ASET: An end-to-end pipeline for quantification and visualization of allele specific expression

ASET: An end-to-end pipeline for quantification and visualization of allele specific expression | 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 ASET: An end-to-end pipeline for quantification and visualization of allele specific expression Weisheng Wu, Kerby Shedden, Claudius Vincenz, Chris Gates, Beverly Strassmann This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6844336/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 21 Oct, 2025 Read the published version in BMC Bioinformatics → Version 1 posted 4 You are reading this latest preprint version Abstract Motivation Allele-specific expression (ASE) analyses from RNA-Seq data provide quantitative insights into imprinting and genetic variants affecting transcription. Robust ASE analysis requires the integration of multiple computational steps, including read alignment, read counting, data visualization, and statistical testing—this complexity creates challenges around reproducibility, scalability, and ease of use. Results Here, we present ASE Toolkit (ASET), an end-to-end pipeline that streamlines SNP-level ASE data generation, visualization, and testing for parent-of-origin (PofO) effect. ASET includes a modular pipeline built with Nextflow for ASE quantification from short-read transcriptome sequencing reads, an R library for data visualization, and a Julia script for PofO testing. ASET performs comprehensive read quality control, SNP-tolerant alignment to reference genomes, read counting with allele and strand resolution, annotation with genes and exons, and estimation of contamination. In sum, ASET provides a complete and easy-to-use solution for molecular and biomedical scientists to identify and interpret patterns in ASE from RNA-Seq data. Availability ASET is available at https://github.com/weishwu/ASET . The ASE data preparation section is implemented in Nextflow with DSL2 syntax. The data visualization functionality is provided as an R library, directly available from the ASET repository or from https://github.com/weishwu/ASEplot . The PofO testing algorithm is implemented in a Julia script. ASET and ASEplot are also accessible as docker containers from Docker Hub: and https://hub.docker.com/repository/docker/weishwu/aseplot. Contact [email protected] Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 1 Introduction Allele-specific expression (ASE) refers to the differential expression between the two alleles measurable at heterozygous single nucleotide polymorphism (SNP) sites. ASE events can arise from different biological mechanisms, including genomic imprinting (Baran et al. , 2015), regulatory genetic variation and eQTLs (Lappalainen et al. , 2013; Aguet et al. , 2017) allele specific methylation or chromatin remodeling (Schmitz et al. , 2013), X chromosome inactivation (Carrel and Willard, 2005), and nonsense-mediated decay (Rivas et al. , 2015). High-throughput RNA-Seq technology has been widely used to measure ASE. Multiple approaches and algorithms have been developed for ASE quantification, focusing on reducing the alignment bias towards reference alleles because the genome reference does not contain the alternative alleles (Degner et al. , 2009). Below are several representative tools and methods that exemplify these strategies. AlleleSeq (Rozowsky et al. , 2011) and SNPsplit (Krueger and Andrews, 2016) can incorporate the alleles of the phased variants into the reference to create two haploid sets of genomes. After alignment against this personalized genome, the reads can be filtered to keep only the ones that are uniquely assigned to one of the haploid genomes. However, this approach requires complete phasing of the variants, which in most cases can only be achieved by sequencing the parental genomes. GSNAP (Wu et al. , 2016) is a SNP-tolerant aligner that treats alternative alleles as matches to the reference, rather than counting them as mismatches, thereby reducing alignment bias toward the reference allele. WASP (Van De Geijn et al. , 2015) is an alignment filtering method that swaps the alleles in SNP-containing reads, and then the reads whose mapping locations change after allele swapping can be eliminated. WASP is integrated into STAR (Dobin et al. , 2012; Asiimwe and Alexander, 2024) which is a frequently used aligner for RNA-Seq reads due to its accuracy and speed. ASEReadCounter is a tool in the widely used GATK toolkit (McKenna et al. , 2010) and is specifically designed for allele-specific RNA-Seq read counting, with many available parameters controlling read filtering and counting criteria. ASElux (Miao et al. , 2018) is an ultra-fast allele-specific read counter that first generates SNP-aware genome indices using only SNP-containing genic regions and then aligns the reads only against these regions for read counting. Allelome.PRO (Andergassen et al. , 2015) is a pipeline for identifying ASE from user-provided RNA-Seq alignments and phased SNP data. It was originally tailored for mouse reciprocal cross samples and was later expanded to diverse biological samples including human datasets. Most of the tools mentioned above have been reviewed, benchmarked, and widely adopted for ASE analyses (Castel et al. , 2015), and the STAR-WASP-ASEReadCounter workflow was used to generate SNP-level ASE data in the Genotype-Tissue Expression (GTEx) project (Lonsdale et al. , 2013; Castel et al. , 2020). Some pipelines have been developed to incorporate some of these tools for ASE quantification, such as the gtex-pipeline (Castel et al. , 2020), mRNAseq from snakePipes (Bhardwaj et al. , 2019), Allele-specific RNA-seq workflow ( https://github.com/yuviaapr/allele-specific_RNA-seq ), RNAseq-VAX ( https://github.com/arontommi/RNAseq-VAX ), and as_analysis ( https://github.com/aryarm/as_analysis ). However, most of these pipelines lack either flexibility or end-to-end analyses; notably, none of these pipelines directly include ASE data visualization or PofO testing. Here we present ASE Toolkit (ASET) for SNP-level ASE quantification. ASET leverages the Nextflow workflow manager (DI Tommaso et al. , 2017) that accepts raw short-read RNA-Seq data and produces SNP-level ASE count data with gene annotation and contamination estimates. ASET integrates multiple alignment options that were designed specifically for ASE analysis, enabling simple usage and customization. It also includes data visualization and PofO testing. ASET provides an easy-to-use suite that streamlines ASE data preparation and visualization, providing the foundation for further interpretation and analysis. 2 Methods The main modules of ASET are implemented using Nextflow, a modern workflow management system that facilitates scalable and reproducible computational pipelines. Nextflow is widely used in the bioinformatics community due to its comprehensive documentation, container support, and support via GitHub and Slack. Leveraging the latest DSL2 syntax, ASET adopts a modular design in which individual analysis steps are implemented as modules. This modularity allows for clean organization, simplified maintenance, and the seamless integration of sub-workflows for alternative analysis paths. ASET also supports containerization through Docker (MerkelDirk, 2014) and Singularity (Kurtzer, Sochat and Bauer, 2017), enabling portable execution across local machines, HPC clusters, and cloud environments. Reproducibility is further enhanced by version-controlled releases, locked software dependencies via containers, and automatic reporting of tool versions and parameters. Both analysis parameters and computational parameters (e.g. CPU and memory usage) can be specified via a configuration file. The data visualization functionality is bundled in an R (R Core Team, 2013) library “ASEplot”. R is a very common platform used for data analysis and visualization. The PofO testing algorithm is provided as a Julia (Bezanson et al. , 2017) script. Julia is a high-performance programming language designed for statistical modeling. An overview of the ASET pipeline is shown in Figure 1 and Supplementary Figure 1. It requires two input files: a sample sheet containing the paths to the read files and SNP VCFs, and a parameter configuration file containing adjustable parameter setting for each tool and the paths to reference files. ASET can be run in two modes: from_fastq or from_bam . In the from_fastq mode, it takes the raw FASTQ reads as input and implements read QC, trimming, and alignment. In the from_bam mode, it takes the provided BAM files and goes directly to alignment filtering and deduplication. Users also need to provide a VCF containing the SNPs for each sample and this VCF will be used for SNP-aware alignment and SNP-level ASE read counting. After read alignment and counting, the data will be concatenated from all the samples to produce an ASE data table, followed by contamination estimation and annotation for genes and exons. The output can be loaded directly into ASEplot for plot generation and data filtering. ASET does not require phasing of the SNPs, but when phased SNPs are available, phasing information can be incorporated, and the phased subset can be analyzed using po_test.jl for PofO testing. The comparison of capabilities among ASET and other available ASE pipelines is summarized in Table 1. The advantages of ASET include: (1) incorporation of four commonly used alignment approaches tailored for ASE analysis, (2) generation of ASE count data in a strand-specific manner, (3) estimation of contamination levels, (4) data visualization, and (5) PofO testing. Table 1. Comparison between ASET versus other available ASE pipelines. (NA means “not directly available”). Feature ASET gtex-pipeline snakePipes Allele-specific RNA-seq workflow RNAseq-VAX as_analysis System Nextflow Cromwell Snakemake Nextflow Nextflow Snakemake Aligner GSNAP or STAR or HISAT2 with N-masked ref STAR with N-masked ref NA STAR+WASP STAR+WASP or STAR+WASP STAR with N-masked ref or ASElux Strand-specific Supported NA Supported NA NA NA Read counting SNP-level # SNP & Haplotype-level # f Gene-level Gene-level SNP-level # SNP-level Contamination estimate Supported NA NA NA NA NA Visualization plots Tailored for ASE NA Tailored for QC and differential expression NA NA NA PofO testing Supported NA* NA NA NA NA # these pipelines use GATK ASEReadCounter for SNP-level allelic read counting from alignments. f The gtex-pipeline has a module for haplotype-specific expression when phased genotypes are available. * The gtex-pipeline has a module for eQTL testing. Individual pipeline steps are explained as follows. 2.1 Read QC ASE data accuracy and robustness depend heavily on the quality of sequencing data, especially the effective coverage of the assayed SNPs, as shown in our previous publication (Wu et al., 2021). ASET uses FastQC (Andrews, 2010) and CollectRnaSeqMetrics from GATK (McKenna et al., 2010) to assess RNA-Seq read quality, and uses Trimmomatic (Bolger, Lohse and Usadel, 2014) to remove adapter contamination and low-quality ends. QC metrics are summarized in both a MultiQC (Ewels et al., 2016) report and a tabular spreadsheet. 2.2 Read alignment ASET currently contains four sub-workflow choices for read alignment. The mapper parameter specified in the configuration file selects one of these alignment approaches: (1) STAR+WASP where the alignment is performed using STAR with the --waspOutputMode parameter to enable WASP filtering; (2) STAR+NMASK where the genome is first N-masked at the SNP sites and then used for STAR alignment; (3) GSNAP where reads are aligned using GSNAP in the SNP-tolerant mode; and (4) ASElux where reads are aligned and counted using ASElux. Note that the provided genome FASTA and GTF files will be indexed by the chosen aligner for splice-aware alignment. 2.3 Alignment filtering, deduplication, and strand split Alignments are filtered based on adjustable flags and mapping quality cutoffs. STAR+WASP-based alignments can additionally exclude alignments flagged as problematic (based on vW tag). Reads are then deduplicated using GATK MarkDuplicates. Deduplicated reads are split into two alignment files based on strand. A strandedness parameter needs to be provided to indicate whether read 1 or read 2 corresponds to the original RNA strand. (ASElux-based alignments skip this step as ASElux integrates both read alignment and counting without emitting the alignment files for manipulation.) 2.4 ASE read counting GATK ASEReadCounter is applied on each alignment file to compute allele-specific read counts on all provided heterozygous and homozygous SNPs, and optionally the genotyped reference sites. Output files on different strands from all samples are concatenated into a single file for each type of site. Base quality cutoffs, mapping quality cutoffs, and the overlap handling scheme are configurable. (As above, ASElux-based alignments skip this step.) 2.5 Contamination estimation The average non-alternative-allele frequency on homozygous SNP sites, and the average non-reference-allele frequency on reference sites (if available), are calculated to serve as an estimate of cross-contamination (or mislabeling) for each sample. For placental samples where maternal contamination is a concern, the average non-reference-allele frequency at the reference sites where the mother has a non-reference genotype is also calculated for each gene individually, with the assumption that the non-reference allele counts arise from contamination by maternal tissue. (ASElux-based alignments skip this step since ASElux only counts reads at exonic heterozygous SNPs.) 2.6 Annotation Based on the provided GTF, the exons from the same gene are merged into a union exon set and then used to annotate a table of SNPs. Each SNP (row) details exon coordinates, gene IDs, symbols, and gene types. When phasing data is provided, paternal and maternal alleles will be indicated, and the paternal allele frequency will be calculated for each SNP that has data. 2.7 ASET outputs The ASET data preparation steps emit an ASE data table that contains count data, contamination estimates, and gene/exon annotation, from all samples; and an R data file that can be used for visualization and downstream PofO testing, using ASEplot and po_test.jl, respectively. Additionally, these steps also produce trimmed FASTQ files, alignment BAM files, MultiQC reports, and a QC tabular spreadsheet. 2.8 Determination of parent-of-origin scores For a given gene with N total read counts and m distinct SNPs, let Y ijk denote the read count for allele k of SNP j for subject i . The alleles are coded k = 0, 1 for the reference and alternative alleles, respectively. Define X ijk = 1/2 when k = 0 and -1/2 when k = 1; and define Z ijk = 1/2 and -1/2 for paternal and maternal allele read counts, respectively. Next, construct an N × m matrix of indicator variables U , where column l of U is defined as U l ijk = 1 if j = l and 0 otherwise. Next, let V denote an N × q matrix consisting of the left singular vectors of U whose singular values are at least 1% of the maximum singular value of U . We fit a cluster-robust quasi-Poisson regression model for each gene in which the indices i , j , k index the N observations, and the explanatory variables are the main effect of parent of origin ( Z ijk ), the main effect of ref/alt status ( X ijk ), main effects for SNP indicators ( V ), and all pairwise interactions between SNP indicators ( V ) and ref/alt status ( X ijk ). Including the X and V main effects and their pairwise interactions allows us to account for genetic ASE, while clustering on subjects ( i ) allows us to account for correlations among read counts within the same individual (e.g. due to linkage disequilibrium). The full model is shown below: We refer to the estimated coefficient for Z as the PofO score and denote it po , with its z-score denoted po_z . Positive and negative po correspond, respectively, to paternally and maternally biased expressions, while 0 denotes a balance. We view | po | > 3 as denoting strong parentally determined ASE, implying at least a 20-fold difference between the two alleles, and | po_z | > 3 as denoting statistical significance. 3 Results We applied ASET on the sequencing data from a set of 244 targeted RNA-Seq samples we previously published (Wu et al. , 2021), using the STAR+WASP alignment approach. This produced a data table with 346,503 exonic SNP × sample × strand data points, observed in 783 genes. Using the ASEplot R library, we can visualize the SNP locations in specific genes (Figure 2 and Supplementary Figure 2), sample-level and gene-level contamination (Figure 3), and exon- and gene-level ASE distribution across different samples, exons, or genes (Figure 4, Figure 5, and Supplementary Figure 3). After data filtering including requiring at least 10 read counts at SNPs and lower than 5% contamination (when measurable), 264,046 data points were retained. The phased subset with 125,772 data points was analyzed using po_test.jl for PofO testing. The results showed that out of 392 genes that were testable, 153 had a strong PofO effect with | po_z | > 3, with 92 biased to paternal expression and 61 biased to the maternal side. Among these genes, 33 had a large difference between the alleles with | po | > 3. 4 Discussion ASET provides an integrated and reproducible framework for the generation and visualization of ASE data, addressing a critical need for streamlined ASE analysis in genomic studies. It combines a robust Nextflow-based workflow for data preprocessing with a dedicated R package for visualization and a statistical algorithm for PofO testing. Compared to other available ASE workflows, ASET provides a more complete solution by including multiple alignment approaches tailored for ASE analysis, support for strand-specific read counting, contamination estimation, data visualization, and PofO testing. ASET employs containerization through Docker and Singularity to boost convenience and reproducibility across different environments. The pipeline's modular structure provides flexibility for further expansion by the addition of more modules. For example, another sub-workflow can be added to enable personalized diploid genome construction and alignment when a complete phased SNP set is available. The current annotation of the SNPs by using the merged exons lacks the ability to interrogate isoform-level ASE. With diploid genome construction and sufficient density of heterozygous SNPs (e.g. from inbred mouse strains), there are approaches to resolve ASE quantification on the isoform-level (Turro et al. , 2011; Perez et al. , 2015). However, the best solution for isoform ASE analysis may lie in full-length transcriptome sequencing using long-read sequencing technologies (Glinos et al., 2022; Tang et al., 2024). The current support provided for downstream data analysis focuses on basic visualization and PofO testing. We realize that there is a variety of methods for downstream analyses, such as eQTL and prediction of cis -acting ncRNA-targets (Hasenbein et al. , 2025). In addition, haplotype-specific expression can be enabled using phASER, especially when long-read RNA-Seq data are available (Castel et al. , 2016). We will be working on adding more functionality to ASET to incorporate diploid alignment, isoform-level ASE measurement, and further statistical analysis especially when phenotype data are available. Overall, compared to the existing alternative pipelines, ASET provides a more comprehensive workflow that bridges the gap between raw data and SNP-level ASE measurement and interpretation, and is particularly valuable for studies such as genomic imprinting, eQTLs, X chromosome inactivation and nonsense-mediated decay, where the preparation of robust ASE data is required. Abbreviations ASE Allele specific expression SNP Single nucleotide polymorphism PofO Parent of-origin Declarations Ethics approval and consent to participate Informed consent or assent was obtained from participants depending on whether they were adults or children. Institutional Review Boards (IRB) approval was obtained from the University of Michigan IRBMED (HUM00043670) and from La Faculté de Médecine de Pharmacie et d'Odontostomatologie (FMPOS) de Bamako in Mali (No2016/68/CD/FMPOS). Consent for publication Not applicable. Availability of data and materials ASET is available at https://github.com/weishwu/ASET . The ASE data preparation section is implemented in Nextflow with DSL2 syntax. The data visualization functionality is provided as an R library, directly available from the ASET repository or from https://github.com/weishwu/ASEplot . The PofO testing algorithm is implemented in a Julia script. ASET and ASEplot are also accessible as docker containers from Docker Hub: and https://hub.docker.com/repository/docker/weishwu/aseplot. The RNA-Seq FASTQ files and the genotype data used to test the pipeline were published in our previous paper (Wu et al. , 2021), and deposited in dbGaP as phs001782.v2. Competing interests The authors declare that they have no competing interests. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results. Funding This research was supported by the Eunice Kennedy Shriver National Institute of Child Health & Human Development (NICHD) of the National Institutes of Health (NIH) (R01HD104676, R01HD088521 and R21HD077465 to B.I.S.); and the John Templeton Foundation (JTF) (52269 to B.I.S.). The content of this study is solely the responsibility of the authors and doesn't necessarily reflect the official views of the JTF, the NICHD, or the NIH. Authors' contributions W.W. developed the Nextflow pipeline and the ASEplot R library, wrote the main manuscript and prepared all figures and tables. K.S. developed the PofO testing model and wrote the section "Determination of parent-of-origin scores". C.V. contributed ideas to some of the pipeline modules and functions. B.S. provided funding and supervised the project. C.G. contributed ideas to pipeline code and manuscript structure. All authors reviewed the manuscript. Acknowledgements The authors acknowledge support from the BRCF Bioinformatics Core at the University of Michigan. References Aguet F et al. (2017) ‘Genetic effects on gene expression across human tissues’, Nature , 550(7675), pp. 204–213. Available at: https://doi.org/10.1038/NATURE24277 . Andergassen D et al. (2015) ‘Allelome.PRO, a pipeline to define allele-specific genomic features from high-throughput sequencing data’, Nucleic Acids Research , 43(21), p. e146. Available at: https://doi.org/10.1093/NAR/GKV727 Asiimwe R, Alexander D. (2024) ‘STAR + WASP reduces reference bias in the allele-specific mapping of RNA-seq reads’, bioRxiv: the preprint server for biology [Preprint]. Available at: https://doi.org/10.1101/2024.01.21.576391 Baran Y et al. (2015) ‘The landscape of genomic imprinting across diverse adult human tissues’, Genome Research , 25(7), pp. 927–936. Available at: https://doi.org/10.1101/GR.192278.115 . Bezanson J et al. (2017) ‘Julia: A fresh approach to numerical computing’, SIAM Review , 59(1), pp. 65–98. Available at: https://doi.org/10.1137/141000671;CTYPE:STRING:JOURNAL Bhardwaj V et al. (2019) ‘SnakePipes: Facilitating flexible, scalable and integrative epigenomic analysis’, Bioinformatics , 35(22), pp. 4757–4759. Available at: https://doi.org/10.1093/BIOINFORMATICS/BTZ436 . Carrel L, Willard HF. (2005) ‘X-inactivation profile reveals extensive variability in X-linked gene expression in females’, Nature , 434(7031), pp. 400–404. Available at: https://doi.org/10.1038/NATURE03479 . Castel SE et al. (2015) ‘Tools and best practices for data processing in allelic expression analysis’, Genome Biology , 16(1). Available at: https://doi.org/10.1186/S13059-015-0762-6 . Castel SE et al. (2016) ‘Rare variant phasing and haplotypic expression from RNA sequencing with phASER’, Nature Communications , 7(1), pp. 1–6. Available at: https://doi.org/10.1038/NCOMMS12817 ;SUBJMETA=114,208,212,631,728,794;KWRD=GENOMICS,HAPLOTYPES,SOFTWARE. Castel SE et al. (2020) ‘A vast resource of allelic expression data spanning human tissues’, Genome Biology , 21(1), pp. 1–12. Available at: https://doi.org/10.1186/S13059-020-02122-Z/FIGURES/2 Degner JF et al. (2009) ‘Effect of read-mapping biases on detecting allele-specific expression from RNA-sequencing data’, Bioinformatics , 25(24), pp. 3207–3212. Available at: https://doi.org/10.1093/BIOINFORMATICS/BTP579 . Dobin A et al. (2012) ‘STAR: ultrafast universal RNA-seq aligner’, Bioinformatics , 29(1), p. 15. Available at: https://doi.org/10.1093/BIOINFORMATICS/BTS635 Van De Geijn B et al. (2015) ‘WASP: Allele-specific software for robust molecular quantitative trait locus discovery’, Nature Methods , 12(11), pp. 1061–1063. Available at: https://doi.org/10.1038/NMETH.3582 . Glinos DA et al. (2022) ‘Transcriptome variation in human tissues revealed by long-read sequencing’, Nature 2022 608:7922 , 608(7922), pp. 353–359. Available at: https://doi.org/10.1038/s41586-022-05035-y Hasenbein TP et al. (2025) ‘Allele-specific genomics decodes gene targets and mechanisms of the non-coding genome’, bioRxiv , p. 2025.03.03.641135. Available at: https://doi.org/10.1101/2025.03.03.641135 Krueger F, Andrews SR. (2016) ‘SNPsplit: Allele-specific splitting of alignments between genomes with known SNP genotypes’, F1000Research , 5. Available at: https://doi.org/10.12688/F1000RESEARCH.9037.2/DOI Kurtzer GM, Sochat V, Bauer MW. (2017) ‘Singularity: Scientific containers for mobility of compute’, PLoS ONE , 12(5). Available at: https://doi.org/10.1371/JOURNAL.PONE.0177459 . Lappalainen T et al. (2013) ‘Transcriptome and genome sequencing uncovers functional variation in humans’, Nature , 501(7468), pp. 506–511. Available at: https://doi.org/10.1038/NATURE12531 . Lonsdale J et al. (2013) ‘The Genotype-Tissue Expression (GTEx) project’, Nature Genetics , 45(6), pp. 580–585. Available at: https://doi.org/10.1038/NG.2653 ;SUBJMETA=114,129,208,631;KWRD=DATABASES,GENETICS. McKenna A et al. (2010) ‘The genome analysis toolkit: A MapReduce framework for analyzing next-generation DNA sequencing data’, Genome Research , 20(9), pp. 1297–1303. Available at: https://doi.org/10.1101/GR.107524.110 . MerkelDirk. (2014) ‘Docker: lightweight linux containers for consistent development and deployment’, Linux Journal [Preprint]. Available at: https://doi.org/10.5555/2600239.2600241 Miao Z et al. (2018) ‘ASElux: An ultra-fast and accurate allelic reads counter’, Bioinformatics , 34(8), pp. 1313–1320. Available at: https://doi.org/10.1093/BIOINFORMATICS/BTX762 . Perez JD et al. (2015) ‘Quantitative and functional interrogation of parent-of-origin allelic expression biases in the brain’, eLife , 4(JULY 2015), p. 41. Available at: https://doi.org/10.7554/ELIFE.07860 . R Core Team. (2013) R: a language and environment for statistical computing , R Foundation for Statistical Computing, Vienna, Austria . Available at: https://www.R-project.org/ (Accessed: 14 May 2025). Rivas MA et al. (2015) ‘Effect of predicted protein-truncating genetic variants on the human transcriptome’, Science , 348(6235), pp. 666–669. Available at: https://doi.org/10.1126/SCIENCE.1261877 . Rozowsky J et al. (2011) ‘AlleleSeq: Analysis of allele-specific expression and binding in a network framework’, Molecular Systems Biology , 7. Available at: https://doi.org/10.1038/MSB.2011.54 . Schmitz RJ et al. (2013) ‘Patterns of population epigenomic diversity’, Nature , 495(7440), pp. 193–198. Available at: https://doi.org/10.1038/NATURE11968 . Tang AD et al. (2024) ‘Detecting haplotype-specific transcript variation in long reads with FLAIR2’, Genome Biology , 25(1), pp. 1–21. Available at: https://doi.org/10.1186/S13059-024-03301-Y/FIGURES/4 Tommaso DI et al. P. (2017) ‘Nextflow enables reproducible computational workflows’, Nature Biotechnology , 35(4), pp. 316–319. Available at: https://doi.org/10.1038/NBT.3820 . Turro E et al. (2011) ‘Haplotype and isoform specific expression estimation using multi-mapping RNA-seq reads’, Genome Biology , 12(2), pp. 1–15. Available at: https://doi.org/10.1186/GB-2011-12-2-R13/TABLES/3 Wu TD et al. (2016) ‘GMAP and GSNAP for genomic sequence alignment: Enhancements to speed, accuracy, and functionality’, Methods in Molecular Biology , 1418, pp. 283–334. Available at: https://doi.org/10.1007/978-1-4939-3578-9_15 . Wu W et al. (2021) ‘Targeted RNA-seq improves efficiency, resolution, and accuracy of allele specific expression for human term placentas’, G3: Genes, Genomes, Genetics , 11(8). Available at: https://doi.org/10.1093/G3JOURNAL/JKAB176 . Additional Declarations No competing interests reported. Supplementary Files ASETmanuscriptsupplement.docx Supplementary information Supplementary materials are in the document “Supplementary_materials.docx”. Cite Share Download PDF Status: Published Journal Publication published 21 Oct, 2025 Read the published version in BMC Bioinformatics → Version 1 posted Editorial decision: Revision requested 17 Jun, 2025 Editor assigned by journal 09 Jun, 2025 Submission checks completed at journal 09 Jun, 2025 First submitted to journal 07 Jun, 2025 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-6844336","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":468863164,"identity":"4cfb0b24-0b17-4c7b-a8e1-e7ec339c9437","order_by":0,"name":"Weisheng Wu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAuUlEQVRIiWNgGAWjYFACxgZmICkHYzA2EKvFmBQtDAwglYkNUAZhLQbHm5s/F7YdTt9w7XDj5wIGG9kNBwhpOXOwTXpm2+HcDbcTm6VnMKQZE9RidiOxjZl3G1hLGzMPw+FEwlruP2z+DNSSbgDR8p8ILTcYG6SBWhKgWg4Q1mJ/JrFNmvdfuuFMkF94DJKNZxLSItl+/PFnnjPW8ny30x9+5qmwk+0jpAUNGJCmfBSMglEwCkYBDgAA5npHNRnb494AAAAASUVORK5CYII=","orcid":"","institution":"University of Michigan–Ann Arbor","correspondingAuthor":true,"prefix":"","firstName":"Weisheng","middleName":"","lastName":"Wu","suffix":""},{"id":468863166,"identity":"77b5c13c-f2d3-4a40-b32e-2e39c4f8ed90","order_by":1,"name":"Kerby Shedden","email":"","orcid":"","institution":"University of Michigan–Ann Arbor","correspondingAuthor":false,"prefix":"","firstName":"Kerby","middleName":"","lastName":"Shedden","suffix":""},{"id":468863167,"identity":"1139a396-e0be-445b-834d-b2f8dbc23475","order_by":2,"name":"Claudius Vincenz","email":"","orcid":"","institution":"University of Michigan–Ann Arbor","correspondingAuthor":false,"prefix":"","firstName":"Claudius","middleName":"","lastName":"Vincenz","suffix":""},{"id":468863170,"identity":"9b828ee4-46a9-466b-a1a9-c9cfaf812f87","order_by":3,"name":"Chris Gates","email":"","orcid":"","institution":"University of Michigan–Ann Arbor","correspondingAuthor":false,"prefix":"","firstName":"Chris","middleName":"","lastName":"Gates","suffix":""},{"id":468863172,"identity":"5195cdb2-c235-47d9-a6a9-78693fa3429d","order_by":4,"name":"Beverly Strassmann","email":"","orcid":"","institution":"University of Michigan–Ann Arbor","correspondingAuthor":false,"prefix":"","firstName":"Beverly","middleName":"","lastName":"Strassmann","suffix":""}],"badges":[],"createdAt":"2025-06-07 18:38:15","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6844336/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6844336/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s12859-025-06282-2","type":"published","date":"2025-10-21T16:17:13+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":84522382,"identity":"22091a74-48c4-428d-940d-0c4f2f9fb994","added_by":"auto","created_at":"2025-06-13 04:07:24","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":256172,"visible":true,"origin":"","legend":"\u003cp\u003eOverview of ASET.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-6844336/v1/2941536c9e5c62266a00f8f6.png"},{"id":84522287,"identity":"77b51423-712f-4bd9-b3a8-fb49c9abd67c","added_by":"auto","created_at":"2025-06-13 03:59:24","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":74976,"visible":true,"origin":"","legend":"\u003cp\u003eSNP locations in the RHOBTB3 gene locus, with isoforms collapsed into a single model for each gene.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-6844336/v1/510e74570e2e92eb5dbf2e36.png"},{"id":84522292,"identity":"8d5a4392-c162-4784-b9ea-fe56d597a8be","added_by":"auto","created_at":"2025-06-13 03:59:24","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":381741,"visible":true,"origin":"","legend":"\u003cp\u003eContamination estimated from opposite allele frequencies on homozygous sites, averaged per sample (A) or per gene (B) (only showing a subset).\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-6844336/v1/75742d2062c82746d9e3446a.png"},{"id":84522290,"identity":"61923774-9f36-4ff7-80a0-487dee55aef0","added_by":"auto","created_at":"2025-06-13 03:59:24","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":27641,"visible":true,"origin":"","legend":"\u003cp\u003eDistribution of gene-level paternal allele frequency, shown as (A) a histogram for one gene with the sample of interest marked; or (B) ridges for multiple genes, with color indicating a tendency for paternal (blue) or maternal (pink) specific expression. Gene-level paternal allele frequency was calculated by summing up paternal and total count data from the exonic SNPs and then taking the ratio.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-6844336/v1/b8ac3f5c23993daf688c663f.png"},{"id":84522296,"identity":"bde9330d-85b0-41cd-9d64-d2fa243847b7","added_by":"auto","created_at":"2025-06-13 03:59:24","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":151602,"visible":true,"origin":"","legend":"\u003cp\u003eDistribution of SNP-level paternal allele frequency across different samples in a gene, shown as a scatter plot where vertical lines represent exon boundaries after merging for each gene. When a sample ID is specified, it is marked as a red triangle where all other samples are shown as gray round dots. The SNP count and the median allele frequency for this sample, plus the gene information, are shown in the title.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-6844336/v1/0ee7afd698a4432c54ef7eac.png"},{"id":94490567,"identity":"740b2192-e6df-4a1f-9e9b-8ac7b0bd140f","added_by":"auto","created_at":"2025-10-27 17:12:14","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1561108,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6844336/v1/dcbfcef4-64e3-481c-8025-dd547c4f130e.pdf"},{"id":84523169,"identity":"6de3d441-cf17-4020-a895-7a2d092fd4b7","added_by":"auto","created_at":"2025-06-13 04:15:25","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":776547,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSupplementary information\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSupplementary materials are in the document “Supplementary_materials.docx”.\u003c/p\u003e","description":"","filename":"ASETmanuscriptsupplement.docx","url":"https://assets-eu.researchsquare.com/files/rs-6844336/v1/97b11dec036ae9276c495eea.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"ASET: An end-to-end pipeline for quantification and visualization of allele specific expression","fulltext":[{"header":"1 Introduction","content":"\u003cp\u003eAllele-specific expression (ASE) refers to the differential expression between the two alleles measurable at heterozygous single nucleotide polymorphism (SNP) sites. ASE events can arise from different biological mechanisms, including genomic imprinting (Baran \u003cem\u003eet al.\u003c/em\u003e, 2015), regulatory genetic variation and eQTLs (Lappalainen \u003cem\u003eet al.\u003c/em\u003e, 2013; Aguet \u003cem\u003eet al.\u003c/em\u003e, 2017) allele specific methylation or chromatin remodeling (Schmitz \u003cem\u003eet al.\u003c/em\u003e, 2013), X chromosome inactivation (Carrel and Willard, 2005), and nonsense-mediated decay (Rivas \u003cem\u003eet al.\u003c/em\u003e, 2015). High-throughput RNA-Seq technology has been widely used to measure ASE. Multiple approaches and algorithms have been developed for ASE quantification, focusing on reducing the alignment bias towards reference alleles because the genome reference does not contain the alternative alleles (Degner \u003cem\u003eet al.\u003c/em\u003e, 2009). Below are several representative tools and methods that exemplify these strategies. AlleleSeq (Rozowsky \u003cem\u003eet al.\u003c/em\u003e, 2011) and SNPsplit (Krueger and Andrews, 2016) can incorporate the alleles of the phased variants into the reference to create two haploid sets of genomes. After alignment against this personalized genome, the reads can be filtered to keep only the ones that are uniquely assigned to one of the haploid genomes. However, this approach requires complete phasing of the variants, which in most cases can only be achieved by sequencing the parental genomes. GSNAP (Wu \u003cem\u003eet al.\u003c/em\u003e, 2016) is a SNP-tolerant aligner that treats alternative alleles as matches to the reference, rather than counting them as mismatches, thereby reducing alignment bias toward the reference allele. WASP (Van De Geijn \u003cem\u003eet al.\u003c/em\u003e, 2015) is an alignment filtering method that swaps the alleles in SNP-containing reads, and then the reads whose mapping locations change after allele swapping can be eliminated. WASP is integrated into STAR (Dobin \u003cem\u003eet al.\u003c/em\u003e, 2012; Asiimwe and Alexander, 2024) which is a frequently used aligner for RNA-Seq reads due to its accuracy and speed. ASEReadCounter is a tool in the widely used GATK toolkit (McKenna \u003cem\u003eet al.\u003c/em\u003e, 2010) and is specifically designed for allele-specific RNA-Seq read counting, with many available parameters controlling read filtering and counting criteria. ASElux (Miao \u003cem\u003eet al.\u003c/em\u003e, 2018) is an ultra-fast allele-specific read counter that first generates SNP-aware genome indices using only SNP-containing genic regions and then aligns the reads only against these regions for read counting. Allelome.PRO (Andergassen \u003cem\u003eet al.\u003c/em\u003e, 2015) is a pipeline for identifying ASE from user-provided RNA-Seq alignments and phased SNP data. It was originally tailored for mouse reciprocal cross samples and was later expanded to diverse biological samples including human datasets. Most of the tools mentioned above have been reviewed, benchmarked, and widely adopted for ASE analyses (Castel \u003cem\u003eet al.\u003c/em\u003e, 2015), and the STAR-WASP-ASEReadCounter workflow was used to generate SNP-level ASE data in the Genotype-Tissue Expression (GTEx) project (Lonsdale \u003cem\u003eet al.\u003c/em\u003e, 2013; Castel \u003cem\u003eet al.\u003c/em\u003e, 2020).\u003c/p\u003e\n\u003cp\u003eSome pipelines have been developed to incorporate some of these tools for ASE quantification, such as the gtex-pipeline (Castel \u003cem\u003eet al.\u003c/em\u003e, 2020), mRNAseq from snakePipes (Bhardwaj \u003cem\u003eet al.\u003c/em\u003e, 2019), Allele-specific RNA-seq workflow (\u003ca href=\"https://github.com/yuviaapr/allele-specific_RNA-seq\"\u003ehttps://github.com/yuviaapr/allele-specific_RNA-seq\u003c/a\u003e), RNAseq-VAX (\u003ca href=\"https://github.com/arontommi/RNAseq-VAX\"\u003ehttps://github.com/arontommi/RNAseq-VAX\u003c/a\u003e), and as_analysis (\u003ca href=\"https://github.com/aryarm/as_analysis\"\u003ehttps://github.com/aryarm/as_analysis\u003c/a\u003e). However, most of these pipelines lack either flexibility or end-to-end analyses; notably, none of these pipelines directly include ASE data visualization or PofO testing.\u003c/p\u003e\n\u003cp\u003eHere we present ASE Toolkit (ASET) for SNP-level ASE quantification. ASET leverages the Nextflow workflow manager (DI Tommaso \u003cem\u003eet al.\u003c/em\u003e, 2017) that accepts raw short-read RNA-Seq data and produces SNP-level ASE count data with gene annotation and contamination estimates. ASET integrates multiple alignment options that were designed specifically for ASE analysis, enabling simple usage and customization. It also includes data visualization and PofO testing. ASET provides an easy-to-use suite that streamlines ASE data preparation and visualization, providing the foundation for further interpretation and analysis.\u003c/p\u003e"},{"header":"2 Methods","content":"\u003cp\u003eThe main modules of ASET are implemented using Nextflow, a modern workflow management system that facilitates scalable and reproducible computational pipelines. Nextflow is widely used in the bioinformatics community due to its comprehensive documentation, container support, and support via GitHub and Slack. Leveraging the latest DSL2 syntax, ASET adopts a modular design in which individual analysis steps are implemented as modules. This modularity allows for clean organization, simplified maintenance, and the seamless integration of sub-workflows for alternative analysis paths. ASET also supports containerization through Docker (MerkelDirk, 2014) and Singularity (Kurtzer, Sochat and Bauer, 2017), enabling portable execution across local machines, HPC clusters, and cloud environments. Reproducibility is further enhanced by version-controlled releases, locked software dependencies via containers, and automatic reporting of tool versions and parameters. Both analysis parameters and computational parameters (e.g. CPU and memory usage) can be specified via a configuration file.\u003c/p\u003e\n\u003cp\u003eThe data visualization functionality is bundled in an R (R Core Team, 2013) library \u0026ldquo;ASEplot\u0026rdquo;. R is a very common platform used for data analysis and visualization. The PofO testing algorithm is provided as a Julia (Bezanson \u003cem\u003eet al.\u003c/em\u003e, 2017) script. Julia is a high-performance programming language designed for statistical modeling.\u003c/p\u003e\n\u003cp\u003eAn overview of the ASET pipeline is shown in Figure 1 and Supplementary Figure 1. It requires two input files: a sample sheet containing the paths to the read files and SNP VCFs, and a parameter configuration file containing adjustable parameter setting for each tool and the paths to reference files. ASET can be run in two modes: \u003cem\u003efrom_fastq\u003c/em\u003e or \u003cem\u003efrom_bam\u003c/em\u003e. In the \u003cem\u003efrom_fastq\u003c/em\u003e mode, it takes the raw FASTQ reads as input and implements read QC, trimming, and alignment. In the \u003cem\u003efrom_bam\u003c/em\u003e mode, it takes the provided BAM files and goes directly to alignment filtering and deduplication. Users also need to provide a VCF containing the SNPs for each sample and this VCF will be used for SNP-aware alignment and SNP-level ASE read counting. After read alignment and counting, the data will be concatenated from all the samples to produce an ASE data table, followed by contamination estimation and annotation for genes and exons. The output can be loaded directly into ASEplot for plot generation and data filtering. ASET does not require phasing of the SNPs, but when phased SNPs are available, phasing information can be incorporated, and the phased subset can be analyzed using po_test.jl for PofO testing.\u003c/p\u003e\n\u003cp\u003eThe comparison of capabilities among ASET and other available ASE pipelines is summarized in Table 1. The advantages of ASET include: (1) incorporation of four commonly used alignment approaches tailored for ASE analysis, (2) generation of ASE count data in a strand-specific manner, (3) estimation of contamination levels, (4) data visualization, and (5) PofO testing. \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTable 1. Comparison between ASET versus other available ASE pipelines. (NA means \u0026ldquo;not directly available\u0026rdquo;).\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"607\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eFeature\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eASET\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003egtex-pipeline\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003esnakePipes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eAllele-specific RNA-seq workflow\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eRNAseq-VAX\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eas_analysis\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eSystem\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNextflow\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eCromwell\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eSnakemake\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNextflow\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNextflow\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eSnakemake\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"3\"\u003e\n \u003cp\u003e\u003cstrong\u003eAligner\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eGSNAP or\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd rowspan=\"3\"\u003e\n \u003cp\u003eSTAR or HISAT2 with N-masked ref\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"3\"\u003e\n \u003cp\u003eSTAR with N-masked ref\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"3\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"3\"\u003e\n \u003cp\u003eSTAR+WASP\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eSTAR+WASP or\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eSTAR+WASP\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eSTAR with N-masked ref or ASElux\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eStrand-specific\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eSupported\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eSupported\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eRead counting\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eSNP-level\u003csup\u003e#\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eSNP \u0026amp; Haplotype-level\u003csup\u003e#\u0026nbsp;\u003c/sup\u003e\u003csup\u003ef\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eGene-level\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eGene-level\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eSNP-level\u003csup\u003e#\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eSNP-level\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eContamination estimate\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eSupported\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eVisualization plots\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eTailored for ASE\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eTailored for QC and differential expression\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003ePofO testing\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eSupported\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNA*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e# these pipelines use GATK ASEReadCounter for SNP-level allelic read counting from alignments.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003ef\u0026nbsp;The gtex-pipeline has a module for haplotype-specific expression when phased genotypes are available.\u003c/p\u003e\n\u003cp\u003e* The gtex-pipeline has a module for eQTL testing.\u003c/p\u003e\n\u003cp\u003eIndividual pipeline steps are explained as follows.\u003c/p\u003e\n\u003cp\u003e2.1 Read QC\u003c/p\u003e\n\u003cp\u003eASE data accuracy and robustness depend heavily on the quality of sequencing data, especially the effective coverage of the assayed SNPs, as shown in our previous publication (Wu et al., 2021). ASET uses FastQC (Andrews, 2010) and CollectRnaSeqMetrics from GATK (McKenna et al., 2010) to assess RNA-Seq read quality, and uses Trimmomatic (Bolger, Lohse and Usadel, 2014) to remove adapter contamination and low-quality ends. QC metrics are summarized in both a MultiQC (Ewels et al., 2016) report and a tabular spreadsheet.\u003c/p\u003e\n\u003cp\u003e2.2 Read alignment\u003c/p\u003e\n\u003cp\u003eASET currently contains four sub-workflow choices for read alignment. The mapper parameter specified in the configuration file selects one of these alignment approaches: \u0026nbsp;(1) STAR+WASP where the alignment is performed using STAR with the --waspOutputMode parameter to enable WASP filtering; (2) STAR+NMASK where the genome is first N-masked at the SNP sites and then used for STAR alignment; (3) GSNAP where reads are aligned using GSNAP in the SNP-tolerant mode; and (4) ASElux where reads are aligned and counted using ASElux. Note that the provided genome FASTA and GTF files will be indexed by the chosen aligner for splice-aware alignment.\u003c/p\u003e\n\u003cp\u003e2.3 Alignment filtering, deduplication, and strand split\u003c/p\u003e\n\u003cp\u003eAlignments are filtered based on adjustable flags and mapping quality cutoffs. STAR+WASP-based alignments can additionally exclude alignments flagged as problematic (based on vW tag). Reads are then deduplicated using GATK MarkDuplicates. Deduplicated reads are split into two alignment files based on strand. A strandedness parameter needs to be provided to indicate whether read 1 or read 2 corresponds to the original RNA strand. (ASElux-based alignments skip this step as ASElux integrates both read alignment and counting without emitting the alignment files for manipulation.)\u003c/p\u003e\n\u003cp\u003e2.4 ASE read counting\u003c/p\u003e\n\u003cp\u003eGATK ASEReadCounter is applied on each alignment file to compute allele-specific read counts on all provided heterozygous and homozygous SNPs, and optionally the genotyped reference sites. Output files on different strands from all samples are concatenated into a single file for each type of site. Base quality cutoffs, mapping quality cutoffs, and the overlap handling scheme are configurable. (As above, ASElux-based alignments skip this step.)\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e2.5 Contamination estimation\u003c/p\u003e\n\u003cp\u003eThe average non-alternative-allele frequency on homozygous SNP sites, and the average non-reference-allele frequency on reference sites (if available), are calculated to serve as an estimate of cross-contamination (or mislabeling) for each sample. For placental samples where maternal contamination is a concern, the average non-reference-allele frequency at the reference sites where the mother has a non-reference genotype is also calculated for each gene individually, with the assumption that the non-reference allele counts arise from contamination by maternal tissue. (ASElux-based alignments skip this step since ASElux only counts reads at exonic heterozygous SNPs.)\u003c/p\u003e\n\u003cp\u003e2.6 Annotation\u003c/p\u003e\n\u003cp\u003eBased on the provided GTF, the exons from the same gene are merged into a union exon set and then used to annotate a table of SNPs. Each SNP (row) details exon coordinates, gene IDs, symbols, and gene types. When phasing data is provided, paternal and maternal alleles will be indicated, and the paternal allele frequency will be calculated for each SNP that has data.\u003c/p\u003e\n\u003cp\u003e2.7 ASET outputs\u003c/p\u003e\n\u003cp\u003eThe ASET data preparation steps emit an ASE data table that contains count data, contamination estimates, and gene/exon annotation, from all samples; and an R data file that can be used for visualization and downstream PofO testing, using ASEplot and po_test.jl, respectively. Additionally, these steps also produce trimmed FASTQ files, alignment BAM files, MultiQC reports, and a QC tabular spreadsheet.\u003c/p\u003e\n\u003cp\u003e2.8 Determination of parent-of-origin scores\u003c/p\u003e\n\u003cp\u003eFor a given gene with \u003cem\u003eN\u003c/em\u003e total read counts and \u003cem\u003em\u003c/em\u003e distinct SNPs, let \u003cem\u003eY\u003csub\u003eijk\u003c/sub\u003e\u003c/em\u003e denote the read count for allele \u003cem\u003ek\u003c/em\u003e of SNP \u003cem\u003ej\u003c/em\u003e for subject \u003cem\u003ei\u003c/em\u003e. The alleles are coded \u003cem\u003ek\u003c/em\u003e = 0, 1 for the reference and alternative alleles, respectively. Define \u003cem\u003eX\u003csub\u003eijk\u003c/sub\u003e\u003c/em\u003e= 1/2 when \u003cem\u003ek\u0026nbsp;\u003c/em\u003e= 0 and -1/2 when \u003cem\u003ek\u003c/em\u003e = 1; and define \u003cem\u003eZ\u003csub\u003eijk\u003c/sub\u003e\u003c/em\u003e = 1/2 and -1/2 for paternal and maternal allele read counts, respectively. Next, construct an \u003cem\u003eN\u003c/em\u003e \u0026times; \u003cem\u003em\u003c/em\u003e matrix of indicator variables \u003cem\u003eU\u003c/em\u003e, where column \u003cem\u003el\u003c/em\u003e of \u003cem\u003eU\u003c/em\u003e is defined as \u003cem\u003eU\u003csup\u003el\u003c/sup\u003e\u003csub\u003eijk\u003c/sub\u003e\u003c/em\u003e= 1 if \u003cem\u003ej\u003c/em\u003e = \u003cem\u003el\u003c/em\u003e and 0 otherwise. Next, let \u003cem\u003eV\u003c/em\u003e denote an \u003cem\u003eN\u003c/em\u003e \u0026times; \u003cem\u003eq\u003c/em\u003e matrix consisting of the left singular vectors of \u003cem\u003eU\u003c/em\u003e whose singular values are at least 1% of the maximum singular value of \u003cem\u003eU\u003c/em\u003e. We fit a cluster-robust quasi-Poisson regression model for each gene in which the indices \u003cem\u003ei\u003c/em\u003e, \u003cem\u003ej\u003c/em\u003e, \u003cem\u003ek\u003c/em\u003e index the \u003cem\u003eN\u003c/em\u003e observations, and the explanatory variables are the main effect of parent of origin (\u003cem\u003eZ\u003csub\u003eijk\u003c/sub\u003e\u003c/em\u003e), the main effect of ref/alt status (\u003cem\u003eX\u003csub\u003eijk\u003c/sub\u003e\u003c/em\u003e), main effects for SNP indicators (\u003cem\u003eV\u003c/em\u003e), and all pairwise interactions between SNP indicators (\u003cem\u003eV\u003c/em\u003e) and ref/alt status (\u003cem\u003eX\u003csub\u003eijk\u003c/sub\u003e\u003c/em\u003e). Including the \u003cem\u003eX\u003c/em\u003e and \u003cem\u003eV\u003c/em\u003e main effects and their pairwise interactions allows us to account for genetic ASE, while clustering on subjects (\u003cem\u003ei\u003c/em\u003e) allows us to account for correlations among read counts within the same individual (e.g. due to linkage disequilibrium). The full model is shown below:\u003c/p\u003e\n\u003cp\u003e\u003cimg width=\"605\" height=\"43\" src=\"data:image/png;base64,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\" alt=\"image\"\u003e\u003c/p\u003e\n\u003cp\u003eWe refer to the estimated coefficient for \u003cem\u003eZ\u003c/em\u003e as the PofO score and denote it \u003cem\u003epo\u003c/em\u003e, with its z-score denoted \u003cem\u003epo_z\u003c/em\u003e. Positive and negative \u003cem\u003epo\u003c/em\u003e correspond, respectively, to paternally and maternally biased expressions, while 0 denotes a balance. We view |\u003cem\u003epo\u003c/em\u003e| \u0026gt; 3 as denoting strong parentally determined ASE, implying at least a 20-fold difference between the two alleles, and |\u003cem\u003epo_z\u003c/em\u003e| \u0026gt; 3 as denoting statistical significance.\u003c/p\u003e"},{"header":"3 Results","content":"\u003cp\u003eWe applied ASET on the sequencing data from a set of 244 targeted RNA-Seq samples we previously published (Wu \u003cem\u003eet al.\u003c/em\u003e, 2021), using the STAR+WASP alignment approach. This produced a data table with 346,503 exonic SNP \u0026times; sample \u0026times; strand data points, observed in 783 genes. Using the ASEplot R library, we can visualize the SNP locations in specific genes (Figure 2 and Supplementary Figure 2), sample-level and gene-level contamination (Figure 3), and exon- and gene-level ASE distribution across different samples, exons, or genes (Figure 4, Figure 5, and Supplementary Figure 3). After data filtering including requiring at least 10 read counts at SNPs and lower than 5% contamination (when measurable), 264,046 data points were retained. The phased subset with 125,772 data points was analyzed using po_test.jl for PofO testing. The results showed that out of 392 genes that were testable, 153 had a strong PofO effect with |\u003cem\u003epo_z\u003c/em\u003e| \u0026gt; 3, with 92 biased to paternal expression and 61 biased to the maternal side. Among these genes, 33 had a large difference between the alleles with |\u003cem\u003epo\u003c/em\u003e| \u0026gt; 3.\u003c/p\u003e"},{"header":"4 Discussion","content":"\u003cp\u003eASET provides an integrated and reproducible framework for the generation and visualization of ASE data, addressing a critical need for streamlined ASE analysis in genomic studies. It combines a robust Nextflow-based workflow for data preprocessing with a dedicated R package for visualization and a statistical algorithm for PofO testing. Compared to other available ASE workflows, ASET provides a more complete solution by including multiple alignment approaches tailored for ASE analysis, support for strand-specific read counting, contamination estimation, data visualization, and PofO testing. ASET employs containerization through Docker and Singularity to boost convenience and reproducibility across different environments. The pipeline's modular structure provides flexibility for further expansion by the addition of more modules. For example, another sub-workflow can be added to enable personalized diploid genome construction and alignment when a complete phased SNP set is available. The current annotation of the SNPs by using the merged exons lacks the ability to interrogate isoform-level ASE. With diploid genome construction and sufficient density of heterozygous SNPs (e.g. from inbred mouse strains), there are approaches to resolve ASE quantification on the isoform-level (Turro \u003cem\u003eet al.\u003c/em\u003e, 2011; Perez \u003cem\u003eet al.\u003c/em\u003e, 2015). However, the best solution for isoform ASE analysis may lie in full-length transcriptome sequencing using long-read sequencing technologies (Glinos et al., 2022; Tang et al., 2024). The current support provided for downstream data analysis focuses on basic visualization and PofO testing. We realize that there is a variety of methods for downstream analyses, such as eQTL and prediction of \u003cem\u003ecis\u003c/em\u003e-acting ncRNA-targets (Hasenbein \u003cem\u003eet al.\u003c/em\u003e, 2025). In addition, haplotype-specific expression can be enabled using phASER, especially when long-read RNA-Seq data are available (Castel \u003cem\u003eet al.\u003c/em\u003e, 2016). We will be working on adding more functionality to ASET to incorporate diploid alignment, isoform-level ASE measurement, and further statistical analysis especially when phenotype data are available. Overall, compared to the existing alternative pipelines, ASET provides a more comprehensive workflow that bridges the gap between raw data and SNP-level ASE measurement and interpretation, and is particularly valuable for studies such as genomic imprinting, eQTLs, X chromosome inactivation and nonsense-mediated decay, where the preparation of robust ASE data is required.\u003c/p\u003e\n"},{"header":"Abbreviations","content":"\u003cdiv class=\"DefinitionList\"\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eASE Allele\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003especific expression\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eSNP Single\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003enucleotide polymorphism\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003ePofO Parent\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eof-origin\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eInformed consent or assent was obtained from participants depending on whether they were adults or children. Institutional Review Boards (IRB) approval was obtained from the University of Michigan IRBMED (HUM00043670) and from La Facult\u0026eacute; de M\u0026eacute;decine de Pharmacie et d\u0026apos;Odontostomatologie (FMPOS) de Bamako in Mali (No2016/68/CD/FMPOS).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eASET is available at \u003cu\u003ehttps://github.com/weishwu/ASET\u003c/u\u003e. The ASE data preparation section is implemented in Nextflow with DSL2 syntax. The data visualization functionality is provided as an R library, directly available from the ASET repository or from \u003cu\u003ehttps://github.com/weishwu/ASEplot\u003c/u\u003e. The PofO testing algorithm is implemented in a Julia script. ASET and ASEplot are also accessible as docker containers from Docker Hub: and https://hub.docker.com/repository/docker/weishwu/aseplot.\u003c/p\u003e\n\u003cp\u003eThe RNA-Seq FASTQ files and the genotype data used to test the pipeline were published in our previous paper (Wu \u003cem\u003eet al.\u003c/em\u003e, 2021), and deposited in dbGaP as phs001782.v2.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research was supported by the Eunice Kennedy Shriver National Institute of Child Health \u0026amp; Human Development (NICHD) of the National Institutes of Health (NIH) (R01HD104676, R01HD088521 and R21HD077465 to B.I.S.); and the John Templeton Foundation (JTF) (52269 to B.I.S.). The content of this study is solely the responsibility of the authors and doesn\u0026apos;t necessarily reflect the official views of the JTF, the NICHD, or the NIH.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026apos; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eW.W. developed the Nextflow pipeline and the ASEplot R library, wrote the main manuscript and prepared all figures and tables. K.S. developed the PofO testing model and wrote the section \u0026quot;Determination of parent-of-origin scores\u0026quot;. \u0026nbsp;C.V. contributed ideas to some of the pipeline modules and functions. B.S. provided funding and supervised the project. C.G. contributed ideas to pipeline code and manuscript structure. All authors reviewed the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors acknowledge support from the BRCF Bioinformatics Core at the University of Michigan.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAguet F et al. (2017) \u0026lsquo;Genetic effects on gene expression across human tissues\u0026rsquo;, \u003cem\u003eNature\u003c/em\u003e, 550(7675), pp. 204\u0026ndash;213. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/NATURE24277\u003c/span\u003e\u003cspan address=\"10.1038/NATURE24277\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAndergassen D et al. (2015) \u0026lsquo;Allelome.PRO, a pipeline to define allele-specific genomic features from high-throughput sequencing data\u0026rsquo;, \u003cem\u003eNucleic Acids Research\u003c/em\u003e, 43(21), p. e146. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/NAR/GKV727\u003c/span\u003e\u003cspan address=\"10.1093/NAR/GKV727\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAsiimwe R, Alexander D. (2024) \u0026lsquo;STAR\u0026thinsp;+\u0026thinsp;WASP reduces reference bias in the allele-specific mapping of RNA-seq reads\u0026rsquo;, \u003cem\u003ebioRxiv: the preprint server for biology\u003c/em\u003e [Preprint]. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1101/2024.01.21.576391\u003c/span\u003e\u003cspan address=\"10.1101/2024.01.21.576391\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBaran Y et al. (2015) \u0026lsquo;The landscape of genomic imprinting across diverse adult human tissues\u0026rsquo;, \u003cem\u003eGenome Research\u003c/em\u003e, 25(7), pp. 927\u0026ndash;936. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1101/GR.192278.115\u003c/span\u003e\u003cspan address=\"10.1101/GR.192278.115\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBezanson J et al. (2017) \u0026lsquo;Julia: A fresh approach to numerical computing\u0026rsquo;, \u003cem\u003eSIAM Review\u003c/em\u003e, 59(1), pp. 65\u0026ndash;98. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1137/141000671;CTYPE:STRING:JOURNAL\u003c/span\u003e\u003cspan address=\"10.1137/141000671;CTYPE:STRING:JOURNAL\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBhardwaj V et al. (2019) \u0026lsquo;SnakePipes: Facilitating flexible, scalable and integrative epigenomic analysis\u0026rsquo;, \u003cem\u003eBioinformatics\u003c/em\u003e, 35(22), pp. 4757\u0026ndash;4759. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/BIOINFORMATICS/BTZ436\u003c/span\u003e\u003cspan address=\"10.1093/BIOINFORMATICS/BTZ436\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCarrel L, Willard HF. (2005) \u0026lsquo;X-inactivation profile reveals extensive variability in X-linked gene expression in females\u0026rsquo;, \u003cem\u003eNature\u003c/em\u003e, 434(7031), pp. 400\u0026ndash;404. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/NATURE03479\u003c/span\u003e\u003cspan address=\"10.1038/NATURE03479\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCastel SE et al. (2015) \u0026lsquo;Tools and best practices for data processing in allelic expression analysis\u0026rsquo;, \u003cem\u003eGenome Biology\u003c/em\u003e, 16(1). Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/S13059-015-0762-6\u003c/span\u003e\u003cspan address=\"10.1186/S13059-015-0762-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCastel SE et al. (2016) \u0026lsquo;Rare variant phasing and haplotypic expression from RNA sequencing with phASER\u0026rsquo;, \u003cem\u003eNature Communications\u003c/em\u003e, 7(1), pp. 1\u0026ndash;6. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/NCOMMS12817\u003c/span\u003e\u003cspan address=\"10.1038/NCOMMS12817\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e;SUBJMETA=114,208,212,631,728,794;KWRD=GENOMICS,HAPLOTYPES,SOFTWARE.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCastel SE et al. (2020) \u0026lsquo;A vast resource of allelic expression data spanning human tissues\u0026rsquo;, \u003cem\u003eGenome Biology\u003c/em\u003e, 21(1), pp. 1\u0026ndash;12. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/S13059-020-02122-Z/FIGURES/2\u003c/span\u003e\u003cspan address=\"10.1186/S13059-020-02122-Z/FIGURES/2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDegner JF et al. (2009) \u0026lsquo;Effect of read-mapping biases on detecting allele-specific expression from RNA-sequencing data\u0026rsquo;, \u003cem\u003eBioinformatics\u003c/em\u003e, 25(24), pp. 3207\u0026ndash;3212. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/BIOINFORMATICS/BTP579\u003c/span\u003e\u003cspan address=\"10.1093/BIOINFORMATICS/BTP579\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDobin A et al. (2012) \u0026lsquo;STAR: ultrafast universal RNA-seq aligner\u0026rsquo;, \u003cem\u003eBioinformatics\u003c/em\u003e, 29(1), p. 15. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/BIOINFORMATICS/BTS635\u003c/span\u003e\u003cspan address=\"10.1093/BIOINFORMATICS/BTS635\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVan De Geijn B et al. (2015) \u0026lsquo;WASP: Allele-specific software for robust molecular quantitative trait locus discovery\u0026rsquo;, \u003cem\u003eNature Methods\u003c/em\u003e, 12(11), pp. 1061\u0026ndash;1063. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/NMETH.3582\u003c/span\u003e\u003cspan address=\"10.1038/NMETH.3582\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGlinos DA et al. (2022) \u0026lsquo;Transcriptome variation in human tissues revealed by long-read sequencing\u0026rsquo;, \u003cem\u003eNature 2022 608:7922\u003c/em\u003e, 608(7922), pp. 353\u0026ndash;359. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41586-022-05035-y\u003c/span\u003e\u003cspan address=\"10.1038/s41586-022-05035-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHasenbein TP et al. (2025) \u0026lsquo;Allele-specific genomics decodes gene targets and mechanisms of the non-coding genome\u0026rsquo;, \u003cem\u003ebioRxiv\u003c/em\u003e, p. 2025.03.03.641135. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1101/2025.03.03.641135\u003c/span\u003e\u003cspan address=\"10.1101/2025.03.03.641135\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKrueger F, Andrews SR. (2016) \u0026lsquo;SNPsplit: Allele-specific splitting of alignments between genomes with known SNP genotypes\u0026rsquo;, \u003cem\u003eF1000Research\u003c/em\u003e, 5. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.12688/F1000RESEARCH.9037.2/DOI\u003c/span\u003e\u003cspan address=\"10.12688/F1000RESEARCH.9037.2/DOI\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKurtzer GM, Sochat V, Bauer MW. (2017) \u0026lsquo;Singularity: Scientific containers for mobility of compute\u0026rsquo;, \u003cem\u003ePLoS ONE\u003c/em\u003e, 12(5). Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/JOURNAL.PONE.0177459\u003c/span\u003e\u003cspan address=\"10.1371/JOURNAL.PONE.0177459\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLappalainen T et al. (2013) \u0026lsquo;Transcriptome and genome sequencing uncovers functional variation in humans\u0026rsquo;, \u003cem\u003eNature\u003c/em\u003e, 501(7468), pp. 506\u0026ndash;511. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/NATURE12531\u003c/span\u003e\u003cspan address=\"10.1038/NATURE12531\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLonsdale J et al. (2013) \u0026lsquo;The Genotype-Tissue Expression (GTEx) project\u0026rsquo;, \u003cem\u003eNature Genetics\u003c/em\u003e, 45(6), pp. 580\u0026ndash;585. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/NG.2653\u003c/span\u003e\u003cspan address=\"10.1038/NG.2653\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e;SUBJMETA=114,129,208,631;KWRD=DATABASES,GENETICS.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcKenna A et al. (2010) \u0026lsquo;The genome analysis toolkit: A MapReduce framework for analyzing next-generation DNA sequencing data\u0026rsquo;, \u003cem\u003eGenome Research\u003c/em\u003e, 20(9), pp. 1297\u0026ndash;1303. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1101/GR.107524.110\u003c/span\u003e\u003cspan address=\"10.1101/GR.107524.110\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMerkelDirk. (2014) \u0026lsquo;Docker: lightweight linux containers for consistent development and deployment\u0026rsquo;, \u003cem\u003eLinux Journal\u003c/em\u003e [Preprint]. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5555/2600239.2600241\u003c/span\u003e\u003cspan address=\"10.5555/2600239.2600241\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMiao Z et al. (2018) \u0026lsquo;ASElux: An ultra-fast and accurate allelic reads counter\u0026rsquo;, \u003cem\u003eBioinformatics\u003c/em\u003e, 34(8), pp. 1313\u0026ndash;1320. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/BIOINFORMATICS/BTX762\u003c/span\u003e\u003cspan address=\"10.1093/BIOINFORMATICS/BTX762\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePerez JD et al. (2015) \u0026lsquo;Quantitative and functional interrogation of parent-of-origin allelic expression biases in the brain\u0026rsquo;, \u003cem\u003eeLife\u003c/em\u003e, 4(JULY 2015), p. 41. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.7554/ELIFE.07860\u003c/span\u003e\u003cspan address=\"10.7554/ELIFE.07860\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eR Core Team. (2013) \u003cem\u003eR: a language and environment for statistical computing\u003c/em\u003e, \u003cem\u003eR Foundation for Statistical Computing, Vienna, Austria\u003c/em\u003e. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.R-project.org/\u003c/span\u003e\u003cspan address=\"https://www.R-project.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (Accessed: 14 May 2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRivas MA et al. (2015) \u0026lsquo;Effect of predicted protein-truncating genetic variants on the human transcriptome\u0026rsquo;, \u003cem\u003eScience\u003c/em\u003e, 348(6235), pp. 666\u0026ndash;669. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1126/SCIENCE.1261877\u003c/span\u003e\u003cspan address=\"10.1126/SCIENCE.1261877\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRozowsky J et al. (2011) \u0026lsquo;AlleleSeq: Analysis of allele-specific expression and binding in a network framework\u0026rsquo;, \u003cem\u003eMolecular Systems Biology\u003c/em\u003e, 7. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/MSB.2011.54\u003c/span\u003e\u003cspan address=\"10.1038/MSB.2011.54\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchmitz RJ et al. (2013) \u0026lsquo;Patterns of population epigenomic diversity\u0026rsquo;, \u003cem\u003eNature\u003c/em\u003e, 495(7440), pp. 193\u0026ndash;198. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/NATURE11968\u003c/span\u003e\u003cspan address=\"10.1038/NATURE11968\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTang AD et al. (2024) \u0026lsquo;Detecting haplotype-specific transcript variation in long reads with FLAIR2\u0026rsquo;, \u003cem\u003eGenome Biology\u003c/em\u003e, 25(1), pp. 1\u0026ndash;21. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/S13059-024-03301-Y/FIGURES/4\u003c/span\u003e\u003cspan address=\"10.1186/S13059-024-03301-Y/FIGURES/4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTommaso DI et al. P. (2017) \u0026lsquo;Nextflow enables reproducible computational workflows\u0026rsquo;, \u003cem\u003eNature Biotechnology\u003c/em\u003e, 35(4), pp. 316\u0026ndash;319. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/NBT.3820\u003c/span\u003e\u003cspan address=\"10.1038/NBT.3820\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTurro E et al. (2011) \u0026lsquo;Haplotype and isoform specific expression estimation using multi-mapping RNA-seq reads\u0026rsquo;, \u003cem\u003eGenome Biology\u003c/em\u003e, 12(2), pp. 1\u0026ndash;15. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/GB-2011-12-2-R13/TABLES/3\u003c/span\u003e\u003cspan address=\"10.1186/GB-2011-12-2-R13/TABLES/3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu TD et al. (2016) \u0026lsquo;GMAP and GSNAP for genomic sequence alignment: Enhancements to speed, accuracy, and functionality\u0026rsquo;, \u003cem\u003eMethods in Molecular Biology\u003c/em\u003e, 1418, pp. 283\u0026ndash;334. Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/978-1-4939-3578-9_15\u003c/span\u003e\u003cspan address=\"10.1007/978-1-4939-3578-9_15\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu W et al. (2021) \u0026lsquo;Targeted RNA-seq improves efficiency, resolution, and accuracy of allele specific expression for human term placentas\u0026rsquo;, \u003cem\u003eG3: Genes, Genomes, Genetics\u003c/em\u003e, 11(8). Available at: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/G3JOURNAL/JKAB176\u003c/span\u003e\u003cspan address=\"10.1093/G3JOURNAL/JKAB176\" 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":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-bioinformatics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"binf","sideBox":"Learn more about [BMC Bioinformatics](http://bmcbioinformatics.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/binf","title":"BMC Bioinformatics","twitterHandle":"@BMC_Bioinformatics","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-6844336/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6844336/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eMotivation\u003c/strong\u003e Allele-specific expression (ASE) analyses from RNA-Seq data provide quantitative insights into imprinting and genetic variants affecting transcription. Robust ASE analysis requires the integration of multiple computational steps, including read alignment, read counting, data visualization, and statistical testing—this complexity creates challenges around reproducibility, scalability, and ease of use.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults\u003c/strong\u003e Here, we present ASE Toolkit (ASET), an end-to-end pipeline that streamlines SNP-level ASE data generation, visualization, and testing for parent-of-origin (PofO) effect. ASET includes a modular pipeline built with Nextflow for ASE quantification from short-read transcriptome sequencing reads, an R library for data visualization, and a Julia script for PofO testing. ASET performs comprehensive read quality control, SNP-tolerant alignment to reference genomes, read counting with allele and strand resolution, annotation with genes and exons, and estimation of contamination. In sum, ASET provides a complete and easy-to-use solution for molecular and biomedical scientists to identify and interpret patterns in ASE from RNA-Seq data.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eASET is available at \u003cu\u003ehttps://github.com/weishwu/ASET\u003c/u\u003e. The ASE data preparation section is implemented in Nextflow with DSL2 syntax. The data visualization functionality is provided as an R library, directly available from the ASET repository or from \u003cu\u003ehttps://github.com/weishwu/ASEplot\u003c/u\u003e. The PofO testing algorithm is implemented in a Julia script. ASET and ASEplot are also accessible as docker containers from Docker Hub: and https://hub.docker.com/repository/docker/weishwu/aseplot.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eContact\u003c/strong\u003e\u003c/p\u003e\n\u003cp\[email protected]\u003c/p\u003e","manuscriptTitle":"ASET: An end-to-end pipeline for quantification and visualization of allele specific expression","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-06-13 03:59:20","doi":"10.21203/rs.3.rs-6844336/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-06-17T14:05:33+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-06-09T23:56:04+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-06-09T23:55:00+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Bioinformatics","date":"2025-06-07T18:32:30+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-bioinformatics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"binf","sideBox":"Learn more about [BMC Bioinformatics](http://bmcbioinformatics.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/binf","title":"BMC Bioinformatics","twitterHandle":"@BMC_Bioinformatics","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"ca570ed6-2f40-4e25-84d4-0891889980c5","owner":[],"postedDate":"June 13th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-10-27T16:35:58+00:00","versionOfRecord":{"articleIdentity":"rs-6844336","link":"https://doi.org/10.1186/s12859-025-06282-2","journal":{"identity":"bmc-bioinformatics","isVorOnly":false,"title":"BMC Bioinformatics"},"publishedOn":"2025-10-21 16:17:13","publishedOnDateReadable":"October 21st, 2025"},"versionCreatedAt":"2025-06-13 03:59:20","video":"","vorDoi":"10.1186/s12859-025-06282-2","vorDoiUrl":"https://doi.org/10.1186/s12859-025-06282-2","workflowStages":[]},"version":"v1","identity":"rs-6844336","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6844336","identity":"rs-6844336","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}