microhaplotypes demonstrated high reliability for first-degree relationship detection, approaching the performance of an established forensic panel comprising 27 short tandem repeats (STRs) plus 94 SNPs (Wu et al. \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Complementary evidence from ancestry inference research identified 120,000 microhaplotypes from nearly one million SNPs and across multiple benchmarks, microhaplotype subsets consistently outperformed SNP panels in resolving ancestry, particularly in complex or admixed populations (Turchi et al. \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2026\u003c/span\u003e). Taken together, these cross-disciplinary insights point toward the value of fully utilizing microhaplotypes in plant breeding, particularly highly complexed polyploid species. These advantages are directly relevant for dosage-aware genomic prediction in polyploids. By encoding each microhaplotype as a copy-number dosage (0\u0026ndash;2 for diploids; 0\u0026ndash;4 for tetraploids, etc.), multi-allelic loci provide richer predictors than biallelic SNPs, capturing local microhaplotype effects and within-locus interactions. Such dosages can be used to build microhaplotype-based relationship matrices for GBLUP or included as regressors in Bayesian whole-genome model to capture more additive variance and reduce phase-related ambiguity compared with individual SNPs.\u003c/p\u003e \u003cp\u003eFrom a methodological perspective, the implementation of a population-aware filtering strategy addresses a challenge in population genetics by ensuring high data quality without discarding private alleles. This way, we preserved genetic signals relevant for local adaptation, breeding history, and germplasm differentiation even when these microhaplotypes were scarce across the global dataset. One advantage of applying this strategy is the preservation of thousands of private microhaplotypes in the sweetpotato Taiwan population, which allowed its clear separation from other populations. The distinct clustering of Taiwanese accessions in the PCA may reflect the long history of sweetpotato breeding in Taiwan. Repeated cycles of crossing and selection within local breeding programs, together with the use of shared founder lines, may have resulted in a relatively distinct breeding pool. In addition, region-specific breeding objectives and relatively limited germplasm exchange with other regions may have further contributed to divergence in allele frequencies.\u003c/p\u003e \u003cp\u003eThe utility of the microhaplotypes has also been demonstrated in alfalfa by others. In a study of 28 populations, Medina et al. (\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) revealed significant population structure based on geographical origin, higher genetic diversity values compared to traditional SNP markers, excess heterozygosity (negative FIS values) in 27 out of 28 populations, and minimal inbreeding in founding populations (Medina et al. \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAnother important application of microhaplotypes is for curation and management of plant genetic resource (PGR) collections. They can be used to: (i) resolve population structure and quantify diversity within and among collections; (ii) identify redundancy and gaps to optimize regeneration, safety-duplication, and sampling strategies; (iii) verify accession identity and detect mixtures or mislabeling to improve passport data and taxonomic assignments; and (iv) define representative core subsets for evaluation and pre-breeding. The standardized IDs and cross-project comparability enable genebanks to benchmark collections across repositories and over time, directly supporting trait discovery and the deployment of novel variation into breeding pipelines.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec29\" class=\"Section2\"\u003e \u003ch2\u003eImplications of microhaplotype analysis and future directions\u003c/h2\u003e \u003cp\u003eLooking forward, the integration of microhaplotypes into both basic and applied research has the potential to markedly improve analytical efficiency and resolution. The efficiency of microhaplotype-based analyses in low-dimensional space makes them particularly attractive for targeted genotyping in resource-limited breeding programs, where computational capacity and sequencing resources are constrained. At present, relatively few tools can fully accommodate multi-allelic data, particularly in complex polyploid species (Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). Expanding the availability and performance of software that can handle multi-allelic datasets in both diploid and polyploid contexts will greatly accelerate the integration of microhaplotypes into breeding decision-making pipelines. Such tool development will enable breeders and researchers to fully leverage the linkage phase information and high allelic diversity inherent to microhaplotype markers.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eList of computational tools available with multi-allelic marker support.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTool Name\u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAnalysis\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSupport Polyploids\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCitation\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePolysat\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePopulation Diversity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(Clark and Jasieniuk \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Clark and Schreier \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2017\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePLINK 2.0\u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHWE, LD, Relationship matrix, GWAS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(Chang et al. \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2015\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003epolyRAD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGenotype Calling\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(Clark et al. \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2019\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRAINBOWR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGWAS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(Hamazaki and Iwata \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2020\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003empQTL\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGWAS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(Th\u0026eacute;r\u0026egrave;se Navarro et al. \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2022\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOneMap\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLinkage Mapping\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(Taniguti et al. \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2022\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eADAM-multi\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePopulation Simulation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(Chu and Jensen \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2025\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003csup\u003ea\u003c/sup\u003e All described tools also support highly heterozygous species.\u003c/p\u003e \u003cp\u003e \u003csup\u003eb\u003c/sup\u003e As of January 10, 2026 release\u003c/p\u003e \u003cp\u003eThe current framework is designed to be species-agnostic and expandable. With HapApp, users can create, update, and maintain species-specific microhaplotype databases for taxa not included in this study by processing MADC files, assigning fixed microhaplotype IDs, and producing updated database. To make the microhaplotype databases more amenable, we are actively developing a user-friendly platform, HapSearch, which could serve as an interactive bridge between raw genomic information and practical breeding decisions. With HapSearch, researchers and breeders can rapidly locate, evaluate, and act upon relevant allelic variations captured in the databases. The envisioned features of HapSearch would enhance the value of microhaplotype resources for plant breeding and genetics. First, the system will allow targeted mining of alleles of interest by crop species, sequence similarity, or project-specific keywords. Second, integrated filtering functions will enable users to narrow search results to specific locus. Multiple sequence alignment will reveal all existing variations for a locus and allow users to identify accessions carrying unique microhaplotype alleles. Third, to promote collaboration and resource exchange, the platform could provide access to PI or germplasm collection curator contact information, enabling rapid follow-up for germplasm acquisition or data verification. The suite of platforms from processing and maintaining microhaplotype databases to user-friendly system for mining microhaplotypes will greatly accelerate the translation of genotypic diversity into practical genetic gain.\u003c/p\u003e \u003c/div\u003e"},{"header":"Conclusion","content":"\u003cp\u003eWe present standardized microhaplotype databases for eight crops, a species‑agnostic, reproducible pipeline, and a no‑code HapApp for assigning fixed IDs and updating records from a targeted amplicon genotyping platform, DArTag. HapApp can also be applied to other targeted genotyping platforms (e.g., GT-seq, AgriSeq, FlexSeq) by pre-processing the raw FASTQ data into MADC-like format. These resources capture extensive allelic diversity, including orthologous and paralogous signals, and provide higher resolution than SNPs for population structure, linkage mapping, and genotype-phenotype analyses. By retaining more informative markers and reducing phasing ambiguity, microhaplotypes are especially powerful in polyploid and highly heterozygous crops and remain efficient for resource‑limited programs. The databases are FAIR and extensible; community updates and the forthcoming HapSearch interface will make allele mining and cross‑project comparisons routine, strengthening PGR curation and accelerating genetic gain across crops.\u003c/p\u003e "},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research was supported through Breeding Insight (RRID: SCR_026645), a USDA-ARS initiative previously hosted by Cornell University under Cooperative Agreements (\u003cu\u003e8062-21000-043-004-A\u003c/u\u003e, \u003cu\u003e8062-21000-052-002-A\u003c/u\u003e, and \u003cu\u003e8062-21000-052-003-A\u003c/u\u003e) and currently hosted at the\u0026nbsp;University of Florida, Gainesville, under a Cooperative Agreement (8062-21000-052-020-A).\u003c/p\u003e\n\u003cp\u003eAdditional support was provided by USDA ARS CRIS Project 2072-21000-059-000D and by the NIFA Alfalfa Seed and Alfalfa Forage Systems Research Program (awards 2019-70005-30361 and 2023-70005-41081 to ECB, BI, HR); the CGIAR Trust Fund contributors (https://www.cgiar.org/funders/ to RS, MK, HLK, and MD); the Bill and Melinda Gates Foundation through the SweetGAINs (OPP1213329) and the RTB breeding (INV-0411050) investments (GCY and SPF); the\u0026nbsp;Cooperative Project No. (to be assigned) Rural Development Administration, Republic of Korea.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDisclaimer\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by\u0026nbsp;USDA-ARS Projects 5062-21500-001-000D,\u0026nbsp;2019-70005-30361 and 2023-70005-41081.\u0026nbsp;Mention of any trade names or commercial products in this article is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the U.S. Department of Agriculture. USDA is an equal opportunity provider and employer,\u0026nbsp;and all agency services are available without discrimination.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors have no relevant financial or non-financial interests to disclose.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDZ: Conceptualization; Data curation; Formal Analysis; Investigation; Methodology; Project administration; Software; Supervision; Validation; Visualization; Writing – original draft preparation. ML: Data curation; Formal Analysis; Investigation; Methodology; Visualization; Writing – original draft preparation. CHT: Data curation; Formal Analysis; Investigation; Methodology; Visualization; Writing – original draft preparation. AMS: Software; Visualization; Writing – original draft preparation. SC: Visualization. DAS: Funding acquisition; Investigation; Resources. ZX, ECB, BMI, HR, DAS, DS, NB, SJC, LMH, EB, JP, JHP, JL, MH, SPF: Resources. CAM: Methodology. XW and AH: Investigation; Resources; Validation. WC: Funding acquisition; Investigation; Resources; Supervision; Validation. PAW: Funding acquisition; Resources. GCY: Funding acquisition; Resources; Investigation; Project administration; Supervision; Validation; SAW: Investigation; Resources; Project administration; Supervision; Validation; THK: Funding acquisition; Resources; RS: Resources; Investigation; Methodology; MK: Conceptualization; Resources; Methodology. MD: Conceptualization; Resources; Investigation; Methodology. HLK: Funding acquisition; Resources; Project administration; Supervision. SYC: Resources; Investigation; Project administration; Supervision. PHL: Resources; Investigation; Project administration. CWL: Resources; Investigation. JRC: Funding acquisition; Resources; Project administration; Supervision. FJM: Resources; Investigation. RCV: Resources; Investigation. CTB and MJS: Funding acquisition; Project administration; Supervision. All authors participated in manuscript Writing – editing. All authors read and approved the final version of the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe gratefully acknowledge our collaborators for generously sharing germplasm for genotyping and for contributing metadata, coordination, and logistical support essential to the development of the microhaplotype database. Jeff Neyhart, Juan Zalapa, and Eric Weisman for their work with cranberry; Yiqun Weng and Savannah Beyer for their work with cucumber; Max Feldman, Noelle Anglin, Xiaohong Wang, Mercedes Ames, and Dennis Halterman for their work with potato; and the USDA-ARS Corvallis, OR location for their support with strawberry. Their openness in sharing material and data made this work possible. We also sincerely thank them for their years of partnership and collaboration with Breeding Insight, which continues to advance, innovate, and provide impact across our shared breeding communities.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eScripts and pipeline for building microhaplotype databases are deposited at GitHub (https://github.com/Breeding-Insight/HapApp_utils; https://github.com/Breeding-Insight/HapApp). The microhaplotype database files of the eight crop species can be downloaded from Zenodo (https://doi.org/10.5281/zenodo.19338534).\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAbdel-Salam EM, Faisal M, Alatar AA, et al (2020) Comparative analysis between wild and cultivated cucumbers reveals transcriptional changes during domestication process. Plants 9:63. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/plants9010063\u003c/span\u003e\u003cspan address=\"10.3390/plants9010063\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAbed A, Belzile F (2019) Comparing single-SNP, multi-SNP, and haplotype-based approaches in association studies for major traits in barley. Plant Genome 12:1\u0026ndash;14. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3835/plantgenome2019.05.0036\u003c/span\u003e\u003cspan address=\"10.3835/plantgenome2019.05.0036\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAmeline-Torregrosa C, Wang B-B, O\u0026rsquo;Bleness MS, et al (2008) Identification and characterization of nucleotide-binding site-leucine-rich repeat genes in the model plant Medicago truncatula. Plant Physiol 146:5\u0026ndash;21. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1104/pp.107.104588\u003c/span\u003e\u003cspan address=\"10.1104/pp.107.104588\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAnderson EC, Clemento AJ, Campbell MA, et al (2025) A multipurpose microhaplotype panel for genetic analysis of California Chinook salmon. Evol Appl 18:e70110. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/eva.70110\u003c/span\u003e\u003cspan address=\"10.1111/eva.70110\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBaetscher DS, Clemento AJ, Ng TC, et al (2018) Microhaplotypes provide increased power from short-read DNA sequences for relationship inference. Mol Ecol Resour 18:296\u0026ndash;305. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/1755-0998.12737\u003c/span\u003e\u003cspan address=\"10.1111/1755-0998.12737\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBlyton MDJ, Brice KL, Heller-Uszynska K, et al (2023) A new genetic method for diet determination from faeces that provides species level resolution in the koala. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1101/2023.02.12.528172\u003c/span\u003e\u003cspan address=\"10.1101/2023.02.12.528172\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBotstein D, White RL, Skolnick M, Davis RW (1980) Construction of a genetic linkage map in man using restriction fragment length polymorphisms. Am J Hum Genet 32:314\u0026ndash;331\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCamacho C, Coulouris G, Avagyan V, et al (2009) BLAST+: architecture and applications. BMC Bioinformatics 10:421. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/1471-2105-10-421\u003c/span\u003e\u003cspan address=\"10.1186/1471-2105-10-421\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCampbell NR, Harmon SA, Narum SR (2015) Genotyping-in-Thousands by sequencing (GT-seq): A cost effective SNP genotyping method based on custom amplicon sequencing. Mol Ecol Resour 15:855\u0026ndash;867. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/1755-0998.12357\u003c/span\u003e\u003cspan address=\"10.1111/1755-0998.12357\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChang CC, Chow CC, Tellier LC, et al (2015) Second-generation PLINK: rising to the challenge of larger and richer datasets. Gigascience 4:7. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s13742-015-0047-8\u003c/span\u003e\u003cspan address=\"10.1186/s13742-015-0047-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChang W, Cheng J, Allaire JJ, et al (2024) shiny: Web Application Framework for R\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen H, Zeng Y, Yang Y, et al (2020) Allele-aware chromosome-level genome assembly and efficient transgene-free genome editing for the autotetraploid cultivated alfalfa. Nat Commun 11:1\u0026ndash;11. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41467-020-16338-x\u003c/span\u003e\u003cspan address=\"10.1038/s41467-020-16338-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen S, Lin M, Taniguti CH, et al (2025) A public mid-density genotyping platform for cultivated cranberry (Vaccinium macrocarpon Aiton). Plant Genome 18:e70118. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/tpg2.70118\u003c/span\u003e\u003cspan address=\"10.1002/tpg2.70118\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen S, Lin M, Zhao D, et al A public mid-density genotyping platform for pecan (Carya illinoinensis (wangenh.) Koch). In preparation\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChu TT, Jensen J (2025) ADAM-multi: software to simulate complex breeding programs for animals and plants with different ploidy levels and generalized genotypic effect models to account for multiple alleles. Front Genet 16:1513615. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fgene.2025.1513615\u003c/span\u003e\u003cspan address=\"10.3389/fgene.2025.1513615\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eClare SJ, Driskill M, Millar TR, et al (2024) Development of a targeted genotyping platform for reproducible results within tetraploid and hexaploid blueberry. Front Hortic 2:1339310. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fhort.2023.1339310\u003c/span\u003e\u003cspan address=\"10.3389/fhort.2023.1339310\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eClark LV, Jasieniuk M (2011) POLYSAT: an R package for polyploid microsatellite analysis. Mol Ecol Resour 11:562\u0026ndash;566. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1755-0998.2011.02985.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1755-0998.2011.02985.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eClark LV, Lipka AE, Sacks EJ (2019) PolyRAD: Genotype calling with uncertainty from sequencing data in polyploids and diploids. G3 (Bethesda) 9:663\u0026ndash;673. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1534/g3.118.200913\u003c/span\u003e\u003cspan address=\"10.1534/g3.118.200913\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eClark LV, Schreier AD (2017) Resolving microsatellite genotype ambiguity in populations of allopolyploid and diploidized autopolyploid organisms using negative correlations between allelic variables. Mol Ecol Resour 17:1090\u0026ndash;1103. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/1755-0998.12639\u003c/span\u003e\u003cspan address=\"10.1111/1755-0998.12639\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eColle M, Leisner CP, Wai CM, et al (2019) Haplotype-phased genome and evolution of phytonutrient pathways of tetraploid blueberry. Gigascience 8:giz012. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/gigascience/giz012\u003c/span\u003e\u003cspan address=\"10.1093/gigascience/giz012\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCruz VMV, Kilian A, Dierig DA (2013) Development of DArT marker platforms and genetic diversity assessment of the U.S. collection of the new oilseed crop lesquerella and related species. PLoS One 8:e64062. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pone.0064062\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0064062\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ede Barros Rodrigues ML, Rodrigues MP, Norton HL, et al (2025) Large-scale selection of highly informative microhaplotypes for ancestry inference and population specific informativeness. Forensic Sci Int Genet 74:103153. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.fsigen.2024.103153\u003c/span\u003e\u003cspan address=\"10.1016/j.fsigen.2024.103153\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDelomas TA, Hollenbeck CM, Matt JL, Thompson NF (2024) Microhaplotypes generate higher breeding value accuracy compared to SNPs for imputation-based breeding strategies. Aquaculture 586:740779. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.aquaculture.2024.740779\u003c/span\u003e\u003cspan address=\"10.1016/j.aquaculture.2024.740779\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDiaz-Garcia L, Garcia-Ortega LF, Gonz\u0026aacute;lez-Rodr\u0026iacute;guez M, et al (2021) Chromosome-Level Genome Assembly of the American Cranberry (Vaccinium macrocarpon Ait.) and Its Wild Relative Vaccinium microcarpum. Front Plant Sci 12:633310. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fpls.2021.633310\u003c/span\u003e\u003cspan address=\"10.3389/fpls.2021.633310\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eElshire RJ, Glaubitz JC, Sun Q, et al (2011) A robust, simple genotyping-by-sequencing (GBS) approach for high diversity species. PLoS One 6:e19379. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pone.0019379\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0019379\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEndelman JB, Kante M, Lindqvist-Kreuze H, et al (2024) Targeted genotyping-by-sequencing of potato and data analysis with R/polyBreedR. Plant Genome e20484. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/tpg2.20484\u003c/span\u003e\u003cspan address=\"10.1002/tpg2.20484\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGabriel SB, Schaffner SF, Nguyen H, et al (2002) The structure of haplotype blocks in the human genome. Science 296:2225\u0026ndash;2229. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1126/science.1069424\u003c/span\u003e\u003cspan address=\"10.1126/science.1069424\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGarrison E, Marth G (2012) Haplotype-based variant detection from short-read sequencing. arXiv [q-bio.GN]\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGerard D, Ferr\u0026atilde;o LFV (2020) Priors for genotyping polyploids. Bioinformatics 36:1795\u0026ndash;1800. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/bioinformatics/btz852\u003c/span\u003e\u003cspan address=\"10.1093/bioinformatics/btz852\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHamazaki K, Iwata H (2020) RAINBOW: Haplotype-based genome-wide association study using a novel SNP-set method. PLoS Comput Biol 16:e1007663. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pcbi.1007663\u003c/span\u003e\u003cspan address=\"10.1371/journal.pcbi.1007663\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHancock JF (2004) Blueberry plant denominated \u0026lsquo;Draper.\u0026rsquo; US Patent\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHardigan MA, Feldmann MJ, Carling J, et al (2023) A medium-density genotyping platform for cultivated strawberry using DArTag technology. Plant Genome 16:e20399. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/tpg2.20399\u003c/span\u003e\u003cspan address=\"10.1002/tpg2.20399\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHardigan MA, Feldmann MJ, Pincot DDA, et al (2021) Blueprint for phasing and assembling the genomes of heterozygous polyploids: Application to the octoploid genome of strawberry. bioRxiv 2021.11.03.467115\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJombart T (2008) adegenet: a R package for the multivariate analysis of genetic markers. Bioinformatics 24:1403\u0026ndash;1405. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/bioinformatics/btn129\u003c/span\u003e\u003cspan address=\"10.1093/bioinformatics/btn129\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKidd KK, Pakstis AJ, Speed WC, et al (2013) Microhaplotype loci are a powerful new type of forensic marker. Forensic Sci Int Genet Suppl Ser 4:e123\u0026ndash;e124. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.fsigss.2013.10.063\u003c/span\u003e\u003cspan address=\"10.1016/j.fsigss.2013.10.063\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKidd KK, Pakstis AJ, Speed WC, et al (2014) Current sequencing technology makes microhaplotypes a powerful new type of genetic marker for forensics. Forensic Sci Int Genet 12:215\u0026ndash;224. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.fsigen.2014.06.014\u003c/span\u003e\u003cspan address=\"10.1016/j.fsigen.2014.06.014\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKnaus BJ, Gr\u0026uuml;nwald NJ (2017) vcfr: a package to manipulate and visualize variant call format data in R. Mol Ecol Resour 17:44\u0026ndash;53. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/1755-0998.12549\u003c/span\u003e\u003cspan address=\"10.1111/1755-0998.12549\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLangmead B, Salzberg SL (2012) Fast gapped-read alignment with Bowtie 2. Nat Methods 9:357\u0026ndash;359. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/nmeth.1923\u003c/span\u003e\u003cspan address=\"10.1038/nmeth.1923\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLovell JT, Bentley NB, Bhattarai G, et al (2021) Four chromosome scale genomes and a pan-genome annotation to accelerate pecan tree breeding. Nat Commun 12:4125. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41467-021-24328-w\u003c/span\u003e\u003cspan address=\"10.1038/s41467-021-24328-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMammadov J, Aggarwal R, Buyyarapu R, Kumpatla S (2012) SNP markers and their impact on plant breeding. Int J Plant Genomics 2012:728398. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1155/2012/728398\u003c/span\u003e\u003cspan address=\"10.1155/2012/728398\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMartin M (2011) Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet.journal 17:10\u0026ndash;12\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMartin M, Patterson M, Garg S, et al (2016) WhatsHap: fast and accurate read-based phasing. bioRxiv\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMatias FI, Galli G, Correia Granato IS, Fritsche-Neto R (2017) Genomic prediction of autogamous and allogamous plants by SNPs and haplotypes. Crop Sci 57:2951\u0026ndash;2958. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2135/cropsci2017.01.0022\u003c/span\u003e\u003cspan address=\"10.2135/cropsci2017.01.0022\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMedina CA, Zhao D, Lin M, et al (2025) Pre-breeding in alfalfa germplasm develops highly differentiated populations, as revealed by genome-wide microhaplotype markers. Scientific Reports 15:1253\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMengist MF, Bostan H, De Paola D, et al (2023) Autopolyploid inheritance and a heterozygous reciprocal translocation shape chromosome genetic behavior in tetraploid blueberry (Vaccinium corymbosum). New Phytol 237:1024\u0026ndash;1039. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/nph.18428\u003c/span\u003e\u003cspan address=\"10.1111/nph.18428\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMr\u0026aacute;zek J, Guo X, Shah A (2007) Simple sequence repeats in prokaryotic genomes. Proc Natl Acad Sci U S A 104:8472\u0026ndash;8477. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1073/pnas.0702412104\u003c/span\u003e\u003cspan address=\"10.1073/pnas.0702412104\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOldoni F, Podini D (2019) Forensic molecular biomarkers for mixture analysis. Forensic Sci Int Genet 41:107\u0026ndash;119. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.fsigen.2019.04.003\u003c/span\u003e\u003cspan address=\"10.1016/j.fsigen.2019.04.003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePham GM, Hamilton JP, Wood JC, et al (2020) Construction of a chromosome-scale long-read reference genome assembly for potato. Gigascience 9:giaa100. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/gigascience/giaa100\u003c/span\u003e\u003cspan address=\"10.1093/gigascience/giaa100\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePreedy KF, Hackett CA (2016) A rapid marker ordering approach for high-density genetic linkage maps in experimental autotetraploid populations using multidimensional scaling. Z\u0026uuml;chter Genet Breed Res 129:2117\u0026ndash;2132. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00122-016-2761-8\u003c/span\u003e\u003cspan address=\"10.1007/s00122-016-2761-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSandercock AM, Peel MD, Taniguti CH, et al (2025) BIGapp: A user-friendly genomic tool kit identified quantitative trait loci for creeping rootedness in alfalfa (Medicago sativa L.). Plant Genome 18:e70067. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/tpg2.70067\u003c/span\u003e\u003cspan address=\"10.1002/tpg2.70067\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSemalaiyappan J, Selvanayagam S, Rathore A, et al (2022) Development of a new AgriSeq 4K mid-density SNP genotyping panel and its utility in pearl millet breeding. Front Plant Sci 13:1068883. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fpls.2022.1068883\u003c/span\u003e\u003cspan address=\"10.3389/fpls.2022.1068883\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTaniguti CH, Taniguti LM, Amadeu RR, et al (2022) Developing best practices for genotyping-by-sequencing analysis in the construction of linkage maps. Gigascience 12:. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/gigascience/giad092\u003c/span\u003e\u003cspan address=\"10.1093/gigascience/giad092\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTh\u0026eacute;r\u0026egrave;se Navarro A, Tumino G, Voorrips RE, et al (2022) Multiallelic models for QTL mapping in diverse polyploid populations. BMC Bioinformatics 23:67. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12859-022-04607-z\u003c/span\u003e\u003cspan address=\"10.1186/s12859-022-04607-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThompson NF, Sutherland BJG, Green TJ, Delomas TA (2025) A free lunch: microhaplotype discovery in an existing amplicon panel improves parentage assignment for the highly polymorphic Pacific oyster. G3 (Bethesda) 15:1\u0026ndash;11. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/g3journal/jkae280\u003c/span\u003e\u003cspan address=\"10.1093/g3journal/jkae280\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTirnaz S, Zandberg J, Thomas WJW, et al (2022) Application of crop wild relatives in modern breeding: An overview of resources, experimental and computational methodologies. Front Plant Sci 13:1008904. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fpls.2022.1008904\u003c/span\u003e\u003cspan address=\"10.3389/fpls.2022.1008904\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTurchi C, Melchionda F, Gentile F, et al (2026) Microhaplotypes in forensic genetics: From exploration to application in degraded DNA specimens. Forensic Sci Int Genet 81:103391. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.fsigen.2025.103391\u003c/span\u003e\u003cspan address=\"10.1016/j.fsigen.2025.103391\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVos P, Hogers R, Bleeker M, et al (1995) AFLP: a new technique for DNA fingerprinting. Nucleic Acids Res 23:4407\u0026ndash;4414. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/nar/23.21.4407\u003c/span\u003e\u003cspan address=\"10.1093/nar/23.21.4407\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang N, Akey JM, Zhang K, et al (2002) Distribution of recombination crossovers and the origin of haplotype blocks: the interplay of population history, recombination, and mutation. Am J Hum Genet 71:1227\u0026ndash;1234. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1086/344398\u003c/span\u003e\u003cspan address=\"10.1086/344398\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWelsh J, McClelland M (1990) Fingerprinting genomes using PCR with arbitrary primers. Nucleic Acids Res 18:7213\u0026ndash;7218. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/nar/18.24.7213\u003c/span\u003e\u003cspan address=\"10.1093/nar/18.24.7213\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWilliams JG, Kubelik AR, Livak KJ, et al (1990) DNA polymorphisms amplified by arbitrary primers are useful as genetic markers. Nucleic Acids Res 18:6531\u0026ndash;6535. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/nar/18.22.6531\u003c/span\u003e\u003cspan address=\"10.1093/nar/18.22.6531\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu R, Chen H, Li R, et al (2021) Pairwise kinship testing with microhaplotypes: Can advancements be made in kinship inference with these markers? Forensic Sci Int 325:110875. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.forsciint.2021.110875\u003c/span\u003e\u003cspan address=\"10.1016/j.forsciint.2021.110875\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu S, Lau KH, Cao Q, et al (2018) Genome sequences of two diploid wild relatives of cultivated sweetpotato reveal targets for genetic improvement. Nat Commun 9:. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41467-018-06983-8\u003c/span\u003e\u003cspan address=\"10.1038/s41467-018-06983-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang L, Koo D-H, Li Y, et al (2012) Chromosome rearrangements during domestication of cucumber as revealed by high-density genetic mapping and draft genome assembly. Plant J 71:895\u0026ndash;906. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1365-313X.2012.05017.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1365-313X.2012.05017.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang H, Mittal N, Leamy LJ, et al (2017) Back into the wild-Apply untapped genetic diversity of wild relatives for crop improvement. Evol Appl 10:5\u0026ndash;24. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/eva.12434\u003c/span\u003e\u003cspan address=\"10.1111/eva.12434\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhao D, Chinchilla-Vargas J, Sandercock AM, et al (2025) A public mid-density genotyping platform for North American Atlantic salmon (Salmo salar L.). GenResJ 6:111\u0026ndash;119. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.46265/genresj.skpz5680\u003c/span\u003e\u003cspan address=\"10.46265/genresj.skpz5680\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhao D, Mejia-Guerra KM, Mollinari M, et al (2023) A public mid-density genotyping platform for alfalfa (Medicago sativa L.). Gen Res J 4:55\u0026ndash;63. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.46265/genresj.EMOR6509\u003c/span\u003e\u003cspan address=\"10.46265/genresj.EMOR6509\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhao D, Sandercock AM, Mejia-Guerra MK, et al (2024a) A public mid-density genotyping platform for hexaploid Sweetpotato (Ipomoea batatas [L.] Lam). Genes (Basel) 15:1047. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/genes15081047\u003c/span\u003e\u003cspan address=\"10.3390/genes15081047\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhao D, Sapkota M, Glaubitz J, et al (2024b) A public mid-density genotyping platform for cultivated blueberry (Vaccinium spp.). GenResJ 5:36\u0026ndash;44. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.46265/genresj.WQZS1824\u003c/span\u003e\u003cspan address=\"10.46265/genresj.WQZS1824\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhao D, Sapkota M, Lin M, et al (2024c) Genetic diversity, population structure, and taxonomic confirmation in annual medic (Medicago spp.) collections from Crimea, Ukraine. Front Plant Sci 15:. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fpls.2024.1339298\u003c/span\u003e\u003cspan address=\"10.3389/fpls.2024.1339298\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"theoretical-and-applied-genetics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"taag","sideBox":"Learn more about [Theoretical and Applied Genetics](https://www.springer.com/journal/122)","snPcode":"122","submissionUrl":"https://submission.nature.com/new-submission/122/3","title":"Theoretical and Applied Genetics","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"microhaplotypes, crop species, HapApp, allelic diversity, multi-allelic markers","lastPublishedDoi":"10.21203/rs.3.rs-9292361/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9292361/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eMicrohaplotypes are short genomic segments that contain multiple tightly linked variants, providing multiallelic data that can enhance genetic resolution compared to traditional biallelic single nucleotide polymorphism (SNP) markers. Here, we present the creation and utilization of separate microhaplotype databases for eight crop species representing diverse genome sizes, ploidy levels, and breeding systems. We developed a standardized, species-agnostic pipeline for processing, filtering, and databasing microhaplotypes generated using the DArTag targeted genotyping platform. To enhance user accessibility, we developed a no-code, user-friendly application, HapApp, that uses an R Shiny front-end interface to allow breeders and researchers to add unique, standardized microhaplotype identities from raw DArTag reports and iteratively update the existing crop-specific database with the newly discovered microhaplotypes. Comparative analyses of these databases highlighted the advantages of microhaplotypes in capturing greater allelic diversity, resolving fine-scale population structures, and improving linkage map construction. This integrated framework provides a reproducible and scalable foundation for managing and exploiting microhaplotype data in plant breeding and genetic research, enabling robust cross-project comparisons and facilitating trait discovery in both simple and complex crop genomes, while enabling comparative genomics and cross-species functional transfer that accelerates genetic gains across all crop species.\u003c/p\u003e","manuscriptTitle":"Development and Characterization of Microhaplotype Databases for Diverse Crop Species","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-17 14:05:48","doi":"10.21203/rs.3.rs-9292361/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-04-19T01:40:24+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"216581980831234141345215110029125094736","date":"2026-04-14T21:39:49+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"250703104573958364258817145501570893914","date":"2026-04-12T17:07:05+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-09T15:06:06+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-03T12:49:14+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-04-02T18:37:11+00:00","index":"","fulltext":""},{"type":"submitted","content":"Theoretical and Applied Genetics","date":"2026-04-01T12:59:01+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"theoretical-and-applied-genetics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"taag","sideBox":"Learn more about [Theoretical and Applied Genetics](https://www.springer.com/journal/122)","snPcode":"122","submissionUrl":"https://submission.nature.com/new-submission/122/3","title":"Theoretical and Applied Genetics","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"738a9a9d-d3ed-4022-80c9-777a04212ad2","owner":[],"postedDate":"April 17th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-04-17T14:05:49+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-17 14:05:48","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9292361","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9292361","identity":"rs-9292361","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}