Genome-wide Association Analysis and Candidate Gene Identification of Plant Height in Dabaigu (Foxtail Millet)

Genome-wide Association Analysis and Candidate Gene Identification of Plant Height in Dabaigu (Foxtail Millet) | 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 Genome-wide Association Analysis and Candidate Gene Identification of Plant Height in Dabaigu (Foxtail Millet) Wei Zhou, Huibin Qin, Rui Huang, Sen Hou, Junjie Wang, Ling Chen, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7333962/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 5 You are reading this latest preprint version Abstract Plant height is pivotal for lodging resistance, harvestability, and yield potential in foxtail millet ( Setaria italica ). We phenotyped 209 diverse Dabaigu accessions across three contrasting environments and performed whole-genome resequencing (mean depth ~38×). After standard filtering, 833,157 high-quality SNPs were retained. Neighbor-joining and principal component analysis consistently resolved four genetic groups with minor differences in sample assignment. A mixed linear model GWAS (EMMAX) detected four loci significantly associated with plant height. Comparative genomics with rice and Arabidopsis shortlisted ten candidates; notably, SETIT_033071mg showed stem-specific high expression at the shooting stage in Yugu1, with low expression elsewhere. Haplotype analysis of SETIT_033071mg in an expanded panel of 313 accessions across five field sites resolved six gene haplotypes (H1–H6), among which H2 was consistently associated with reduced plant height across environments. In parallel, we developed a KASP assay (kPH8552) tagging the lead SNP at qPH2.2 (distinct from the SETIT_033071mg locus). In the Dabaigu panel, the assay achieved an 80.4% call rate (G:G=74; T:T=94), and the T allele was consistently associated with reduced height by −4.48 cm (BLUE; P = 0.044). These findings refine the genomic basis of plant height in foxtail millet and provide actionable targets and a candidate marker for breeding lodging-resistant, semi-dwarf cultivars, pending validation in independent populations. Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Key Message Multi-environment GWAS reveals a dwarfing haplotype at SETIT_033071mg and a small-effect marker at qPH2.2 , informing selection for lodging-resistant foxtail millet. Introduction The mid-twentieth-century Green Revolution—anchored by semi-dwarf, lodging-resistant wheat and rice cultivars and supported by irrigation and agrochemical inputs—greatly increased cereal yields in the developing world (Ameen and Raza 2017 ). Meeting the food needs of a projected 9.7 billion people by 2050 will nevertheless require an additional 50–60% rise in global production (Falcon et al. 2022 ). Plant height remains a pivotal breeding target: suitably reduced stature enhances lodging resistance, raises the harvest index, and enables fully mechanised harvesting, traits that underpinned previous yield breakthroughs and will be vital for future gains (Fernandez et al. 2009 ; Hedden 2003 ). Therefore, identifying and utilizing height-regulating genes is pivotal for driving the next generation of productivity gains in cereal crops. Foxtail millet ( Setaria italica ) was domesticated from green foxtail in the Yellow River basin ~ 16 000 YBP and, thanks to its small diploid genome, short life cycle and strict self-pollination, now serves both as a climate-resilient staple in arid regions and as a model for Panicoideae genomics (Bennetzen et al. 2012 ; Diao and Jia 2016 ; Goron and Raizada 2015 ; Panaud 2006 ; Peng et al. 1999 ). Richer in protein, vitamins and minerals than rice or wheat and exceptionally tolerant of drought and poor soils (Das and Rakshit 2016 ; Kalita et al. 2025 ), foxtail millet nonetheless suffers severe lodging at the grain-filling stage because dense planting meets intrinsically tall culms (Tian et al. 2017 ). Breeding shorter, lodging-resistant cultivars is therefore urgent, and systematically dissecting the genetic basis of plant height will provide the essential targets and tools for molecular improvement. Extensive QTL dissection for plant height has been achieved in major cereals. Rice harbours more than 100 QTLs across its 12 chromosomes (Sabouri et al. 2010 ; Sandhu et al. 2021 ; Sowadan et al. 2018 ; Srividhya et al. 2011 ; Zhao et al. 2011 ), with sd1 , d18 and d35 conferring large, stable effects (Itoh et al. 2004 ; Itoh et al. 2002 ; Monna et al. 2002 ). Wheat carries in excess of 50 height loci (Griffiths et al. 2012 ; Würschum et al. 2015 ), and both linkage mapping and GWAS repeatedly identify the Rht-B1 / Rht-D1 allelic series as the Green-Revolution foundation (Peng et al. 1999 ). In foxtail millet, studies remain recent and centered on F₂/RIL/natural-population panels, yielding over 100 mapped plant height QTLs across nine chromosomes; however, only 24 replicate consistently and many exhibit marked environment dependence (Fan et al. 2017 ; Han et al. 2024 ; He et al. 2021 ; Mauro-Herrera and Doust 2016 ; Zhang et al. 2017 ; Zhu et al. 2023 ). Cloning of plant-height genes has revolutionized cereal breeding. In rice, sd1 (GA20-oxidase)(Monna et al. 2002 ) and SLR1 (DELLA)(Ikeda et al. 2001 ), together with the wheat Rht-B1b/Rht-D1b allelic series(Peng et al. 1999 ), demonstrate that attenuating gibberellin (GA) signaling is a universal strategy for generating semi-dwarf phenotypes. In maize, D8/D9 (Lawit et al. 2010 ), and in sorghum, Dw3 (Multani et al. 2003 ), further implicate additional pathways involving abscisic acid (ABA), auxin, and brassinosteroid/strigolactone signaling. In foxtail millet, however, only a handful of GA-related genes— SiSD1 (Ni et al. 2017 ), SiGID1 (Zhao et al. 2019 ), and Sidwarf3 (Fan et al. 2017 )—have been characterized, mostly at the transcriptional level and without comprehensive mutant validation. Pathways related to cell-wall biosynthesis, cross-talk among GA, auxin, and brassinosteroids, or strigolactone signaling remain virtually unexplored. This GA-centric focus leaves other hormonal and structural mechanisms (e.g., cell-wall signaling) underexplored, creating a gap between locus discovery and breeding-ready targets. Here, we conduct a multi-environment GWAS in a Dabaigu panel and, from ~ 833k high-quality SNPs, identify four plant-height loci. Guided by cross-species annotation and shooting-stage stem expression in Yugu1, we prioritize SETIT_033071mg ( WAK4 ) within qPH2.1 and perform a coding-SNP haplotype analysis in 313 accessions, highlighting an H2 haplotype associated with shorter stature. In parallel, we develop a KASP assay (kPH8552) for the lead SNP at qPH2.2 (distinct from the SETIT_033071mg interval) as a candidate tool for marker-assisted improvement; its small but consistent effect (BLUE Δ = −4.48 cm; P = 0.044) motivates independent validation. Materials and methods Plant materials and field trials A diversity panel of 209 foxtail millet ( Setaria italica ) landraces (hereafter “Dabaigu”) was obtained from the Center for Agricultural Genetic Resources Research, Shanxi Agricultural University (Taiyuan, China). Trials were conducted for three consecutive years (2020–2022) at Jinzhong, Shanxi, China (37.68°N, 112.75°E). Within each year, accessions were randomized and planted as a single 6-m row per accession with 30 cm row spacing; standard local management (irrigation, manual weeding and integrated pest control) was applied. At physiological maturity, plant height (ground to panicle tip) was measured on three randomly chosen plants per row and averaged for analyses. Phenotypic analysis, BLUEs and broad-sense heritability Summary statistics (mean, SD, CV, range, skewness, kurtosis) were computed in R; histograms with overlaid density was drawn with ggplot2 (v3.5.1). Across-environment BLUEs were obtained using lme4 (v1.1-37) with the model: $$\:\text{P}\text{H}\:\sim\:\:\text{Env}\:+\:\left(1\:∣\:\text{Genotype}\right)\:+\:\left(1\:∣\:\text{Genotype}:\text{Env}\right)$$ treating Env as fixed and Genotype and Genotype×Env as random; conditional genotype means (BLUEs) were extracted with emmeans. Variance components were obtained from the same mixed model, and broad-sense heritability on an entry-mean basis was computed as: $$\:{H}^{2}=\frac{{{\sigma\:}}_{g}^{2}}{{\sigma\:}_{g}^{2}+\frac{{\sigma\:}_{ge}^{2}}{n}+\frac{{{\sigma\:}}_{e}^{2}}{nr}}$$ with n = 3 environments and r = 1 row per environment. DNA extraction, library preparation and whole-genome resequencing Leaf Young leaves were collected at the shooting stage in 2022. Genomic DNA was extracted using a modified CTAB method, quality-checked by agarose electrophoresis and Qubit fluorometry (Thermo Fisher). Libraries (insert ~ 500 bp) were prepared (end repair, A-tailing, adapter ligation), circularized, amplified as DNA nanoballs and sequenced on the BGI DNBSEQ™ platform to generate 150-bp paired-end reads. Read processing, alignment and variant calling Adapters, reads with > 10% Ns, and reads with > 50% bases of Q ≤ 12 were removed by SOAPnuke (v0.20.0). Clean reads were aligned to the Yugu1 reference genome (GCF_000263155.2, v2.0) using BWA-MEM (v0.7.17). SAM files were converted, sorted and deduplicated with SAMtools and Picard; only reads with MAPQ ≥ 30 were retained. Variants were called per sample with GATK (v4.6.2) HaplotypeCaller (gVCF mode), combined with CombineGVCFs and jointly genotyped with GenotypeGVCFs. Hard filters followed GATK best practices (QD ≥ 2.0, FS ≤ 60, SOR ≤ 3.0, MQ ≥ 40, MQRankSum ≥ − 12.5, ReadPosRankSum ≥ − 8.0). SNPs were retained if biallelic, MAF ≥ 0.05, missing rate ≤ 10%; HWE filtering was not applied given the selfing mating system and population structure. Population structure, PCA and linkage disequilibrium For structure analyses, we used an LD-pruned SNP set (PLINK 1.9, --indep-pairwise 50 5 0.2). PCA was performed with PLINK; population structure was inferred by ADMIXTURE (v1.3.0) with K = 2–8, and the optimal K was chosen by CV error minimization. Genome-wide LD decay was estimated by PopLDdecay (v3.43) as mean pairwise r² vs physical distance; LD decay distance was defined as the first point where mean r² dropped below 0.20. GWAS for Plant height Genome-wide association studies (GWAS) were conducted using EMMAX to identify SNPs significantly associated with plant height. A mixed linear model (MLM) was applied. The genome-wide significance threshold was set at P 7.22) based on Bonferroni correction (0.05/833,157). To avoid excessive false negatives resulting from this conservative threshold, SNPs with P 7) were also considered suggestively associated, following previous studies (Kang et al. 2010 ; Khound et al. 2024 ). GWAS analyses were performed on four datasets (20JZ, 21JZ, 22JZ, and BLUE), and significant loci were determined by integrating these results with prior findings. Candidate gene identification and functional annotation Candidate genes within each significant QTL region were identified using the Yugu1 reference genome ( https://www.ncbi.nlm.nih.gov/datasets/genome/GCF_00026315 5.2/). Genes located in the defined intervals (including an additional 100-kb flanking region) were extracted and functionally annotated based on UniProt ( https://www.uniprot.org/ ) and NCBI databases ( https://www.ncbi.nlm.nih.gov/ ). Furthermore, orthologous gene searches were performed against rice ( https://rapdb.dna.affrc.go.jp/ ) and Arabidopsis ( https://www.arabidopsis.org/ ) protein sequences using the online NCBI BLASTP tool ( https://blast.ncbi.nlm.nih.gov/Bla st.cgi) to infer potential gene functions. Relevant literature was also consulted to support the functional prediction and prioritize candidate genes. Tissue-Specific Expression Analysis and Haplotype Variation Analysis . Expression profiles of candidate genes in stems, leaves, nodes, panicles, and seeds at different developmental stages (seeding, three-leaf, booting, shooting, flowering, and maturity) were retrieved from the Setaria-db database ( http://www.setariadb.com/millet ). Heatmaps illustrating the expression patterns across organs and stages were generated using the R package pheatmap (v1.0.12). Haplotype analysis of the SETIT_033071mg gene was performed using resequencing data from 313 core foxtail millet accessions (unpublished). Plant height measurements were collected in Jinzhong (2018, 2019, 2020), Datong (2020), and Yuncheng (2020). The associations between haplotype variants and phenotypic traits were evaluated using the geneHapR package. KASP marker development The lead SNP at qPH2.2 was converted to a KASP assay (kPH8552) following Biosearch Technologies guidelines. Allele-specific primers were designed with PolyMarker (Supplementary Table 1); each forward primer carried the universal FAM (5′-GAAGGTGACCAAGTTCATGCT-3′) or HEX (5′-GAAGGTCGGAGTCAACGGATT-3′) tail. PCRs were run on a Roche LightCycler® 480 in 10 µL reactions (50–100 ng DNA, 2× KASP master mix, 0.14 µL primer cocktail) with touchdown cycling (94°C 15 min; 10 cycles 94°C 20 s, 61→55°C − 0.6°C/cycle 60 s; then 26 cycles 94°C 20 s, 55°C 60 s; final 37°C 1 min). Results Phenotypic variations of plant height PH was measured in 209 foxtail millet accessions across three experimental environments (Supplementary Table 2). The PH phenotypes exhibited approximately normal distributions (Fig. 1A) with substantial variation (Fig. 1B). The mean PH ranged from 125.04 cm (22JZ) to 161.01 cm (21JZ), with overall phenotypic variation spanning 90.00–194.42 cm. The coefficients of variation (CV) varied slightly across environments, ranging from 9.85% (21JZ, lowest variability) to 12.02% (20JZ, highest variability). Skewness and kurtosis values indicated that PH distributions were generally negatively skewed across all environments, with kurtosis values negative for 21JZ and 22JZ, while slightly positive for 20JZ. Plant height showed a high broad-sense heritability ( H ²) of 93.6%, indicating that genetic effects account for the vast majority of the observed phenotypic variation. Figure 1. Frequency distribution and descriptive statistics of plant height across three environments. A: Frequency distribution of PH in 209 foxtail millet accessions evaluated in three environments: 20JZ (Jinzhong, 2020), 21JZ (Jinzhong, 2021), and 22JZ (Jinzhong, 2022). Curves indicate kernel-density estimates. B: Statistical summary of PH data across the three environments. CV, coefficient of variation; H ², broad-sense heritability. Sequencing, SNP identification Table 1 Summary of marker distribution among whole foxtail millet genome Chromosome Length (Bp) SNP numbers SNPs/Mb 1 42,145,358 79,358 1,882 2 49,199,692 96,374 1,959 3 50,651,948 110,235 2,177 4 40,407,787 96,163 2,380 5 47,252,504 81,593 1,726 6 36,014,412 85,545 2,375 7 35,964,117 65,385 1,818 8 40,688,952 125,255 3,079 9 58,970,299 93,249 1,581 Total 401,295,069 833,157 2,109 To comprehensively characterize the genomic variation in foxtail millet, we conducted deep resequencing of 209 accessions, generating a total of 12.33 billion clean reads, with an average of 59 million reads per accession (Supplementary Table 3). The average GC content, Q20, and Q30 values were 45.2%, 97.3%, and 91.9%, respectively. More than 98% of reads were successfully mapped to the reference genome version 2 (Bennetzen et al. 2012 ), with a coverage rate exceeding 93% and an average sequencing depth ≈ 38× (mean = 37.6×) across accessions. After stringent filtering a total of 833,157 high-quality SNPs were retained (Fig. 2 A). These SNPs were uniformly distributed across the genome, with an average density of 2,109 SNPs per megabase (Mb). Chromosome 8 exhibited the highest SNP density (3,079 SNPs/Mb), whereas chromosome 9 had the lowest (1,581 SNPs/Mb) (Table 1 ). Based on functional annotation, the majority of SNPs (approximately 62.9%) were located in intergenic regions. Genic regions accounted for 33.6% of SNPs, while only 1.9% and 1.5% were found within pseudogenes and lncRNA regions, respectively (Fig. 2 B). Phylogenetic relationships, population structure, principal component analysis, and linkage disequilibrium analysis An NJ tree constructed from pairwise genetic distances resolved four groups (Fig. 3 A). Group sizes were uneven, with Group 2 the largest and Group 1 the smallest (Supplementary Table 4). Principal component analysis (PCA) recapitulated the same four-cluster pattern (Fig. 3 B), although sample assignments were not identical between methods, consistent with admixture and a complex genetic background. Further analysis using ADMIXTURE revealed that population boundaries were most distinct at K = 4. At this optimal value, some accessions exhibited pure ancestry, while others displayed admixed genetic backgrounds, suggesting extensive gene flow among groups (Fig. 3 D). The cross-validation (CV) error sharply decreased from K = 2 to K = 4 and reached its lowest point at K = 4, then gradually increased with higher K values (Fig. S1), supporting K = 4 as the optimal number of subpopulations. To investigate linkage disequilibrium (LD) patterns, we analyzed LD decay across the entire panel and within each cluster. In the whole panel, LD declined to R² = 0.2 at approximately 35 kb (Fig. 3 C). Each cluster exhibited distinct LD decay rates: Cluster 1 displayed the slowest decay, indicating a higher degree of genetic homogeneity, while Cluster 2 showed the fastest decay, suggesting more frequent recombination events and higher genetic diversity. GWAS analysis Using 833,157 high-quality SNPs, we conducted GWAS for plant height across three environments (20JZ, 21JZ, and 22JZ) and the BLUE values. In total, four stable QTL loci significantly associated with plant height were identified (Fig. 4 ; Supplementary Table 5). On chromosome 8, four significant SNPs (chr8:910320, chr8:911022, chr8:911033, and chr8:911623) were located within a 1.3 kb interval, which is much smaller than the estimated LD decay distance (~ 35 kb) in our panel. Therefore, these SNPs were merged into a single QTL locus ( qPH8 ), and chr8:911022, with the lowest P value, was considered as the lead SNP. Among the identified loci, two were located on chromosome 2 ( qPH2.1 and qPH2.2 ), while chromosomes 4 and 8 each harbored one locus ( qPH4 and qPH8 , respectively). The strongest association signal was detected at qPH4 on chromosome 4, which had the lowest P value among all loci. The QTLs qPH2.1 , qPH2.2 , qPH4 , and qPH8 were detected in one, two, three, and three environments, respectively. The PVE ranged from 10.62% ( qPH2.1 ) to 17.12% ( qPH2.2 ), with qPH2.2 exhibiting the highest PVE (Table 2 ). The quantile–quantile (QQ) plots indicated that the observed p-values generally conformed to the expected distribution under the null hypothesis, suggesting that the population structure and relatedness were well controlled in the GWAS model. Only the tail deviations represented true association signals (Fig. 4 ). Candidate gene Table 2 Stable QTL loci and candidate genes identified in this study QTL Chr QTL region (bp) QTN marker P value Environment PVE (%) Candidate gene Annotation qPH2.1 2 5091820–5291820 chr2:5191820 7.734E-08 20JZ 10.62 SETIT_033071mg wall-associated kinase 4 SETIT_029387mg FAD-linked oxidases family protein SETIT_028865mg NB-ARC domain-containing disease resistance protein qPH2.2 2 24636879–24836879 chr2:24736879 3.20E-08 20JZ/BLUE 17.12 SETIT_030569mg Chlorophyll A-B binding family protein qPH4 4 8128049–8328049 chr4:8228049 3.25E-09 20JZ/22JZ/BLUE 13.19 SETIT_007662mg basic helix-loop-helix (bHLH) DNA-binding superfamily protein SETIT_007340mg HVA22 homologue A qPH8 8 811022–1011022 chr8:911022 4.51E-09 20JZ/22JZ/BLUE 12.14 SETIT_026086mg with no lysine (K) kinase 5 SETIT_026573mg dsRNA-binding domain-like superfamily protein SETIT_027872mg Tetratricopeptide repeat (TPR)-like superfamily protein SETIT_026360mg Protein phosphatase 2C family protein Based on the estimated LD decay (R² = 0.2 at approximately 35 kb), a conservative ± 100 kb window around each significant GWAS locus was used to define candidate gene regions, as suggested in previous studies (Dai et al. 2024 ; Jia et al. 2013 ; Liu et al. 2022 ). Within these intervals we identified a total of 61 candidate genes (Supplementary Table 6). Through functional annotation of genes within these regions, several promising candidate genes potentially related to plant height regulation were identified (Table 2 ). Notably, in qPH2.1 on chromosome 2, we identified SETIT_033071mg , encoding a wall-associated kinase ( WAK ), which is involved in cell wall integrity and signal transduction, and SETIT_029387mg , encoding an FAD-linked oxidases family protein. Additionally, SETIT_028865mg , encoding an NB-ARC domain-containing disease resistance protein. For qPH2.2 , SETIT_030569mg , encoding a chlorophyll A-B binding family protein. In qPH4, we identified SETIT_007662mg , encoding a basic helix-loop-helix (bHLH) DNA-binding superfamily protein, as well as SETIT_007340mg , encoding HVA22 homologue A. In qPH8 on chromosome 8, multiple candidate genes were detected, including SETIT_026086mg , encoding a with-no-lysine (WNK) kinase 5 potentially; SETIT_026573mg , encoding a dsRNA-binding domain-like superfamily protein; SETIT_027872mg , encoding a tetratricopeptide repeat (TPR)-like superfamily protein; and SETIT_026360mg , encoding a protein phosphatase 2C family protein. Tissue- and stage-specific expression profiles of candidate gene Expression profiling of ten plant-height candidate genes across six developmental stages (seedling, three-leaf, shooting, booting, flowering, and maturity) and six tissues (seed, root, stem, node, leaf, and panicle) revealed distinct spatiotemporal patterns (Fig. S2). Specifically, SETIT_029387mg exhibited elevated expression in leaves and panicles from shooting to flowering stages. Genes SETIT_026573mg , SETIT_007662mg , SETIT_026360mg , SETIT_027872mg , and SETIT_007340mg displayed high expression in panicles during booting and flowering stages. Additionally, SETIT_028865mg and SETIT_027872mg showed elevated expression in stems and nodes at the booting stage. Notably, SETIT_033071mg exhibited high expression exclusively in stems at the shooting stage, while remaining lowly expressed across all other tissues and developmental stages. Haplotype and functional marker analysis of SETIT_033071mg Based on expression profiles and functional annotation within the qPH2.1 interval, SETIT_033071mg ( WAK4 ) was prioritized for haplotype analysis. Twenty-eight high-confidence coding SNPs partitioned 313 accessions into six major haplotypes (H1–H6). Across five field environments (18JZ, 19JZ, 20JZ, 20DT, 20YC), pairwise comparisons indicated that H2 tended to exhibit shorter plants than H1, H4 and H5, with significance in most environments (Wilcoxon tests, P < 0.05; Fig. 6 C). In parallel, we designed a KASP assay (kPH8552) for the lead SNP at qPH2.2 . In the Dabaigu panel (n = 209), the assay achieved an 80.4% call rate (G:G = 74; T:T = 94) and produced two well-separated homozygous clusters (Fig. 6 B). Based on across-environment BLUE estimates, the T allele was associated with a small but statistically significant reduction in plant height (− 4.48 cm; P = 0.044), with year-wise effects of − 4.15, − 1.13 and − 3.27 cm in 20JZ, 21JZ and 22JZ, respectively (Supplementary Table 7). These results indicate that kPH8552 tags a small-effect locus at qPH2.2 , with a consistent direction of effect across environments Discussion Well-powered GWAS enabled by multi-year phenotyping of a diverse panel The present study assessed plant height in a genetically diverse panel of 209 Dabaigu foxtail millet accessions sourced from contrasting eco-regions of northern China and grown for three successive seasons (2020–2022) under standardised field conditions in Jinzhong. Jinzhong, experiences a typical temperate continental monsoon climate and is representative of the northern Chinese dry-land millet agro-ecosystem (Sun et al. 2022 ). A mixed linear model (MLM) fitted across three successive seasons yielded a broad-sense heritability ( H ²) of 93.6% for plant height, indicating strong genetic control while retaining some environmental responsiveness. The resulting BLUE values thus provide a robust phenotypic foundation for high-resolution GWAS and reliable QTL localisation. Plant stature integrates lodging resistance, canopy light capture and synchronised grain filling (Dineshkumar et al. 1992 ; Tian et al. 2017 ); moderate dwarfism therefore enhances harvest index and facilitates mechanised harvesting while reducing labour and energy inputs (Pearce 2021 ). In the context of increasingly volatile climates, semi-dwarf, lodging-resistant foxtail millet cultivars are vital for sustaining food and feed production across arid and semi-arid zones(Choudhary et al. 2023 ; Yu et al. 2024 ). The continual expansion of genetic resources has likely enhanced the resolution with which plant height in foxtail millet can be dissected. Three material systems now drive progress. (i) Biparental populations. In a 124 plant Hongmiaozhangu × Changnong35 F₂ family, Wang et al. ( 2017 ) delimited the major locus qPH1.1 . He et al. ( 2021 ) later used 333 Ai 88 × Liaogu 1 RILs to uncover 26 QTL on nine chromosomes, underscoring the value of deep recombination for capturing both large- and moderate-effect loci. (ii) Diversity panels. Resequencing 916 globally sourced accessions, Jia et al. ( 2013 ) performed five-environment GWAS and pinpointed three stable height QTL, illustrating how abundant allelic diversity and rapid LD decay deliver high-resolution mapping. (iii) Functional mutants and multi-parent resources. The dwarf mutant Sidwarf2 of Yugu 1 was resolved by BSA-seq and fine mapping to a 52.7-kb window on chromosome 3, implicating a cytochrome P450 gene (Xue et al. 2016 ). Together, these resources create a seamless pipeline—from coarse mapping and fine localisation to causal validation and elite-allele assembly. Leveraging high-stability, single-site multi-year phenotypes from 209 diverse accessions, our GWAS identified environmentally stable height loci and deployable markers. The resulting genetic targets and rich allelic variation underpin semi-dwarf breeding for northern dry-land foxtail millet and provide a solid springboard for subsequent mutant validation and allele pyramiding. A stable chromosome-2 hotspot and additional loci not reported previously In earlier work, Gao et al. ( 2025 ) mapped a major QTL, qPH2-1 (8.10–28.52 Mb), on chromosome 2 in a 300 plant F₂ population from Jingu 28 × Ai 88, an interval that encompasses and extends beyond our qPH2.2 (24.64–24.84 Mb). A neighbouring segment on the same chromosome was likewise detected by He et al. ( 2021 ) in an Ai 88 × Liaogu 1 RIL population (26.04–28.91 Mb) and by Dai et al. ( 2024 ) in a 408 accession GWAS panel ( qPH02.13 , marker LDB_2_24921046) (Supplementary Table 5). Convergent evidence from F₂, RIL and natural populations thus pinpoints the 24–29 Mb region of chromosome 2 as a key hotspot controlling plant height in foxtail millet. By contrast, the additional loci we discovered— qPH2.1 , qPH4 , and qPH8 —have not appeared in previous RIL linkage maps, natural-panel GWAS (Jaiswal et al. 2019 ; Jia et al. 2013 ), or the 1,844-accession graph-genome GWAS of He et al. ( 2023 ), and therefore represent genuinely novel additions to the foxtail-millet height landscape. Among them, qPH4 yielded the lowest P -value and, together with qPH8 , was repeatedly detected in all three environments. qPH2.2 accounted for the largest proportion of phenotypic variance (17.12%), while qPH2.1 still explained 10.62%, underscoring the practical breeding value of these loci. Reliance on a 209 lines Dabaigu core set, phenotyped for three consecutive seasons at a single dry-land site, likely improved QTL resolution and reproducibility by combining spatial uniformity with inter-annual climatic variation. The new loci enlarge the current catalogue of foxtail-millet height QTL and offer fresh entry points for map-based cloning; the associated molecular markers may serve as useful tools for stacking semi-dwarf alleles, potentially underpinning lodging-resistant, high-harvest-index cultivars in arid and semi-arid regions. Future work will adopt the cross-genome collinearity framework of Sandhu et al. ( 2021 ) in rice Meta-QTL analysis, together with functional-mutant validation, to clarify the underlying biology and facilitate efficient aggregation of superior alleles. Candidate-gene landscape suggests GA-dependent and GA-independent contributions Early foxtail-millet height studies focused heavily on the GA pathway, He et al. ( 2021 ) highlighted Seita.1G242300 (GA2-oxidase-8) within qPH1.3 and proposed six additional GA-biosynthesis/signalling genes plus fifteen F-box genes across other QTL. Using RIL and F₂ data, Ni et al. ( 2017 ) mapped Z3ph1 , an sd1 / GA20ox homologue (89% identity), in bin2021 on chromosome 2, reinforcing GA20ox as a core regulator. More recently, Dai et al. ( 2024 ) identified WAK / Bph30 , the NAC factor OMTN3, phospholipase pPLAIII, and the IAA-glucosidase TGW6 in qPH03.14 and qPH06.10 , implicating cell-wall plasticity and auxin/cytokinin homeostasis. Within qPH2.1 we pinpointed SETIT_033071mg ( WAK4 ) and SETIT_029387mg (FAD-linked oxidase); the former guards wall integrity, the latter may modulate ROS-lignin balance (Kanneganti and Gupta 2008 ; Schmülling et al. 2003 ). qPH2.2 houses SETIT_030569mg , a chlorophyll a-b binding protein gene, suggesting that photosynthetic carbon flux can indirectly influence internode elongation (Green and Durnford 1996 ). qPH4 contains a bHLH transcription factor ( SETIT_007662mg ) (Toledo-Ortiz et al. 2003 ) and HVA22a ( SETIT_007340mg ) (Brands and Ho 2002 ) linked to hormone crosstalk and ER–vesicle homeostasis, while qPH8 features WNK5 (Manuka et al. 2015 ; Urano et al. 2012 ), PP2C (Leung et al. 1997 ; Umezawa et al. 2009 ), and RNA-binding/TPR proteins (Allan and Ratajczak 2011 ; Blatch and Lässle 1999 ), pointing to BR/ABA signalling and RNA metabolism. Except for the functional parallel between WAK4 and the WAK reported by Dai et al. ( 2024 ) these genes have not surfaced in earlier height studies, widening the regulatory horizon beyond GA. Taken together, our findings indicate that, while GA signalling is still central, plant stature in foxtail millet may also be influenced by additional mechanisms—including cell-wall remodelling, light-energy allocation, BR/ABA signalling, and aspects of RNA metabolism—warranting further functional verification. Declarations Author contribution statement WZ and HQ performed the data analyses and drafted the manuscript. ZQ, ZM and HW conceived and supervised the study, developed the 209-line population, and revised the manuscript. RH and SH managed the field trials and developed the KASP markers. JW, LC and XT collected the phenotypic data. All authors read and approved the final manuscript. Data, materials and code availability All raw resequencing genotype data and supplementary files generated in this study are available in Zenodo at https://doi.org/10.5281/zenodo.16778358 . Conflict of interest The authors declare that they have no relevant financial or non-financial interests to disclose. Funding This work was supported by the Major Special Science and Technology Project of Shanxi Province (202101140601027) and the National Natural Science Foundation of China (32241041). Acknowledgement We thank Dr. X Diao and Dr. Q He (Institute of Crop Sciences, Chinese Academy of Agricultural Sciences) for their valuable support and assistance with data processing. References Allan RK, Ratajczak T (2011) Versatile TPR domains accommodate different modes of target protein recognition and function. Cell Stress Chaperones 16:353–367 Ameen A, Raza S (2017) Green revolution: a review. Int J Adv Sci Res 3:129–137 Bennetzen JL, Schmutz J, Wang H, Percifield R, Hawkins J, Pontaroli AC, Estep M, Feng L, Vaughn JN, Grimwood J (2012) Reference genome sequence of the model plant Setaria. Nat Biotechnol 30:555–561 Blatch GL, Lässle M (1999) The tetratricopeptide repeat: a structural motif mediating protein-protein interactions. BioEssays 21:932–939 Brands A, Ho T-hD (2002) Function of a plant stress-induced gene, HVA22. Synthetic enhancement screen with its yeast homolog reveals its role in vesicular traffic. Plant Physiol 130:1121–1131 Choudhary P, Shukla P, Muthamilarasan M (2023) Genetic enhancement of climate-resilient traits in small millets: a review. Heliyon 9 Dai K, Wang X, Liu H, Qiao P, Wang J, Shi W, Guo J, Diao X (2024) Efficient identification of QTL for agronomic traits in foxtail millet (Setaria italica) using RTM-and MLM-GWAS. Theor Appl Genet 137:18 Das I, Rakshit S (2016) Millets, their importance, and production constraints. Biotic stress resistance in millets. Elsevier, pp 3–19 Diao X, Jia G (2016) Origin and domestication of foxtail millet. Genetics and genomics of Setaria. Springer, pp 61–72 Dineshkumar S, Shashidhar V, Ravikumar R, Seetharam A, Gowda B (1992) Indentification of true genetic dwarfing sources in foxtail millet (Setaria italica Beauv). Euphytica 60:207–212 Falcon WP, Naylor RL, Shankar ND (2022) Rethinking global food demand for 2050. Popul Dev Rev 48:921–957 Fan X, Tang S, Zhi H, He M, Ma W, Jia Y, Zhao B, Jia G, Diao X (2017) Identification and fine mapping of SiDWARF3 (D3), a pleiotropic locus controlling environment-independent dwarfism in foxtail millet. Crop Sci 57:2431–2442 Fernandez MGS, Becraft PW, Yin Y, Lübberstedt T (2009) From dwarves to giants? Plant height manipulation for biomass yield. Trends Plant Sci 14:454–461 Gao L, Zhu Q, Li H, Wang S, Fan J, Wang T, Yang L, Zhao Y, Ma Y, Chen L (2025) Construction of a genetic linkage map and QTL mapping of the agronomic traits in Foxtail millet (Setaria italica). BMC Genomics 26:152 Goron TL, Raizada MN (2015) Genetic diversity and genomic resources available for the small millet crops to accelerate a New Green Revolution. Front Plant Sci 6:157 Green BR, Durnford DG (1996) The chlorophyll-carotenoid proteins of oxygenic photosynthesis. Annu Rev Plant Biol 47:685–714 Griffiths S, Simmonds J, Leverington M, Wang Y, Fish L, Sayers L, Alibert L, Orford S, Wingen L, Snape J (2012) Meta-QTL analysis of the genetic control of crop height in elite European winter wheat germplasm. Mol Breeding 29:159–171 Han K, Wang Z, Shen L, Du X, Lian S, Li Y, Li Y, Tang C, Li H, Zhang L (2024) Mapping of dynamic quantitative trait loci for plant height in a RIL population of foxtail millet (Setaria italica L). Front Plant Sci 15:1418328 He Q, Tang S, Zhi H, Chen J, Zhang J, Liang H, Alam O, Li H, Zhang H, Xing L (2023) A graph-based genome and pan-genome variation of the model plant Setaria. Nat Genet 55:1232–1242 He Q, Zhi H, Tang S, Xing L, Wang S, Wang H, Zhang A, Li Y, Gao M, Zhang H (2021) QTL mapping for foxtail millet plant height in multi-environment using an ultra-high density bin map. Theor Appl Genet 134:557–572 Hedden P (2003) The genes of the Green Revolution. Trends Genet 19:5–9 Ikeda A, Ueguchi-Tanaka M, Sonoda Y, Kitano H, Koshioka M, Futsuhara Y, Matsuoka M, Yamaguchi J (2001) slender rice, a constitutive gibberellin response mutant, is caused by a null mutation of the SLR1 gene, an ortholog of the height-regulating gene GAI/RGA/RHT/D8. Plant Cell 13:999–1010 Itoh H, Tatsumi T, Sakamoto T, Otomo K, Toyomasu T, Kitano H, Ashikari M, Ichihara S, Matsuoka M (2004) A rice semi-dwarf gene, Tan-Ginbozu (D35), encodes the gibberellin biosynthesis enzyme, ent-kaurene oxidase. Plant Mol Biol 54:533–547 Itoh H, Ueguchi-Tanaka M, Sakamoto T, Kayano T, Tanaka H, Ashikari M, Matsuoka M (2002) Modification of rice plant height by suppressing the height-controlling gene, D18, in rice. Breed Sci 52:215–218 Jaiswal V, Gupta S, Gahlaut V, Muthamilarasan M, Bandyopadhyay T, Ramchiary N, Prasad M (2019) Genome-wide association study of major agronomic traits in foxtail millet (Setaria italica L.) using ddRAD sequencing. Sci Rep 9:5020 Jia G, Huang X, Zhi H, Zhao Y, Zhao Q, Li W, Chai Y, Yang L, Liu K, Lu H (2013) A haplotype map of genomic variations and genome-wide association studies of agronomic traits in foxtail millet (Setaria italica). Nat Genet 45:957–961 Kalita D, Taifa P, Nickhil C, Gogoi N (2025) Comparative nutritional analysis of foxtail millet and rice: a case for millet as a superior alternative. Discover Food 5:196 Kang HM, Sul JH, Service SK, Zaitlen NA, Kong S-y, Freimer NB, Sabatti C, Eskin E (2010) Variance component model to account for sample structure in genome-wide association studies. Nat Genet 42:348–354 Kanneganti V, Gupta AK (2008) Wall associated kinases from plants—an overview. Physiol Mol Biology Plants 14:109–118 Khound R, Rajput SG, Schnable JC, Vetriventhan M, Santra DK (2024) Genome-wide association study reveals marker–trait associations for major agronomic traits in proso millet (Panicum miliaceum L). Planta 260:44 Lawit SJ, Wych HM, Xu D, Kundu S, Tomes DT (2010) Maize DELLA proteins dwarf plant8 and dwarf plant9 as modulators of plant development. Plant Cell Physiol 51:1854–1868 Leung J, Merlot S, Giraudat J (1997) The Arabidopsis ABSCISIC ACID-INSENSITIVE2 (ABI2) and ABI1 genes encode homologous protein phosphatases 2C involved in abscisic acid signal transduction. Plant Cell 9:759–771 Liu X, Yang Y, Hou S, Men Y, Han Y (2022) The integration of genome-wide association study and homology analysis to explore the genomic regions and candidate genes for panicle-related traits in foxtail millet. Int J Mol Sci 23:14735 Manuka R, Saddhe AA, Kumar K (2015) Genome-wide identification and expression analysis of WNK kinase gene family in rice. Comput Biol Chem 59:56–66 Mauro-Herrera M, Doust AN (2016) Development and genetic control of plant architecture and biomass in the panicoid grass, Setaria. PLoS ONE 11:e0151346 Monna L, Kitazawa N, Yoshino R, Suzuki J, Masuda H, Maehara Y, Tanji M, Sato M, Nasu S, Minobe Y (2002) Positional cloning of rice semidwarfing gene, sd-1: rice green revolution gene encodes a mutant enzyme involved in gibberellin synthesis. DNA Res 9:11–17 Multani DS, Briggs SP, Chamberlin MA, Blakeslee JJ, Murphy AS, Johal GS (2003) Loss of an MDR transporter in compact stalks of maize br2 and sorghum dw3 mutants. Science 302:81–84 Ni X, Xia Q, Zhang H, Cheng S, Li H, Fan G, Guo T, Huang P, Xiang H, Chen Q (2017) Updated foxtail millet genome assembly and gene mapping of nine key agronomic traits by resequencing a RIL population. GigaScience 6:giw005 Panaud O (2006) Foxtail millet. Cereals and millets. Springer, pp 325–332 Pearce S (2021) Towards the replacement of wheat ‘Green Revolution’genes. J Exp Bot 72:157–160 Peng J, Richards DE, Hartley NM, Murphy GP, Devos KM, Flintham JE, Beales J, Fish LJ, Worland AJ, Pelica F (1999) Green revolution’genes encode mutant gibberellin response modulators. Nature 400:256–261 Sabouri A, Toorchi M, Rabiei B, Aharizad S, Moumeni A, Singh R (2010) Identification and mapping of QTLs for agronomic traits in indica—indica cross of rice (Oryza sativa L). Cereal Res Commun 38:317–326 Sandhu N, Pruthi G, Prakash Raigar O, Singh MP, Phagna K, Kumar A, Sethi M, Singh J, Ade PA, Saini DK (2021) Meta-QTL analysis in rice and cross-genome talk of the genomic regions controlling nitrogen use efficiency in cereal crops revealing phylogenetic relationship. Front Genet 12:807210 Schmülling T, Werner T, Riefler M, Krupková E, Bartrina y Manns I (2003) Structure and function of cytokinin oxidase/dehydrogenase genes of maize, rice, Arabidopsis and other species. J Plant Res 116:241–252 Sowadan O, Li D, Zhang Y, Zhu S, Hu X, Bhanbhro LB, Edzesi WM, Dang X, Hong D (2018) Mining of favorable alleles for lodging resistance traits in rice (Oryza sativa) through association mapping. Planta 248:155–169 Srividhya A, Vemireddy LR, Sridhar S, Jayaprada M, Ramanarao PV, Hariprasad AS, Reddy HK, Anuradha G, Siddiq E (2011) Molecular mapping of QTLs for yield and its components under two water supply conditions in rice (Oryza sativa L). J Crop Sci Biotechnol 14:45–56 Sun M, Guo M, Guo S, Li Y, Dong S, Song X, Shi X, Yuan X (2022) Effects of mesotrione on the control efficiency and chlorophyll fluorescence parameters of Chenopodium album under simulated rainfall conditions. PLoS ONE 17:e0267649 Tian B, Liu Y, Zhang L, Li H (2017) Stem lodging parameters of the basal three internodes associated with plant population densities and developmental stages in foxtail millet (Setaria italica) cultivars differing in resistance to lodging. Crop Pasture Sci 68:349–357 Toledo-Ortiz G, Huq E, Quail PH (2003) The Arabidopsis basic/helix-loop-helix transcription factor family. Plant Cell 15:1749–1770 Umezawa T, Sugiyama N, Mizoguchi M, Hayashi S, Myouga F, Yamaguchi-Shinozaki K, Ishihama Y, Hirayama T, Shinozaki K (2009) Type 2C protein phosphatases directly regulate abscisic acid-activated protein kinases in Arabidopsis. Proceedings of the National Academy of sciences 106:17588–17593 Urano D, Phan N, Jones JC, Yang J, Huang J, Grigston J, Philip Taylor J, Jones AM (2012) Endocytosis of the seven-transmembrane RGS1 protein activates G-protein-coupled signalling in Arabidopsis. Nat Cell Biol 14:1079–1088 Wang J, Wang Z, Du X, Yang H, Han F, Han Y, Yuan F, Zhang L, Peng S, Guo E (2017) A high-density genetic map and QTL analysis of agronomic traits in foxtail millet [Setaria italica (L.) P. Beauv.] using RAD-seq. PLoS ONE 12:e0179717 Würschum T, Langer SM, Longin CFH (2015) Genetic control of plant height in European winter wheat cultivars. Theor Appl Genet 128:865–874 Xue C, Zhi H, Fang X, Liu X, Tang S, Chai Y, Zhao B, Jia G, Diao X (2016) Characterization and fine mapping of SiDWARF2 (D2) in foxtail millet. Crop Sci 56:95–103 Yu J, Bai X, Zhang K, Feng L, Yu Z, Jiao X, Guo Y (2024) Assessment of Breeding Potential of Foxtail Millet Varieties Using a TOPSIS Model Constructed Based on Distinctness, Uniformity, and Stability Test Characteristics. Plants 13:2102 Zhang K, Fan G, Zhang X, Zhao F, Wei W, Du G, Feng X, Wang X, Wang F, Song G (2017) Identification of QTLs for 14 agronomically important traits in Setaria italica based on SNPs generated from high-throughput sequencing. G3: Genes, Genomes, Genetics 7:1587–1594 Zhao K, Tung C-W, Eizenga GC, Wright MH, Ali ML, Price AH, Norton GJ, Islam MR, Reynolds A, Mezey J (2011) Genome-wide association mapping reveals a rich genetic architecture of complex traits in Oryza sativa. Nat Commun 2:467 Zhao M, Zhi H, Zhang X, Jia G, Diao X (2019) Retrotransposon-mediated DELLA transcriptional reprograming underlies semi-dominant dwarfism in foxtail millet. Crop J 7:458–468 Zhu M, He Q, Lyu M, Shi T, Gao Q, Zhi H, Wang H, Jia G, Tang S, Cheng X (2023) Integrated genomic and transcriptomic analysis reveals genes associated with plant height of foxtail millet. Crop J 11:593–604 Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Major revisions 06 Jan, 2026 Reviewers agreed at journal 29 Sep, 2025 Reviewers invited by journal 28 Sep, 2025 Editor assigned by journal 12 Aug, 2025 First submitted to journal 09 Aug, 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-7333962","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":521874085,"identity":"3f8610cb-8180-418a-a38c-1c305f524bcb","order_by":0,"name":"Wei Zhou","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAuElEQVRIiWNgGAWjYFACxsYHCQZscmzs7QeI1cJ82OBBAZ8xH8+ZBGK1sKUJPvgglzhPwsGAOA3mEjlmDAkGZultEgwJDD8qthHWYtlzxgzol7TcNunGA4w9Z24T1mJwvMfcIMHgWG6bzIEEZsY2YrQc5jGTSDD4n84GJInUcrwtDaiYLYEELWcOHwY6jM2wDRjIB4nzy43Exoc//rDJy7e3H3zwo4IILSjgAInqR8EoGAWjYBTgAgAMBTyi+3EA2gAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0009-0006-6155-5562","institution":"Shanxi Agricultural University","correspondingAuthor":true,"prefix":"","firstName":"Wei","middleName":"","lastName":"Zhou","suffix":""},{"id":521874086,"identity":"87ce2e4e-8baf-4b10-824d-0b3515221bcc","order_by":1,"name":"Huibin Qin","email":"","orcid":"","institution":"Shanxi Agricultural University","correspondingAuthor":false,"prefix":"","firstName":"Huibin","middleName":"","lastName":"Qin","suffix":""},{"id":521874087,"identity":"49bfecbb-f05c-4702-8d24-d0791c9a7a66","order_by":2,"name":"Rui Huang","email":"","orcid":"","institution":"Shanxi Agricultural University","correspondingAuthor":false,"prefix":"","firstName":"Rui","middleName":"","lastName":"Huang","suffix":""},{"id":521874088,"identity":"e91d9b4d-7dd9-4e9f-823f-84b9406210af","order_by":3,"name":"Sen Hou","email":"","orcid":"","institution":"Shanxi Agricultural University","correspondingAuthor":false,"prefix":"","firstName":"Sen","middleName":"","lastName":"Hou","suffix":""},{"id":521874089,"identity":"0996815a-7f0c-46b9-a8e5-2e46a5bcb209","order_by":4,"name":"Junjie Wang","email":"","orcid":"","institution":"Shanxi Agricultural University","correspondingAuthor":false,"prefix":"","firstName":"Junjie","middleName":"","lastName":"Wang","suffix":""},{"id":521874090,"identity":"01141f67-d882-4c32-86ba-a7a3b1dda47f","order_by":5,"name":"Ling Chen","email":"","orcid":"","institution":"Shanxi Agricultural University","correspondingAuthor":false,"prefix":"","firstName":"Ling","middleName":"","lastName":"Chen","suffix":""},{"id":521874091,"identity":"aefc5226-d993-460c-b840-1a24783241f1","order_by":6,"name":"Xiang Tian","email":"","orcid":"","institution":"Shanxi Agricultural University","correspondingAuthor":false,"prefix":"","firstName":"Xiang","middleName":"","lastName":"Tian","suffix":""},{"id":521874092,"identity":"e4be0b00-808b-46db-bec5-944933ed377e","order_by":7,"name":"Zhijun Qiao","email":"","orcid":"","institution":"Shanxi Agricultural University","correspondingAuthor":false,"prefix":"","firstName":"Zhijun","middleName":"","lastName":"Qiao","suffix":""},{"id":521874093,"identity":"e083b43b-09d6-419f-a03d-1f242a859c68","order_by":8,"name":"Zhixin Mu","email":"","orcid":"","institution":"Shanxi Agricultural University","correspondingAuthor":false,"prefix":"","firstName":"Zhixin","middleName":"","lastName":"Mu","suffix":""},{"id":521874094,"identity":"cb977851-82ed-4057-96d7-fe75a388c371","order_by":9,"name":"Haigang Wang","email":"","orcid":"https://orcid.org/0009-0005-1731-8669","institution":"Center for Agricultural Genetic Resources Research/Key Laboratory of Crop Gene Resources and Germplasm Enhancement on the Loess Plateau, Ministry of Agriculture and Rural Affairs, Shanxi Agricultural University","correspondingAuthor":false,"prefix":"","firstName":"Haigang","middleName":"","lastName":"Wang","suffix":""}],"badges":[],"createdAt":"2025-08-09 12:48:38","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7333962/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7333962/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":93266648,"identity":"5ea5cdd8-39c1-45fa-b915-741a52e436be","added_by":"auto","created_at":"2025-10-10 20:17:03","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":11355,"visible":true,"origin":"","legend":"","description":"","filename":"taagTAAGD2500725.xml","url":"https://assets-eu.researchsquare.com/files/rs-7333962/v1/f818bccaa2445ddb2ad6d7e7.xml"},{"id":93266319,"identity":"52e3bfb3-84e5-401c-8efa-cc47622566c5","added_by":"auto","created_at":"2025-10-10 20:09:03","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":980,"visible":true,"origin":"","legend":"","description":"","filename":"TAAGD250072519007.go.xml","url":"https://assets-eu.researchsquare.com/files/rs-7333962/v1/34c0a010f129f9b4cb82226e.xml"},{"id":93266318,"identity":"45dae2ea-c748-460a-8d72-0b15e2346fc3","added_by":"auto","created_at":"2025-10-10 20:09:03","extension":"xml","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":816,"visible":true,"origin":"","legend":"","description":"","filename":"TAAGD2500725Import.xml","url":"https://assets-eu.researchsquare.com/files/rs-7333962/v1/0544af9ba99200cd05cc9c41.xml"},{"id":93266324,"identity":"31ad19ae-5f2a-45a9-92b0-e9e9e45b6af3","added_by":"auto","created_at":"2025-10-10 20:09:03","extension":"xml","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":148450,"visible":true,"origin":"","legend":"","description":"","filename":"TAAGD25007250enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7333962/v1/263f3dca7dfe4a3b8beff5f0.xml"},{"id":93266325,"identity":"adbf4109-bc44-4db2-8544-d88a434d2b2b","added_by":"auto","created_at":"2025-10-10 20:09:03","extension":"jpeg","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":510196,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7333962/v1/bd406b908de75723f5c373cd.jpeg"},{"id":93266328,"identity":"3aa56861-2c76-456e-8a54-0e4582862a7c","added_by":"auto","created_at":"2025-10-10 20:09:03","extension":"jpeg","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2627386,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7333962/v1/0c06352d025ec2f4b6326bec.jpeg"},{"id":93266330,"identity":"4212d68c-2947-4254-a2eb-2ed9aba23fed","added_by":"auto","created_at":"2025-10-10 20:09:03","extension":"jpeg","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":114352,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7333962/v1/aa246af7c537e600fe8dc2b2.jpeg"},{"id":93266331,"identity":"aaa4561f-7cdb-4a2e-b148-c409e4580f7c","added_by":"auto","created_at":"2025-10-10 20:09:03","extension":"jpeg","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":516648,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7333962/v1/e0eee193cca461506fabdaff.jpeg"},{"id":93266651,"identity":"5f537478-1b3b-4da4-b529-89b5cd2b773e","added_by":"auto","created_at":"2025-10-10 20:17:03","extension":"jpeg","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":27337072,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7333962/v1/0b5fe62890da80242eb2062e.jpeg"},{"id":93266332,"identity":"39b42fe4-f7ca-4144-9232-e604c50d0a77","added_by":"auto","created_at":"2025-10-10 20:09:03","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":60689,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7333962/v1/7eacdd58c2be217e07a752e5.png"},{"id":93266333,"identity":"6f46efa7-0400-45e0-9b73-27876bd6a956","added_by":"auto","created_at":"2025-10-10 20:09:04","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":286051,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7333962/v1/08bbb6362042845495902af5.png"},{"id":93266321,"identity":"b681195a-f1fc-4bb7-af94-1fa08906db8e","added_by":"auto","created_at":"2025-10-10 20:09:03","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":21770,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7333962/v1/2167bf2b1205b2c66c921ca4.png"},{"id":93266323,"identity":"19941a76-e653-4503-ac48-2d0d4bacd355","added_by":"auto","created_at":"2025-10-10 20:09:03","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":101058,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7333962/v1/e2fb0b922a71e5003c525c25.png"},{"id":93266322,"identity":"dd8661d4-4eb6-4c85-bdb2-bcb145a85616","added_by":"auto","created_at":"2025-10-10 20:09:03","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":194927,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7333962/v1/84fd8c0ca38dd4f7d707c8c9.png"},{"id":93266650,"identity":"49bd8590-44ee-4b8a-910d-8b65d732fac0","added_by":"auto","created_at":"2025-10-10 20:17:03","extension":"xml","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":146578,"visible":true,"origin":"","legend":"","description":"","filename":"TAAGD25007250structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7333962/v1/7c88918be1a0e59be8c9af6b.xml"},{"id":93266326,"identity":"a98c009b-c6e6-4d90-8a55-d34c9679ef84","added_by":"auto","created_at":"2025-10-10 20:09:03","extension":"html","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":157142,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7333962/v1/870606368b971df0b830c54d.html"},{"id":93266314,"identity":"484d079d-3b9d-40e9-a0f7-7bc620db8bbc","added_by":"auto","created_at":"2025-10-10 20:09:03","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":234116,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFrequency distribution and descriptive statistics of plant height across three environments.\u003c/strong\u003eA: Frequency distribution of PH in 209 foxtail millet accessions evaluated in three environments: 20JZ (Jinzhong, 2020), 21JZ (Jinzhong, 2021), and 22JZ (Jinzhong, 2022). Curves indicate kernel-density estimates. B: Statistical summary of PH data across the three environments. CV, coefficient of variation; \u003cem\u003eH\u003c/em\u003e², broad-sense heritability.\u003c/p\u003e","description":"","filename":"1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7333962/v1/79d25d302fd9ff8a7c80ff20.jpg"},{"id":93266313,"identity":"f3c209f6-acb0-466e-8f96-2c8f04a785e8","added_by":"auto","created_at":"2025-10-10 20:09:03","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1213615,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eGenome-wide distribution of SNPs in 209 foxtail millet.\u003c/strong\u003e A: SNP density distribution across the nine chromosomes of foxtail millet. The red lines represent SNP density along each chromosome. B: Proportion of SNPs distributed in different genomic regions, including intergenic, genic region, lncRNA, and pseudogene.\u003c/p\u003e","description":"","filename":"2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7333962/v1/71af7564364e135b40c0744f.jpg"},{"id":93266316,"identity":"3fcd133c-7a0f-4e51-9108-9563169dc2f1","added_by":"auto","created_at":"2025-10-10 20:09:03","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":144939,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eGenomic diversity and population structure of 209 Dabaigu foxtail millet accessions based on SNP data. \u003c/strong\u003eA: Neighbor-joining phylogenetic tree of the 209 foxtail millet accessions, with four major groups indicated by different colors. B: Principal component analysis (PCA) of the 209 accessions, showing the distribution of individuals along the first three principal components (PC1, PC2, PC3). Colors represent genetic clusters inferred from population structure analysis, and point shapes correspond to phylogenetic groups. C: Decay of linkage disequilibrium (LD, measured as r²) with physical distance (kb) for each subpopulation. D: Population structure inferred from ADMIXTURE analysis at K = 4. Each vertical bar represents an individual, and different colors indicate the proportion of ancestral components assigned to each cluster.\u003c/p\u003e","description":"","filename":"3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7333962/v1/5e69ce9baaa51dcd883aafe7.jpg"},{"id":93266317,"identity":"ad02ca34-3f6a-4ca1-a524-bd8625c6d030","added_by":"auto","created_at":"2025-10-10 20:09:03","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":159193,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eManhattan and quantile–quantile (Q-Q) plots of PH in foxtail millet. \u003c/strong\u003eThe Manhattan plot (left) displays the −log₁₀(P) values for marker–trait associations across all nine chromosomes. The horizontal red dashed line indicates the genome-wide significance threshold. (BLUE: Best Linear Unbiased Estimate). The Q-Q plot (right) compares the observed and expected −log₁₀(P) values for each dataset. Different colors represent GWAS results from each year or the BLUE values.\u003c/p\u003e","description":"","filename":"4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7333962/v1/0b9cbe4c59a2723c600c8e15.jpg"},{"id":93266649,"identity":"efa16c35-94ae-406f-951d-844c8d747389","added_by":"auto","created_at":"2025-10-10 20:17:03","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":252360,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eHaplotype analysis of \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eSETIT_033071mg\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e and evaluation of a KASP marker at qPH2.2. \u003c/strong\u003eA: Gene model and haplotype matrix derived from coding-region SNPs within \u003cem\u003eSETIT_033071mg\u003c/em\u003e. The REF row lists reference alleles and the ALT row lists alternative alleles; H1–H6 denote the six major haplotypes. Cells identical to the reference are highlighted in orange. The right-most column shows the number of accessions per haplotype. B: KASP clustering for the lead SNP at \u003cem\u003eqPH2.2\u003c/em\u003e (marker kPH8552) in the Dabaigu panel (n=209), resolving two homozygous classes (G:G and T:T). C: Violin plots of plant-height distributions for each haplotype across five field environments; boxes within violins indicate medians and inter-quartile ranges. Asterisks denote Wilcoxon rank-sum test significance (*\u003cem\u003eP\u003c/em\u003e\u0026lt; 0.05, **\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.01, ***\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.001). Environment codes: 18JZ, 19JZ, and 20JZ correspond to Jinzhong trials in 2018–2020; 20DT to Datong (2020); and 20YC to Yuncheng (2020).\u003c/p\u003e","description":"","filename":"5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7333962/v1/f1aaa1deac8b665d122bc434.jpg"},{"id":93266652,"identity":"6d54fa63-ff95-4c01-bfca-422f45ae401d","added_by":"auto","created_at":"2025-10-10 20:17:09","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3275811,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7333962/v1/d64b9f99-637c-44fa-9efb-99a370e61be4.pdf"}],"financialInterests":"","formattedTitle":"Genome-wide Association Analysis and Candidate Gene Identification of Plant Height in Dabaigu (Foxtail Millet)","fulltext":[{"header":"Key Message","content":"\u003cp\u003eMulti-environment GWAS reveals a dwarfing haplotype at \u003cem\u003eSETIT_033071mg\u003c/em\u003e and a small-effect marker at \u003cem\u003eqPH2.2\u003c/em\u003e, informing selection for lodging-resistant foxtail millet.\u003c/p\u003e"},{"header":"Introduction","content":"\u003cp\u003eThe mid-twentieth-century Green Revolution\u0026mdash;anchored by semi-dwarf, lodging-resistant wheat and rice cultivars and supported by irrigation and agrochemical inputs\u0026mdash;greatly increased cereal yields in the developing world (Ameen and Raza \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Meeting the food needs of a projected 9.7\u0026nbsp;billion people by 2050 will nevertheless require an additional 50\u0026ndash;60% rise in global production (Falcon et al. \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Plant height remains a pivotal breeding target: suitably reduced stature enhances lodging resistance, raises the harvest index, and enables fully mechanised harvesting, traits that underpinned previous yield breakthroughs and will be vital for future gains (Fernandez et al. \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Hedden \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2003\u003c/span\u003e). Therefore, identifying and utilizing height-regulating genes is pivotal for driving the next generation of productivity gains in cereal crops.\u003c/p\u003e\u003cp\u003eFoxtail millet (\u003cem\u003eSetaria italica\u003c/em\u003e) was domesticated from green foxtail in the Yellow River basin\u0026thinsp;~\u0026thinsp;16 000 YBP and, thanks to its small diploid genome, short life cycle and strict self-pollination, now serves both as a climate-resilient staple in arid regions and as a model for Panicoideae genomics (Bennetzen et al. \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Diao and Jia \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Goron and Raizada \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Panaud \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2006\u003c/span\u003e; Peng et al. \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e1999\u003c/span\u003e). Richer in protein, vitamins and minerals than rice or wheat and exceptionally tolerant of drought and poor soils (Das and Rakshit \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Kalita et al. \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), foxtail millet nonetheless suffers severe lodging at the grain-filling stage because dense planting meets intrinsically tall culms (Tian et al. \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Breeding shorter, lodging-resistant cultivars is therefore urgent, and systematically dissecting the genetic basis of plant height will provide the essential targets and tools for molecular improvement.\u003c/p\u003e\u003cp\u003eExtensive QTL dissection for plant height has been achieved in major cereals. Rice harbours more than 100 QTLs across its 12 chromosomes (Sabouri et al. \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Sandhu et al. \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Sowadan et al. \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Srividhya et al. \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Zhao et al. \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2011\u003c/span\u003e), with \u003cem\u003esd1\u003c/em\u003e, \u003cem\u003ed18\u003c/em\u003e and \u003cem\u003ed35\u003c/em\u003e conferring large, stable effects (Itoh et al. \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2004\u003c/span\u003e; Itoh et al. \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2002\u003c/span\u003e; Monna et al. \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2002\u003c/span\u003e). Wheat carries in excess of 50 height loci (Griffiths et al. \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; W\u0026uuml;rschum et al. \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2015\u003c/span\u003e), and both linkage mapping and GWAS repeatedly identify the \u003cem\u003eRht-B1\u003c/em\u003e/\u003cem\u003eRht-D1\u003c/em\u003e allelic series as the Green-Revolution foundation (Peng et al. \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e1999\u003c/span\u003e). In foxtail millet, studies remain recent and centered on F₂/RIL/natural-population panels, yielding over 100 mapped plant height QTLs across nine chromosomes; however, only 24 replicate consistently and many exhibit marked environment dependence (Fan et al. \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Han et al. \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; He et al. \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Mauro-Herrera and Doust \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Zhang et al. \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Zhu et al. \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eCloning of plant-height genes has revolutionized cereal breeding. In rice, \u003cem\u003esd1\u003c/em\u003e (GA20-oxidase)(Monna et al. \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2002\u003c/span\u003e) and \u003cem\u003eSLR1\u003c/em\u003e (DELLA)(Ikeda et al. \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2001\u003c/span\u003e), together with the wheat \u003cem\u003eRht-B1b/Rht-D1b\u003c/em\u003e allelic series(Peng et al. \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e1999\u003c/span\u003e), demonstrate that attenuating gibberellin (GA) signaling is a universal strategy for generating semi-dwarf phenotypes. In maize, \u003cem\u003eD8/D9\u003c/em\u003e(Lawit et al. \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2010\u003c/span\u003e), and in sorghum, \u003cem\u003eDw3\u003c/em\u003e(Multani et al. \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2003\u003c/span\u003e), further implicate additional pathways involving abscisic acid (ABA), auxin, and brassinosteroid/strigolactone signaling. In foxtail millet, however, only a handful of GA-related genes\u0026mdash;\u003cem\u003eSiSD1\u003c/em\u003e(Ni et al. \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), \u003cem\u003eSiGID1\u003c/em\u003e(Zhao et al. \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), and \u003cem\u003eSidwarf3\u003c/em\u003e (Fan et al. \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2017\u003c/span\u003e)\u0026mdash;have been characterized, mostly at the transcriptional level and without comprehensive mutant validation. Pathways related to cell-wall biosynthesis, cross-talk among GA, auxin, and brassinosteroids, or strigolactone signaling remain virtually unexplored. This GA-centric focus leaves other hormonal and structural mechanisms (e.g., cell-wall signaling) underexplored, creating a gap between locus discovery and breeding-ready targets.\u003c/p\u003e\u003cp\u003eHere, we conduct a multi-environment GWAS in a Dabaigu panel and, from ~\u0026thinsp;833k high-quality SNPs, identify four plant-height loci. Guided by cross-species annotation and shooting-stage stem expression in Yugu1, we prioritize \u003cem\u003eSETIT_033071mg\u003c/em\u003e (\u003cem\u003eWAK4\u003c/em\u003e) within \u003cem\u003eqPH2.1\u003c/em\u003e and perform a coding-SNP haplotype analysis in 313 accessions, highlighting an H2 haplotype associated with shorter stature. In parallel, we develop a KASP assay (kPH8552) for the lead SNP at \u003cem\u003eqPH2.2\u003c/em\u003e (distinct from the \u003cem\u003eSETIT_033071mg\u003c/em\u003e interval) as a candidate tool for marker-assisted improvement; its small but consistent effect (BLUE Δ = \u0026minus;4.48 cm; \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.044) motivates independent validation.\u003c/p\u003e"},{"header":"Materials and methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003ePlant materials and field trials\u003c/h2\u003e\u003cp\u003eA diversity panel of 209 foxtail millet (\u003cem\u003eSetaria italica\u003c/em\u003e) landraces (hereafter \u0026ldquo;Dabaigu\u0026rdquo;) was obtained from the Center for Agricultural Genetic Resources Research, Shanxi Agricultural University (Taiyuan, China). Trials were conducted for three consecutive years (2020\u0026ndash;2022) at Jinzhong, Shanxi, China (37.68\u0026deg;N, 112.75\u0026deg;E). Within each year, accessions were randomized and planted as a single 6-m row per accession with 30 cm row spacing; standard local management (irrigation, manual weeding and integrated pest control) was applied. At physiological maturity, plant height (ground to panicle tip) was measured on three randomly chosen plants per row and averaged for analyses.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003ePhenotypic analysis, BLUEs and broad-sense heritability\u003c/h3\u003e\n\u003cp\u003eSummary statistics (mean, SD, CV, range, skewness, kurtosis) were computed in R; histograms with overlaid density was drawn with ggplot2 (v3.5.1). Across-environment BLUEs were obtained using lme4 (v1.1-37) with the model:\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:\\text{P}\\text{H}\\:\\sim\\:\\:\\text{Env}\\:+\\:\\left(1\\:∣\\:\\text{Genotype}\\right)\\:+\\:\\left(1\\:∣\\:\\text{Genotype}:\\text{Env}\\right)$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003etreating Env as fixed and Genotype and Genotype\u0026times;Env as random; conditional genotype means (BLUEs) were extracted with emmeans.\u003c/p\u003e\u003cp\u003eVariance components were obtained from the same mixed model, and broad-sense heritability on an entry-mean basis was computed as:\u003cdiv id=\"Equb\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equb\" name=\"EquationSource\"\u003e\n$$\\:{H}^{2}=\\frac{{{\\sigma\\:}}_{g}^{2}}{{\\sigma\\:}_{g}^{2}+\\frac{{\\sigma\\:}_{ge}^{2}}{n}+\\frac{{{\\sigma\\:}}_{e}^{2}}{nr}}$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003ewith n\u0026thinsp;=\u0026thinsp;3 environments and r\u0026thinsp;=\u0026thinsp;1 row per environment.\u003c/p\u003e\u003cp\u003e\u003cb\u003eDNA extraction, library preparation and whole-genome resequencing\u003c/b\u003eLeaf\u003c/p\u003e\u003cp\u003eYoung leaves were collected at the shooting stage in 2022. Genomic DNA was extracted using a modified CTAB method, quality-checked by agarose electrophoresis and Qubit fluorometry (Thermo Fisher). Libraries (insert\u0026thinsp;~\u0026thinsp;500 bp) were prepared (end repair, A-tailing, adapter ligation), circularized, amplified as DNA nanoballs and sequenced on the BGI DNBSEQ\u0026trade; platform to generate 150-bp paired-end reads.\u003c/p\u003e\n\u003ch3\u003eRead processing, alignment and variant calling\u003c/h3\u003e\n\u003cp\u003eAdapters, reads with \u0026gt;\u0026thinsp;10% Ns, and reads with \u0026gt;\u0026thinsp;50% bases of Q\u0026thinsp;\u0026le;\u0026thinsp;12 were removed by SOAPnuke (v0.20.0). Clean reads were aligned to the Yugu1 reference genome (GCF_000263155.2, v2.0) using BWA-MEM (v0.7.17). SAM files were converted, sorted and deduplicated with SAMtools and Picard; only reads with MAPQ\u0026thinsp;\u0026ge;\u0026thinsp;30 were retained.\u003c/p\u003e\u003cp\u003eVariants were called per sample with GATK (v4.6.2) HaplotypeCaller (gVCF mode), combined with CombineGVCFs and jointly genotyped with GenotypeGVCFs. Hard filters followed GATK best practices (QD\u0026thinsp;\u0026ge;\u0026thinsp;2.0, FS\u0026thinsp;\u0026le;\u0026thinsp;60, SOR\u0026thinsp;\u0026le;\u0026thinsp;3.0, MQ\u0026thinsp;\u0026ge;\u0026thinsp;40, MQRankSum\u0026thinsp;\u0026ge;\u0026thinsp;\u0026minus;\u0026thinsp;12.5, ReadPosRankSum\u0026thinsp;\u0026ge;\u0026thinsp;\u0026minus;\u0026thinsp;8.0). SNPs were retained if biallelic, MAF\u0026thinsp;\u0026ge;\u0026thinsp;0.05, missing rate\u0026thinsp;\u0026le;\u0026thinsp;10%; HWE filtering was not applied given the selfing mating system and population structure.\u003c/p\u003e\n\u003ch3\u003ePopulation structure, PCA and linkage disequilibrium\u003c/h3\u003e\n\u003cp\u003eFor structure analyses, we used an LD-pruned SNP set (PLINK 1.9, --indep-pairwise 50 5 0.2). PCA was performed with PLINK; population structure was inferred by ADMIXTURE (v1.3.0) with K\u0026thinsp;=\u0026thinsp;2\u0026ndash;8, and the optimal K was chosen by CV error minimization.\u003c/p\u003e\u003cp\u003eGenome-wide LD decay was estimated by PopLDdecay (v3.43) as mean pairwise r\u0026sup2; vs physical distance; LD decay distance was defined as the first point where mean r\u0026sup2; dropped below 0.20.\u003c/p\u003e\n\u003ch3\u003eGWAS for Plant height\u003c/h3\u003e\n\u003cp\u003eGenome-wide association studies (GWAS) were conducted using EMMAX to identify SNPs significantly associated with plant height. A mixed linear model (MLM) was applied. The genome-wide significance threshold was set at \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;6.00 \u0026times; 10⁻⁸ (\u0026minus;\u0026thinsp;log₁₀(P)\u0026thinsp;\u0026gt;\u0026thinsp;7.22) based on Bonferroni correction (0.05/833,157). To avoid excessive false negatives resulting from this conservative threshold, SNPs with \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;1 \u0026times; 10⁻⁷ (\u0026minus;\u0026thinsp;log₁₀(P)\u0026thinsp;\u0026gt;\u0026thinsp;7) were also considered suggestively associated, following previous studies (Kang et al. \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Khound et al. \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). GWAS analyses were performed on four datasets (20JZ, 21JZ, 22JZ, and BLUE), and significant loci were determined by integrating these results with prior findings.\u003c/p\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003eCandidate gene identification and functional annotation\u003c/h2\u003e\u003cp\u003eCandidate genes within each significant QTL region were identified using the \u003cem\u003eYugu1\u003c/em\u003e reference genome (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.ncbi.nlm.nih.gov/datasets/genome/GCF_00026315\u003c/span\u003e\u003cspan address=\"https://www.ncbi.nlm.nih.gov/datasets/genome/GCF_00026315\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e 5.2/). Genes located in the defined intervals (including an additional 100-kb flanking region) were extracted and functionally annotated based on UniProt (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.uniprot.org/\u003c/span\u003e\u003cspan address=\"https://www.uniprot.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) and NCBI databases (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.ncbi.nlm.nih.gov/\u003c/span\u003e\u003cspan address=\"https://www.ncbi.nlm.nih.gov/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Furthermore, orthologous gene searches were performed against rice (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://rapdb.dna.affrc.go.jp/\u003c/span\u003e\u003cspan address=\"https://rapdb.dna.affrc.go.jp/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) and Arabidopsis (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.arabidopsis.org/\u003c/span\u003e\u003cspan address=\"https://www.arabidopsis.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) protein sequences using the online NCBI BLASTP tool (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://blast.ncbi.nlm.nih.gov/Bla\u003c/span\u003e\u003cspan address=\"https://blast.ncbi.nlm.nih.gov/Bla\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e st.cgi) to infer potential gene functions. Relevant literature was also consulted to support the functional prediction and prioritize candidate genes.\u003c/p\u003e\u003cp\u003e\u003cb\u003eTissue-Specific Expression Analysis and Haplotype Variation Analysis\u003c/b\u003e.\u003c/p\u003e\u003cp\u003eExpression profiles of candidate genes in stems, leaves, nodes, panicles, and seeds at different developmental stages (seeding, three-leaf, booting, shooting, flowering, and maturity) were retrieved from the Setaria-db database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.setariadb.com/millet\u003c/span\u003e\u003cspan address=\"http://www.setariadb.com/millet\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Heatmaps illustrating the expression patterns across organs and stages were generated using the R package pheatmap (v1.0.12).\u003c/p\u003e\u003cp\u003eHaplotype analysis of the \u003cem\u003eSETIT_033071mg\u003c/em\u003e gene was performed using resequencing data from 313 core foxtail millet accessions (unpublished). Plant height measurements were collected in Jinzhong (2018, 2019, 2020), Datong (2020), and Yuncheng (2020). The associations between haplotype variants and phenotypic traits were evaluated using the \u003cem\u003egeneHapR\u003c/em\u003e package.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eKASP marker development\u003c/h3\u003e\n\u003cp\u003eThe lead SNP at \u003cem\u003eqPH2.2\u003c/em\u003e was converted to a KASP assay (kPH8552) following Biosearch Technologies guidelines. Allele-specific primers were designed with PolyMarker (Supplementary Table\u0026nbsp;1); each forward primer carried the universal FAM (5\u0026prime;-GAAGGTGACCAAGTTCATGCT-3\u0026prime;) or HEX (5\u0026prime;-GAAGGTCGGAGTCAACGGATT-3\u0026prime;) tail. PCRs were run on a Roche LightCycler\u0026reg; 480 in 10 \u0026micro;L reactions (50\u0026ndash;100 ng DNA, 2\u0026times; KASP master mix, 0.14 \u0026micro;L primer cocktail) with touchdown cycling (94\u0026deg;C 15 min; 10 cycles 94\u0026deg;C 20 s, 61\u0026rarr;55\u0026deg;C \u0026minus;\u0026thinsp;0.6\u0026deg;C/cycle 60 s; then 26 cycles 94\u0026deg;C 20 s, 55\u0026deg;C 60 s; final 37\u0026deg;C 1 min).\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003ePhenotypic variations of plant height\u003c/h2\u003e\u003cp\u003ePH was measured in 209 foxtail millet accessions across three experimental environments (Supplementary Table\u0026nbsp;2). The PH phenotypes exhibited approximately normal distributions (Fig.\u0026nbsp;1A) with substantial variation (Fig.\u0026nbsp;1B). The mean PH ranged from 125.04 cm (22JZ) to 161.01 cm (21JZ), with overall phenotypic variation spanning 90.00\u0026ndash;194.42 cm. The coefficients of variation (CV) varied slightly across environments, ranging from 9.85% (21JZ, lowest variability) to 12.02% (20JZ, highest variability). Skewness and kurtosis values indicated that PH distributions were generally negatively skewed across all environments, with kurtosis values negative for 21JZ and 22JZ, while slightly positive for 20JZ. Plant height showed a high broad-sense heritability (\u003cem\u003eH\u003c/em\u003e\u0026sup2;) of 93.6%, indicating that genetic effects account for the vast majority of the observed phenotypic variation.\u003c/p\u003e\u003cp\u003e\u003cb\u003eFigure 1. Frequency distribution and descriptive statistics of plant height across three environments.\u003c/b\u003e A: Frequency distribution of PH in 209 foxtail millet accessions evaluated in three environments: 20JZ (Jinzhong, 2020), 21JZ (Jinzhong, 2021), and 22JZ (Jinzhong, 2022). Curves indicate kernel-density estimates. B: Statistical summary of PH data across the three environments. CV, coefficient of variation; \u003cem\u003eH\u003c/em\u003e\u0026sup2;, broad-sense heritability.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003eSequencing, SNP identification\u003c/h2\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eSummary of marker distribution among whole foxtail millet genome\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\u003eChromosome\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eLength (Bp)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eSNP numbers\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eSNPs/Mb\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e42,145,358\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e79,358\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1,882\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e49,199,692\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e96,374\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1,959\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e50,651,948\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e110,235\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e2,177\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e40,407,787\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e96,163\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e2,380\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e47,252,504\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e81,593\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1,726\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e36,014,412\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e85,545\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e2,375\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e35,964,117\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e65,385\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1,818\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e40,688,952\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e125,255\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e3,079\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e58,970,299\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e93,249\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1,581\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTotal\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e401,295,069\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e833,157\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e2,109\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 comprehensively characterize the genomic variation in foxtail millet, we conducted deep resequencing of 209 accessions, generating a total of 12.33\u0026nbsp;billion clean reads, with an average of 59\u0026nbsp;million reads per accession (Supplementary Table\u0026nbsp;3). The average GC content, Q20, and Q30 values were 45.2%, 97.3%, and 91.9%, respectively. More than 98% of reads were successfully mapped to the reference genome version 2 (Bennetzen et al. \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2012\u003c/span\u003e), with a coverage rate exceeding 93% and an average sequencing depth\u0026thinsp;\u0026asymp;\u0026thinsp;38\u0026times; (mean\u0026thinsp;=\u0026thinsp;37.6\u0026times;) across accessions.\u003c/p\u003e\u003cp\u003eAfter stringent filtering a total of 833,157 high-quality SNPs were retained (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). These SNPs were uniformly distributed across the genome, with an average density of 2,109 SNPs per megabase (Mb). Chromosome 8 exhibited the highest SNP density (3,079 SNPs/Mb), whereas chromosome 9 had the lowest (1,581 SNPs/Mb) (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eBased on functional annotation, the majority of SNPs (approximately 62.9%) were located in intergenic regions. Genic regions accounted for 33.6% of SNPs, while only 1.9% and 1.5% were found within pseudogenes and lncRNA regions, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e2\u003c/span\u003eB).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003ePhylogenetic relationships, population structure, principal component analysis, and linkage disequilibrium analysis\u003c/h2\u003e\u003cp\u003eAn NJ tree constructed from pairwise genetic distances resolved four groups (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e3\u003c/span\u003eA). Group sizes were uneven, with Group 2 the largest and Group 1 the smallest (Supplementary Table\u0026nbsp;4). Principal component analysis (PCA) recapitulated the same four-cluster pattern (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e3\u003c/span\u003eB), although sample assignments were not identical between methods, consistent with admixture and a complex genetic background.\u003c/p\u003e\u003cp\u003eFurther analysis using ADMIXTURE revealed that population boundaries were most distinct at K\u0026thinsp;=\u0026thinsp;4. At this optimal value, some accessions exhibited pure ancestry, while others displayed admixed genetic backgrounds, suggesting extensive gene flow among groups (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e3\u003c/span\u003eD). The cross-validation (CV) error sharply decreased from K\u0026thinsp;=\u0026thinsp;2 to K\u0026thinsp;=\u0026thinsp;4 and reached its lowest point at K\u0026thinsp;=\u0026thinsp;4, then gradually increased with higher K values (Fig. S1), supporting K\u0026thinsp;=\u0026thinsp;4 as the optimal number of subpopulations.\u003c/p\u003e\u003cp\u003eTo investigate linkage disequilibrium (LD) patterns, we analyzed LD decay across the entire panel and within each cluster. In the whole panel, LD declined to R\u0026sup2; = 0.2 at approximately 35 kb (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e3\u003c/span\u003eC). Each cluster exhibited distinct LD decay rates: Cluster 1 displayed the slowest decay, indicating a higher degree of genetic homogeneity, while Cluster 2 showed the fastest decay, suggesting more frequent recombination events and higher genetic diversity.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003eGWAS analysis\u003c/h2\u003e\u003cp\u003eUsing 833,157 high-quality SNPs, we conducted GWAS for plant height across three environments (20JZ, 21JZ, and 22JZ) and the BLUE values. In total, four stable QTL loci significantly associated with plant height were identified (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e4\u003c/span\u003e; Supplementary Table\u0026nbsp;5). On chromosome 8, four significant SNPs (chr8:910320, chr8:911022, chr8:911033, and chr8:911623) were located within a 1.3 kb interval, which is much smaller than the estimated LD decay distance (~\u0026thinsp;35 kb) in our panel. Therefore, these SNPs were merged into a single QTL locus (\u003cem\u003eqPH8\u003c/em\u003e), and chr8:911022, with the lowest P value, was considered as the lead SNP.\u003c/p\u003e\u003cp\u003eAmong the identified loci, two were located on chromosome 2 (\u003cem\u003eqPH2.1\u003c/em\u003e and \u003cem\u003eqPH2.2\u003c/em\u003e), while chromosomes 4 and 8 each harbored one locus (\u003cem\u003eqPH4\u003c/em\u003e and \u003cem\u003eqPH8\u003c/em\u003e, respectively). The strongest association signal was detected at \u003cem\u003eqPH4\u003c/em\u003e on chromosome 4, which had the lowest P value among all loci. The QTLs \u003cem\u003eqPH2.1\u003c/em\u003e, \u003cem\u003eqPH2.2\u003c/em\u003e, \u003cem\u003eqPH4\u003c/em\u003e, and \u003cem\u003eqPH8\u003c/em\u003e were detected in one, two, three, and three environments, respectively. The PVE ranged from 10.62% (\u003cem\u003eqPH2.1\u003c/em\u003e) to 17.12% (\u003cem\u003eqPH2.2\u003c/em\u003e), with \u003cem\u003eqPH2.2\u003c/em\u003e exhibiting the highest PVE (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe quantile\u0026ndash;quantile (QQ) plots indicated that the observed p-values generally conformed to the expected distribution under the null hypothesis, suggesting that the population structure and relatedness were well controlled in the GWAS model. Only the tail deviations represented true association signals (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e4\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\u003ch2\u003eCandidate gene\u003c/h2\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\u003eStable QTL loci and candidate genes identified in this study\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=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" 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=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eQTL\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eChr\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eQTL region (bp)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eQTN marker\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eP value\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eEnvironment\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003ePVE (%)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c8\"\u003e\u003cp\u003eCandidate gene\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c9\"\u003e\u003cp\u003eAnnotation\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e\u003cp\u003e\u003cem\u003eqPH2.1\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\" morerows=\"2\" rowspan=\"3\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\" morerows=\"2\" rowspan=\"3\"\u003e\u003cp\u003e5091820\u0026ndash;5291820\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\" morerows=\"2\" rowspan=\"3\"\u003e\u003cp\u003echr2:5191820\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\" morerows=\"2\" rowspan=\"3\"\u003e\u003cp\u003e7.734E-08\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\" morerows=\"2\" rowspan=\"3\"\u003e\u003cp\u003e20JZ\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\" morerows=\"2\" rowspan=\"3\"\u003e\u003cp\u003e10.62\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cem\u003eSETIT_033071mg\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003ewall-associated kinase 4\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cem\u003eSETIT_029387mg\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003eFAD-linked oxidases family protein\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cem\u003eSETIT_028865mg\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003eNB-ARC domain-containing disease resistance protein\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cem\u003eqPH2.2\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e24636879\u0026ndash;24836879\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003echr2:24736879\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e3.20E-08\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e20JZ/BLUE\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e17.12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cem\u003eSETIT_030569mg\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003eChlorophyll A-B binding family protein\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e\u003cem\u003eqPH4\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e8128049\u0026ndash;8328049\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003echr4:8228049\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e3.25E-09\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e20JZ/22JZ/BLUE\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e13.19\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cem\u003eSETIT_007662mg\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003ebasic helix-loop-helix (bHLH) DNA-binding superfamily protein\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cem\u003eSETIT_007340mg\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003eHVA22 homologue A\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e\u003cp\u003e\u003cem\u003eqPH8\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\" morerows=\"3\" rowspan=\"4\"\u003e\u003cp\u003e8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\" morerows=\"3\" rowspan=\"4\"\u003e\u003cp\u003e811022\u0026ndash;1011022\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\" morerows=\"3\" rowspan=\"4\"\u003e\u003cp\u003echr8:911022\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\" morerows=\"3\" rowspan=\"4\"\u003e\u003cp\u003e4.51E-09\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\" morerows=\"3\" rowspan=\"4\"\u003e\u003cp\u003e20JZ/22JZ/BLUE\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\" morerows=\"3\" rowspan=\"4\"\u003e\u003cp\u003e12.14\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cem\u003eSETIT_026086mg\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003ewith no lysine (K) kinase 5\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cem\u003eSETIT_026573mg\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003edsRNA-binding domain-like superfamily protein\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cem\u003eSETIT_027872mg\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003eTetratricopeptide repeat (TPR)-like superfamily protein\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cem\u003eSETIT_026360mg\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003eProtein phosphatase 2C family protein\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\u003eBased on the estimated LD decay (R\u0026sup2; = 0.2 at approximately 35 kb), a conservative\u0026thinsp;\u0026plusmn;\u0026thinsp;100 kb window around each significant GWAS locus was used to define candidate gene regions, as suggested in previous studies (Dai et al. \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Jia et al. \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Liu et al. \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Within these intervals we identified a total of 61 candidate genes (Supplementary Table\u0026nbsp;6).\u003c/p\u003e\u003cp\u003eThrough functional annotation of genes within these regions, several promising candidate genes potentially related to plant height regulation were identified (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Notably, in \u003cem\u003eqPH2.1\u003c/em\u003e on chromosome 2, we identified \u003cem\u003eSETIT_033071mg\u003c/em\u003e, encoding a wall-associated kinase (\u003cem\u003eWAK\u003c/em\u003e), which is involved in cell wall integrity and signal transduction, and \u003cem\u003eSETIT_029387mg\u003c/em\u003e, encoding an FAD-linked oxidases family protein. Additionally, \u003cem\u003eSETIT_028865mg\u003c/em\u003e, encoding an NB-ARC domain-containing disease resistance protein.\u003c/p\u003e\u003cp\u003eFor \u003cem\u003eqPH2.2\u003c/em\u003e, \u003cem\u003eSETIT_030569mg\u003c/em\u003e, encoding a chlorophyll A-B binding family protein. In qPH4, we identified \u003cem\u003eSETIT_007662mg\u003c/em\u003e, encoding a basic helix-loop-helix (bHLH) DNA-binding superfamily protein, as well as \u003cem\u003eSETIT_007340mg\u003c/em\u003e, encoding HVA22 homologue A. In \u003cem\u003eqPH8\u003c/em\u003e on chromosome 8, multiple candidate genes were detected, including \u003cem\u003eSETIT_026086mg\u003c/em\u003e, encoding a with-no-lysine (WNK) kinase 5 potentially; \u003cem\u003eSETIT_026573mg\u003c/em\u003e, encoding a dsRNA-binding domain-like superfamily protein; \u003cem\u003eSETIT_027872mg\u003c/em\u003e, encoding a tetratricopeptide repeat (TPR)-like superfamily protein; and \u003cem\u003eSETIT_026360mg\u003c/em\u003e, encoding a protein phosphatase 2C family protein.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\u003ch2\u003eTissue- and stage-specific expression profiles of candidate gene\u003c/h2\u003e\u003cp\u003eExpression profiling of ten plant-height candidate genes across six developmental stages (seedling, three-leaf, shooting, booting, flowering, and maturity) and six tissues (seed, root, stem, node, leaf, and panicle) revealed distinct spatiotemporal patterns (Fig. S2). Specifically, \u003cem\u003eSETIT_029387mg\u003c/em\u003e exhibited elevated expression in leaves and panicles from shooting to flowering stages. Genes \u003cem\u003eSETIT_026573mg\u003c/em\u003e, \u003cem\u003eSETIT_007662mg\u003c/em\u003e, \u003cem\u003eSETIT_026360mg\u003c/em\u003e, \u003cem\u003eSETIT_027872mg\u003c/em\u003e, and \u003cem\u003eSETIT_007340mg\u003c/em\u003e displayed high expression in panicles during booting and flowering stages. Additionally, \u003cem\u003eSETIT_028865mg\u003c/em\u003e and \u003cem\u003eSETIT_027872mg\u003c/em\u003e showed elevated expression in stems and nodes at the booting stage. Notably, \u003cem\u003eSETIT_033071mg\u003c/em\u003e exhibited high expression exclusively in stems at the shooting stage, while remaining lowly expressed across all other tissues and developmental stages.\u003c/p\u003e\u003cp\u003e\u003cb\u003eHaplotype and functional marker analysis of\u003c/b\u003e \u003cb\u003eSETIT_033071mg\u003c/b\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eBased on expression profiles and functional annotation within the \u003cem\u003eqPH2.1\u003c/em\u003e interval, \u003cem\u003eSETIT_033071mg\u003c/em\u003e (\u003cem\u003eWAK4\u003c/em\u003e) was prioritized for haplotype analysis. Twenty-eight high-confidence coding SNPs partitioned 313 accessions into six major haplotypes (H1\u0026ndash;H6). Across five field environments (18JZ, 19JZ, 20JZ, 20DT, 20YC), pairwise comparisons indicated that H2 tended to exhibit shorter plants than H1, H4 and H5, with significance in most environments (Wilcoxon tests, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05; Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e6\u003c/span\u003eC).\u003c/p\u003e\u003cp\u003eIn parallel, we designed a KASP assay (kPH8552) for the lead SNP at \u003cem\u003eqPH2.2\u003c/em\u003e. In the Dabaigu panel (n\u0026thinsp;=\u0026thinsp;209), the assay achieved an 80.4% call rate (G:G\u0026thinsp;=\u0026thinsp;74; T:T\u0026thinsp;=\u0026thinsp;94) and produced two well-separated homozygous clusters (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e6\u003c/span\u003eB). Based on across-environment BLUE estimates, the T allele was associated with a small but statistically significant reduction in plant height (\u0026minus;\u0026thinsp;4.48 cm; \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.044), with year-wise effects of \u0026minus;\u0026thinsp;4.15, \u0026minus;\u0026thinsp;1.13 and \u0026minus;\u0026thinsp;3.27 cm in 20JZ, 21JZ and 22JZ, respectively (Supplementary Table\u0026nbsp;7). These results indicate that kPH8552 tags a small-effect locus at \u003cem\u003eqPH2.2\u003c/em\u003e, with a consistent direction of effect across environments\u003c/p\u003e\u003c/div\u003e"},{"header":"Discussion","content":"\u003cdiv id=\"Sec18\" class=\"Section2\"\u003e\u003ch2\u003eWell-powered GWAS enabled by multi-year phenotyping of a diverse panel\u003c/h2\u003e\u003cp\u003eThe present study assessed plant height in a genetically diverse panel of 209 Dabaigu foxtail millet accessions sourced from contrasting eco-regions of northern China and grown for three successive seasons (2020\u0026ndash;2022) under standardised field conditions in Jinzhong. Jinzhong, experiences a typical temperate continental monsoon climate and is representative of the northern Chinese dry-land millet agro-ecosystem (Sun et al. \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). A mixed linear model (MLM) fitted across three successive seasons yielded a broad-sense heritability (\u003cem\u003eH\u003c/em\u003e\u0026sup2;) of 93.6% for plant height, indicating strong genetic control while retaining some environmental responsiveness. The resulting BLUE values thus provide a robust phenotypic foundation for high-resolution GWAS and reliable QTL localisation.\u003c/p\u003e\u003cp\u003ePlant stature integrates lodging resistance, canopy light capture and synchronised grain filling (Dineshkumar et al. \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e1992\u003c/span\u003e; Tian et al. \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2017\u003c/span\u003e); moderate dwarfism therefore enhances harvest index and facilitates mechanised harvesting while reducing labour and energy inputs (Pearce \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). In the context of increasingly volatile climates, semi-dwarf, lodging-resistant foxtail millet cultivars are vital for sustaining food and feed production across arid and semi-arid zones(Choudhary et al. \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Yu et al. \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe continual expansion of genetic resources has likely enhanced the resolution with which plant height in foxtail millet can be dissected. Three material systems now drive progress. (i) Biparental populations. In a 124 plant Hongmiaozhangu \u003cem\u003e\u0026times;\u003c/em\u003e Changnong35 F₂ family, Wang et al. (\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) delimited the major locus \u003cem\u003eqPH1.1\u003c/em\u003e. He et al. (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) later used 333 Ai 88 \u0026times; Liaogu 1 RILs to uncover 26 QTL on nine chromosomes, underscoring the value of deep recombination for capturing both large- and moderate-effect loci. (ii) Diversity panels. Resequencing 916 globally sourced accessions, Jia et al. (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2013\u003c/span\u003e) performed five-environment GWAS and pinpointed three stable height QTL, illustrating how abundant allelic diversity and rapid LD decay deliver high-resolution mapping. (iii) Functional mutants and multi-parent resources. The dwarf mutant \u003cem\u003eSidwarf2\u003c/em\u003e of Yugu 1 was resolved by BSA-seq and fine mapping to a 52.7-kb window on chromosome 3, implicating a cytochrome P450 gene (Xue et al. \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2016\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eTogether, these resources create a seamless pipeline\u0026mdash;from coarse mapping and fine localisation to causal validation and elite-allele assembly. Leveraging high-stability, single-site multi-year phenotypes from 209 diverse accessions, our GWAS identified environmentally stable height loci and deployable markers. The resulting genetic targets and rich allelic variation underpin semi-dwarf breeding for northern dry-land foxtail millet and provide a solid springboard for subsequent mutant validation and allele pyramiding.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec19\" class=\"Section2\"\u003e\u003ch2\u003eA stable chromosome-2 hotspot and additional loci not reported previously\u003c/h2\u003e\u003cp\u003eIn earlier work, Gao et al. (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) mapped a major QTL, \u003cem\u003eqPH2-1\u003c/em\u003e (8.10\u0026ndash;28.52 Mb), on chromosome 2 in a 300 plant F₂ population from Jingu 28 \u0026times; Ai 88, an interval that encompasses and extends beyond our \u003cem\u003eqPH2.2\u003c/em\u003e (24.64\u0026ndash;24.84 Mb). A neighbouring segment on the same chromosome was likewise detected by He et al. (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) in an Ai 88 \u0026times; Liaogu 1 RIL population (26.04\u0026ndash;28.91 Mb) and by Dai et al. (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) in a 408 accession GWAS panel (\u003cem\u003eqPH02.13\u003c/em\u003e, marker LDB_2_24921046) (Supplementary Table\u0026nbsp;5). Convergent evidence from F₂, RIL and natural populations thus pinpoints the 24\u0026ndash;29 Mb region of chromosome 2 as a key hotspot controlling plant height in foxtail millet.\u003c/p\u003e\u003cp\u003eBy contrast, the additional loci we discovered\u0026mdash;\u003cem\u003eqPH2.1\u003c/em\u003e, \u003cem\u003eqPH4\u003c/em\u003e, and \u003cem\u003eqPH8\u003c/em\u003e\u0026mdash;have not appeared in previous RIL linkage maps, natural-panel GWAS (Jaiswal et al. \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Jia et al. \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2013\u003c/span\u003e), or the 1,844-accession graph-genome GWAS of He et al. (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), and therefore represent genuinely novel additions to the foxtail-millet height landscape. Among them, \u003cem\u003eqPH4\u003c/em\u003e yielded the lowest \u003cem\u003eP\u003c/em\u003e-value and, together with \u003cem\u003eqPH8\u003c/em\u003e, was repeatedly detected in all three environments. \u003cem\u003eqPH2.2\u003c/em\u003e accounted for the largest proportion of phenotypic variance (17.12%), while \u003cem\u003eqPH2.1\u003c/em\u003e still explained 10.62%, underscoring the practical breeding value of these loci.\u003c/p\u003e\u003cp\u003eReliance on a 209 lines Dabaigu core set, phenotyped for three consecutive seasons at a single dry-land site, likely improved QTL resolution and reproducibility by combining spatial uniformity with inter-annual climatic variation. The new loci enlarge the current catalogue of foxtail-millet height QTL and offer fresh entry points for map-based cloning; the associated molecular markers may serve as useful tools for stacking semi-dwarf alleles, potentially underpinning lodging-resistant, high-harvest-index cultivars in arid and semi-arid regions. Future work will adopt the cross-genome collinearity framework of Sandhu et al. (\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) in rice Meta-QTL analysis, together with functional-mutant validation, to clarify the underlying biology and facilitate efficient aggregation of superior alleles.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec20\" class=\"Section2\"\u003e\u003ch2\u003eCandidate-gene landscape suggests GA-dependent and GA-independent contributions\u003c/h2\u003e\u003cp\u003eEarly foxtail-millet height studies focused heavily on the GA pathway, He et al. (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) highlighted \u003cem\u003eSeita.1G242300\u003c/em\u003e (GA2-oxidase-8) within \u003cem\u003eqPH1.3\u003c/em\u003e and proposed six additional GA-biosynthesis/signalling genes plus fifteen F-box genes across other QTL. Using RIL and F₂ data, Ni et al. (\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) mapped \u003cem\u003eZ3ph1\u003c/em\u003e, an \u003cem\u003esd1\u003c/em\u003e/\u003cem\u003eGA20ox\u003c/em\u003e homologue (89% identity), in bin2021 on chromosome 2, reinforcing GA20ox as a core regulator. More recently, Dai et al. (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) identified \u003cem\u003eWAK\u003c/em\u003e/\u003cem\u003eBph30\u003c/em\u003e, the NAC factor OMTN3, phospholipase pPLAIII, and the IAA-glucosidase TGW6 in \u003cem\u003eqPH03.14\u003c/em\u003e and \u003cem\u003eqPH06.10\u003c/em\u003e, implicating cell-wall plasticity and auxin/cytokinin homeostasis.\u003c/p\u003e\u003cp\u003eWithin \u003cem\u003eqPH2.1\u003c/em\u003e we pinpointed \u003cem\u003eSETIT_033071mg\u003c/em\u003e (\u003cem\u003eWAK4\u003c/em\u003e) and \u003cem\u003eSETIT_029387mg\u003c/em\u003e (FAD-linked oxidase); the former guards wall integrity, the latter may modulate ROS-lignin balance (Kanneganti and Gupta \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Schm\u0026uuml;lling et al. \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2003\u003c/span\u003e). \u003cem\u003eqPH2.2\u003c/em\u003e houses \u003cem\u003eSETIT_030569mg\u003c/em\u003e, a chlorophyll a-b binding protein gene, suggesting that photosynthetic carbon flux can indirectly influence internode elongation (Green and Durnford \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e1996\u003c/span\u003e). \u003cem\u003eqPH4\u003c/em\u003e contains a bHLH transcription factor (\u003cem\u003eSETIT_007662mg\u003c/em\u003e) (Toledo-Ortiz et al. \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2003\u003c/span\u003e) and \u003cem\u003eHVA22a\u003c/em\u003e (\u003cem\u003eSETIT_007340mg\u003c/em\u003e) (Brands and Ho \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2002\u003c/span\u003e) linked to hormone crosstalk and ER\u0026ndash;vesicle homeostasis, while \u003cem\u003eqPH8\u003c/em\u003e features \u003cem\u003eWNK5\u003c/em\u003e (Manuka et al. \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Urano et al. \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2012\u003c/span\u003e), \u003cem\u003ePP2C\u003c/em\u003e (Leung et al. \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e1997\u003c/span\u003e; Umezawa et al. \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2009\u003c/span\u003e), and RNA-binding/TPR proteins (Allan and Ratajczak \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Blatch and L\u0026auml;ssle \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e1999\u003c/span\u003e), pointing to BR/ABA signalling and RNA metabolism. Except for the functional parallel between \u003cem\u003eWAK4\u003c/em\u003e and the \u003cem\u003eWAK\u003c/em\u003e reported by Dai et al. (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) these genes have not surfaced in earlier height studies, widening the regulatory horizon beyond GA.\u003c/p\u003e\u003cp\u003eTaken together, our findings indicate that, while GA signalling is still central, plant stature in foxtail millet may also be influenced by additional mechanisms\u0026mdash;including cell-wall remodelling, light-energy allocation, BR/ABA signalling, and aspects of RNA metabolism\u0026mdash;warranting further functional verification.\u003c/p\u003e\u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003ch2\u003eAuthor contribution statement\u003c/h2\u003e\u003cp\u003eWZ and HQ performed the data analyses and drafted the manuscript. ZQ, ZM and HW conceived and supervised the study, developed the 209-line population, and revised the manuscript. RH and SH managed the field trials and developed the KASP markers. JW, LC and XT collected the phenotypic data. All authors read and approved the final manuscript.\u003c/p\u003e\u003c/p\u003e\u003cp\u003e\u003ch2\u003eData, materials and code availability\u003c/h2\u003e\u003cp\u003eAll raw resequencing genotype data and supplementary files generated in this study are available in Zenodo at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5281/zenodo.16778358\u003c/span\u003e\u003cspan address=\"10.5281/zenodo.16778358\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/p\u003e\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eConflict of interest\u003c/strong\u003e\u003cp\u003eThe authors declare that they have no relevant financial or non-financial interests to disclose.\u003c/p\u003e\u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e\u003cp\u003eThis work was supported by the Major Special Science and Technology Project of Shanxi Province (202101140601027) and the National Natural Science Foundation of China (32241041).\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eWe thank Dr. X Diao and Dr. Q He (Institute of Crop Sciences, Chinese Academy of Agricultural Sciences) for their valuable support and assistance with data processing.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAllan RK, Ratajczak T (2011) Versatile TPR domains accommodate different modes of target protein recognition and function. Cell Stress Chaperones 16:353\u0026ndash;367\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAmeen A, Raza S (2017) Green revolution: a review. Int J Adv Sci Res 3:129\u0026ndash;137\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBennetzen JL, Schmutz J, Wang H, Percifield R, Hawkins J, Pontaroli AC, Estep M, Feng L, Vaughn JN, Grimwood J (2012) Reference genome sequence of the model plant Setaria. Nat Biotechnol 30:555\u0026ndash;561\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBlatch GL, L\u0026auml;ssle M (1999) The tetratricopeptide repeat: a structural motif mediating protein-protein interactions. BioEssays 21:932\u0026ndash;939\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBrands A, Ho T-hD (2002) Function of a plant stress-induced gene, HVA22. Synthetic enhancement screen with its yeast homolog reveals its role in vesicular traffic. Plant Physiol 130:1121\u0026ndash;1131\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChoudhary P, Shukla P, Muthamilarasan M (2023) Genetic enhancement of climate-resilient traits in small millets: a review. Heliyon 9\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDai K, Wang X, Liu H, Qiao P, Wang J, Shi W, Guo J, Diao X (2024) Efficient identification of QTL for agronomic traits in foxtail millet (Setaria italica) using RTM-and MLM-GWAS. Theor Appl Genet 137:18\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDas I, Rakshit S (2016) Millets, their importance, and production constraints. Biotic stress resistance in millets. Elsevier, pp 3\u0026ndash;19\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDiao X, Jia G (2016) Origin and domestication of foxtail millet. Genetics and genomics of Setaria. Springer, pp 61\u0026ndash;72\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDineshkumar S, Shashidhar V, Ravikumar R, Seetharam A, Gowda B (1992) Indentification of true genetic dwarfing sources in foxtail millet (Setaria italica Beauv). Euphytica 60:207\u0026ndash;212\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFalcon WP, Naylor RL, Shankar ND (2022) Rethinking global food demand for 2050. Popul Dev Rev 48:921\u0026ndash;957\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFan X, Tang S, Zhi H, He M, Ma W, Jia Y, Zhao B, Jia G, Diao X (2017) Identification and fine mapping of SiDWARF3 (D3), a pleiotropic locus controlling environment-independent dwarfism in foxtail millet. Crop Sci 57:2431\u0026ndash;2442\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFernandez MGS, Becraft PW, Yin Y, L\u0026uuml;bberstedt T (2009) From dwarves to giants? Plant height manipulation for biomass yield. Trends Plant Sci 14:454\u0026ndash;461\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGao L, Zhu Q, Li H, Wang S, Fan J, Wang T, Yang L, Zhao Y, Ma Y, Chen L (2025) Construction of a genetic linkage map and QTL mapping of the agronomic traits in Foxtail millet (Setaria italica). BMC Genomics 26:152\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGoron TL, Raizada MN (2015) Genetic diversity and genomic resources available for the small millet crops to accelerate a New Green Revolution. Front Plant Sci 6:157\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGreen BR, Durnford DG (1996) The chlorophyll-carotenoid proteins of oxygenic photosynthesis. Annu Rev Plant Biol 47:685\u0026ndash;714\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGriffiths S, Simmonds J, Leverington M, Wang Y, Fish L, Sayers L, Alibert L, Orford S, Wingen L, Snape J (2012) Meta-QTL analysis of the genetic control of crop height in elite European winter wheat germplasm. Mol Breeding 29:159\u0026ndash;171\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHan K, Wang Z, Shen L, Du X, Lian S, Li Y, Li Y, Tang C, Li H, Zhang L (2024) Mapping of dynamic quantitative trait loci for plant height in a RIL population of foxtail millet (Setaria italica L). Front Plant Sci 15:1418328\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHe Q, Tang S, Zhi H, Chen J, Zhang J, Liang H, Alam O, Li H, Zhang H, Xing L (2023) A graph-based genome and pan-genome variation of the model plant Setaria. Nat Genet 55:1232\u0026ndash;1242\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHe Q, Zhi H, Tang S, Xing L, Wang S, Wang H, Zhang A, Li Y, Gao M, Zhang H (2021) QTL mapping for foxtail millet plant height in multi-environment using an ultra-high density bin map. Theor Appl Genet 134:557\u0026ndash;572\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHedden P (2003) The genes of the Green Revolution. Trends Genet 19:5\u0026ndash;9\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eIkeda A, Ueguchi-Tanaka M, Sonoda Y, Kitano H, Koshioka M, Futsuhara Y, Matsuoka M, Yamaguchi J (2001) slender rice, a constitutive gibberellin response mutant, is caused by a null mutation of the SLR1 gene, an ortholog of the height-regulating gene GAI/RGA/RHT/D8. Plant Cell 13:999\u0026ndash;1010\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eItoh H, Tatsumi T, Sakamoto T, Otomo K, Toyomasu T, Kitano H, Ashikari M, Ichihara S, Matsuoka M (2004) A rice semi-dwarf gene, Tan-Ginbozu (D35), encodes the gibberellin biosynthesis enzyme, ent-kaurene oxidase. Plant Mol Biol 54:533\u0026ndash;547\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eItoh H, Ueguchi-Tanaka M, Sakamoto T, Kayano T, Tanaka H, Ashikari M, Matsuoka M (2002) Modification of rice plant height by suppressing the height-controlling gene, D18, in rice. Breed Sci 52:215\u0026ndash;218\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJaiswal V, Gupta S, Gahlaut V, Muthamilarasan M, Bandyopadhyay T, Ramchiary N, Prasad M (2019) Genome-wide association study of major agronomic traits in foxtail millet (Setaria italica L.) using ddRAD sequencing. Sci Rep 9:5020\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJia G, Huang X, Zhi H, Zhao Y, Zhao Q, Li W, Chai Y, Yang L, Liu K, Lu H (2013) A haplotype map of genomic variations and genome-wide association studies of agronomic traits in foxtail millet (Setaria italica). Nat Genet 45:957\u0026ndash;961\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKalita D, Taifa P, Nickhil C, Gogoi N (2025) Comparative nutritional analysis of foxtail millet and rice: a case for millet as a superior alternative. Discover Food 5:196\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKang HM, Sul JH, Service SK, Zaitlen NA, Kong S-y, Freimer NB, Sabatti C, Eskin E (2010) Variance component model to account for sample structure in genome-wide association studies. Nat Genet 42:348\u0026ndash;354\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKanneganti V, Gupta AK (2008) Wall associated kinases from plants\u0026mdash;an overview. Physiol Mol Biology Plants 14:109\u0026ndash;118\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKhound R, Rajput SG, Schnable JC, Vetriventhan M, Santra DK (2024) Genome-wide association study reveals marker\u0026ndash;trait associations for major agronomic traits in proso millet (Panicum miliaceum L). Planta 260:44\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLawit SJ, Wych HM, Xu D, Kundu S, Tomes DT (2010) Maize DELLA proteins dwarf plant8 and dwarf plant9 as modulators of plant development. Plant Cell Physiol 51:1854\u0026ndash;1868\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLeung J, Merlot S, Giraudat J (1997) The Arabidopsis ABSCISIC ACID-INSENSITIVE2 (ABI2) and ABI1 genes encode homologous protein phosphatases 2C involved in abscisic acid signal transduction. Plant Cell 9:759\u0026ndash;771\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLiu X, Yang Y, Hou S, Men Y, Han Y (2022) The integration of genome-wide association study and homology analysis to explore the genomic regions and candidate genes for panicle-related traits in foxtail millet. Int J Mol Sci 23:14735\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eManuka R, Saddhe AA, Kumar K (2015) Genome-wide identification and expression analysis of WNK kinase gene family in rice. Comput Biol Chem 59:56\u0026ndash;66\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMauro-Herrera M, Doust AN (2016) Development and genetic control of plant architecture and biomass in the panicoid grass, Setaria. PLoS ONE 11:e0151346\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMonna L, Kitazawa N, Yoshino R, Suzuki J, Masuda H, Maehara Y, Tanji M, Sato M, Nasu S, Minobe Y (2002) Positional cloning of rice semidwarfing gene, sd-1: rice green revolution gene encodes a mutant enzyme involved in gibberellin synthesis. DNA Res 9:11\u0026ndash;17\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMultani DS, Briggs SP, Chamberlin MA, Blakeslee JJ, Murphy AS, Johal GS (2003) Loss of an MDR transporter in compact stalks of maize br2 and sorghum dw3 mutants. Science 302:81\u0026ndash;84\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNi X, Xia Q, Zhang H, Cheng S, Li H, Fan G, Guo T, Huang P, Xiang H, Chen Q (2017) Updated foxtail millet genome assembly and gene mapping of nine key agronomic traits by resequencing a RIL population. GigaScience 6:giw005\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePanaud O (2006) Foxtail millet. Cereals and millets. Springer, pp 325\u0026ndash;332\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePearce S (2021) Towards the replacement of wheat \u0026lsquo;Green Revolution\u0026rsquo;genes. J Exp Bot 72:157\u0026ndash;160\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePeng J, Richards DE, Hartley NM, Murphy GP, Devos KM, Flintham JE, Beales J, Fish LJ, Worland AJ, Pelica F (1999) Green revolution\u0026rsquo;genes encode mutant gibberellin response modulators. Nature 400:256\u0026ndash;261\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSabouri A, Toorchi M, Rabiei B, Aharizad S, Moumeni A, Singh R (2010) Identification and mapping of QTLs for agronomic traits in indica\u0026mdash;indica cross of rice (Oryza sativa L). Cereal Res Commun 38:317\u0026ndash;326\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSandhu N, Pruthi G, Prakash Raigar O, Singh MP, Phagna K, Kumar A, Sethi M, Singh J, Ade PA, Saini DK (2021) Meta-QTL analysis in rice and cross-genome talk of the genomic regions controlling nitrogen use efficiency in cereal crops revealing phylogenetic relationship. Front Genet 12:807210\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSchm\u0026uuml;lling T, Werner T, Riefler M, Krupkov\u0026aacute; E, Bartrina y Manns I (2003) Structure and function of cytokinin oxidase/dehydrogenase genes of maize, rice, Arabidopsis and other species. J Plant Res 116:241\u0026ndash;252\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSowadan O, Li D, Zhang Y, Zhu S, Hu X, Bhanbhro LB, Edzesi WM, Dang X, Hong D (2018) Mining of favorable alleles for lodging resistance traits in rice (Oryza sativa) through association mapping. Planta 248:155\u0026ndash;169\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSrividhya A, Vemireddy LR, Sridhar S, Jayaprada M, Ramanarao PV, Hariprasad AS, Reddy HK, Anuradha G, Siddiq E (2011) Molecular mapping of QTLs for yield and its components under two water supply conditions in rice (Oryza sativa L). J Crop Sci Biotechnol 14:45\u0026ndash;56\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSun M, Guo M, Guo S, Li Y, Dong S, Song X, Shi X, Yuan X (2022) Effects of mesotrione on the control efficiency and chlorophyll fluorescence parameters of Chenopodium album under simulated rainfall conditions. PLoS ONE 17:e0267649\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTian B, Liu Y, Zhang L, Li H (2017) Stem lodging parameters of the basal three internodes associated with plant population densities and developmental stages in foxtail millet (Setaria italica) cultivars differing in resistance to lodging. Crop Pasture Sci 68:349\u0026ndash;357\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eToledo-Ortiz G, Huq E, Quail PH (2003) The Arabidopsis basic/helix-loop-helix transcription factor family. Plant Cell 15:1749\u0026ndash;1770\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eUmezawa T, Sugiyama N, Mizoguchi M, Hayashi S, Myouga F, Yamaguchi-Shinozaki K, Ishihama Y, Hirayama T, Shinozaki K (2009) Type 2C protein phosphatases directly regulate abscisic acid-activated protein kinases in Arabidopsis. Proceedings of the National Academy of sciences 106:17588\u0026ndash;17593\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eUrano D, Phan N, Jones JC, Yang J, Huang J, Grigston J, Philip Taylor J, Jones AM (2012) Endocytosis of the seven-transmembrane RGS1 protein activates G-protein-coupled signalling in Arabidopsis. Nat Cell Biol 14:1079\u0026ndash;1088\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang J, Wang Z, Du X, Yang H, Han F, Han Y, Yuan F, Zhang L, Peng S, Guo E (2017) A high-density genetic map and QTL analysis of agronomic traits in foxtail millet [Setaria italica (L.) P. Beauv.] using RAD-seq. PLoS ONE 12:e0179717\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eW\u0026uuml;rschum T, Langer SM, Longin CFH (2015) Genetic control of plant height in European winter wheat cultivars. Theor Appl Genet 128:865\u0026ndash;874\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eXue C, Zhi H, Fang X, Liu X, Tang S, Chai Y, Zhao B, Jia G, Diao X (2016) Characterization and fine mapping of SiDWARF2 (D2) in foxtail millet. Crop Sci 56:95\u0026ndash;103\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYu J, Bai X, Zhang K, Feng L, Yu Z, Jiao X, Guo Y (2024) Assessment of Breeding Potential of Foxtail Millet Varieties Using a TOPSIS Model Constructed Based on Distinctness, Uniformity, and Stability Test Characteristics. Plants 13:2102\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhang K, Fan G, Zhang X, Zhao F, Wei W, Du G, Feng X, Wang X, Wang F, Song G (2017) Identification of QTLs for 14 agronomically important traits in Setaria italica based on SNPs generated from high-throughput sequencing. G3: Genes, Genomes, Genetics 7:1587\u0026ndash;1594\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhao K, Tung C-W, Eizenga GC, Wright MH, Ali ML, Price AH, Norton GJ, Islam MR, Reynolds A, Mezey J (2011) Genome-wide association mapping reveals a rich genetic architecture of complex traits in Oryza sativa. Nat Commun 2:467\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhao M, Zhi H, Zhang X, Jia G, Diao X (2019) Retrotransposon-mediated DELLA transcriptional reprograming underlies semi-dominant dwarfism in foxtail millet. Crop J 7:458\u0026ndash;468\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhu M, He Q, Lyu M, Shi T, Gao Q, Zhi H, Wang H, Jia G, Tang S, Cheng X (2023) Integrated genomic and transcriptomic analysis reveals genes associated with plant height of foxtail millet. Crop J 11:593\u0026ndash;604\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":"","lastPublishedDoi":"10.21203/rs.3.rs-7333962/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7333962/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003ePlant height is pivotal for lodging resistance, harvestability, and yield potential in foxtail millet (\u003cem\u003eSetaria italica\u003c/em\u003e). We phenotyped 209 diverse Dabaigu accessions across three contrasting environments and performed whole-genome resequencing (mean depth ~38×). After standard filtering, 833,157 high-quality SNPs were retained. Neighbor-joining and principal component analysis consistently resolved four genetic groups with minor differences in sample assignment. A mixed linear model GWAS (EMMAX) detected four loci significantly associated with plant height. Comparative genomics with rice and Arabidopsis shortlisted ten candidates; notably, \u003cem\u003eSETIT_033071mg\u003c/em\u003e showed stem-specific high expression at the shooting stage in Yugu1, with low expression elsewhere. Haplotype analysis of \u003cem\u003eSETIT_033071mg\u003c/em\u003e in an expanded panel of 313 accessions across five field sites resolved six gene haplotypes (H1–H6), among which H2 was consistently associated with reduced plant height across environments. In parallel, we developed a KASP assay (kPH8552) tagging the lead SNP at \u003cem\u003eqPH2.2\u003c/em\u003e (distinct from the \u003cem\u003eSETIT_033071mg\u003c/em\u003elocus). In the Dabaigu panel, the assay achieved an 80.4% call rate (G:G=74; T:T=94), and the T allele was consistently associated with reduced height by −4.48 cm (BLUE; \u003cem\u003eP\u003c/em\u003e = 0.044). These findings refine the genomic basis of plant height in foxtail millet and provide actionable targets and a candidate marker for breeding lodging-resistant, semi-dwarf cultivars, pending validation in independent populations.\u003c/p\u003e","manuscriptTitle":"Genome-wide Association Analysis and Candidate Gene Identification of Plant Height in Dabaigu (Foxtail Millet)","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-10 19:36:28","doi":"10.21203/rs.3.rs-7333962/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Major revisions","date":"2026-01-06T20:43:27+00:00","index":"","fulltext":""},{"type":"reviewerAgreed","content":"","date":"2025-09-29T08:30:38+00:00","index":0,"fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-09-28T23:22:48+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-08-12T14:55:13+00:00","index":"","fulltext":""},{"type":"submitted","content":"Theoretical and Applied Genetics","date":"2025-08-09T08:48:24+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":"cf241b1e-7321-4a10-b84e-3075e13ecb00","owner":[],"postedDate":"October 10th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-01-13T06:58:23+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-10 19:36:28","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7333962","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7333962","identity":"rs-7333962","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}