The research focused on four fundamental measurements used in livestock breeding to assess body conformation: body weight, body height, body oblique length, and chest circumference. Each of these traits captures a different aspect of an animal’s physical development. Body weight reflects overall mass and is often the single most economically important figure for meat production. Body height and body oblique length describe skeletal frame size, while chest circumference provides a proxy for body depth and volume. Together, they form the standard toolkit of quantitative measures that breeders use to judge whether an animal is maturing well. Understanding which genes influence these numbers is crucial for designing breeding programs that can improve productivity while preserving the breed’s adaptation to its extreme environment.

To gather the raw material for the analysis, the researchers sampled 536 Tianzhu White Yaks representing three distinct growth stages: animals at one year of age, animals at two and a half years, and animals aged four years or older. This staged design matters because growth is not a single event but a trajectory. Genes that govern early skeletal development may not be the same ones that control later muscle deposition and fat accumulation, and including animals from different life stages gives the study a window onto that changing biology. Blood samples were collected with ethical approval from the Lanzhou Institute, in strict compliance with China’s Guidelines for Ethical Treatment of Experimental Animals, and the four body measurements were recorded for each animal.

The genomic strategy the team chose was low-coverage resequencing, a technique that reads the genome at relatively shallow depth across hundreds of individuals rather than deeply sequencing a small number. The advantage is scale: with enough animals, statistical methods can reconstruct each individual’s full genotype computationally in a process called genotype imputation. After imputation and rigorous quality control, the study retained a staggering total of 24,383,269 single nucleotide polymorphisms, or SNPs, the single-letter differences in DNA sequence that pepper the yak genome. That depth of marker information, nearly two dozen million variants, is what gives a GWAS the resolving power to home in on genomic regions that genuinely influence a trait rather than hitchhiking along with them.

Armed with this dense variant map and the body measurements, the researchers applied a mixed linear model for the association testing. Mixed linear models are the workhorse of modern GWAS because they simultaneously test the link between each SNP and a trait while statistically accounting for the complex family structure and relatedness that are almost unavoidable in livestock populations. Without such correction, related animals sharing large stretches of genome would generate false associations. The team also used standard quality checks such as principal component analysis to characterize population structure and quantile-quantile plots to verify that their model was behaving as expected, with the observed test statistics aligning closely with what would be predicted under their assumptions.

The scan paid off. Across the four traits, the study identified a total of 366 significant SNP loci, which the researchers annotated to 32 candidate genes. For body weight, the candidates included PLCB1, SAR1B, and SRSF6. Body height pointed toward BRPF3, MAPK13, and ZNF76. Body oblique length was associated with variants near ABCD3 and NTM, while chest circumference flagged LMNTD1 and ACVR1. The latter gene is particularly notable to anyone familiar with developmental biology: ACVR1 encodes a receptor in the bone morphogenetic protein signaling pathway, a family of pathways long known to regulate skeletal growth, bone formation, and patterning during development. Its appearance in a chest circumference analysis, a measure closely tied to skeletal dimensions, fits neatly with what biologists already understand about how the body frame is built.

To move from a list of gene names toward biological meaning, the team performed functional enrichment analysis, asking whether the candidate genes cluster in particular cellular processes more than chance would predict. Several themes emerged. Multiple genes were implicated in the negative regulation of the G1/S transition of the mitotic cell cycle, the checkpoint that decides whether a cell commits to dividing. That process is fundamental to how organisms grow, since the number and timing of cell divisions in developing tissues directly shape final organ and body size. Another enriched process was fatty acid beta-oxidation, the metabolic pathway by which cells break down fats to generate energy, a fitting connection for traits like body weight and chest circumference that are influenced by both lean tissue and fat reserves, especially in an animal that must survive long, cold winters on sparse forage.

One of the most intriguing findings involved three genes, TEAD3, ZNF76, and RBL1, that were significantly enriched in the transcription regulator complex, a cellular component consisting of assemblies of proteins that control which genes are switched on or off. Transcription regulators sit high in the hierarchy of cellular control, and their involvement suggests that variation in these loci could influence cell proliferation, differentiation, and tissue formation, the underlying processes that ultimately determine how large and well-formed an animal becomes. RBL1, for instance, is part of the retinoblastoma family of proteins that act as gatekeepers of the cell cycle, while TEAD proteins are effectors of the Hippo signaling pathway, an evolutionarily conserved organ-size control system. Such connections, if confirmed by functional studies, would tie the breed’s growth biology to some of the deepest mechanisms known in developmental biology.

The practical implications reach well beyond the laboratory. Tianzhu White Yaks are a valuable genetic resource, prized on the plateau for their white coat and their role in the livelihoods of herding communities, and the study’s authors emphasize that identifying markers associated with growth traits is crucial for breeding. Marker-assisted selection and genomic selection both depend on precisely this kind of catalog: a validated set of DNA variants that predict how an animal will grow before it reaches maturity. By genotyping young calves for the significant SNPs identified here, breeders could, in principle, make earlier and more accurate selection decisions, accelerating genetic improvement while maintaining the genetic diversity that makes the breed resilient. The authors note that the identified SNPs and candidate genes will facilitate the development and utilization of Tianzhu White Yak resources.

The study also carries broader scientific weight. Yaks occupy a unique position among livestock as the large mammals best adapted to life above 3,000 meters, and every well-powered GWAS in the species adds to a comparative framework that spans cattle, yak, and their hybrids. Because the analysis spans animals of different ages, the same dataset could support future work on how gene effects change across development, and the 24 million SNP resource itself is a legacy asset for the yak genetics community. For now, the 366 loci and 32 candidate genes stand as a hypothesis list, the starting point for the functional experiments the researchers say should follow. From a blood draw on a windswept plateau to nearly two dozen million genetic letters, the study shows how modern genomics can translate the centuries-old practice of herding into a molecular map of what makes an animal grow.

Subject of Research: Genome-wide association analysis of growth traits in the Tianzhu White Yak

Article Title: Genome-wide association analysis of growth traits in Tianzhu White Yak

Article References: Yang, G., Qi, X., La, Y., Ma, X., Liu, Y., Wang, Z., Luo, W., Ren, W., Zhang, Z., Chu, M., Wu, X., Guo, X., Qi, W., & Liang, C. (2026). Genome-wide association analysis of growth traits in Tianzhu White Yak. BMC Genomics. https://doi.org/10.1186/s12864-026-13444-w

Image Credits: AI Generated

DOI: 10.1186/s12864-026-13444-w

Keywords: Tianzhu White Yak, GWAS, growth traits, SNPs, low-coverage sequencing, genotype imputation, livestock breeding, BMC Genomics, body conformation, ACVR1, transcription regulation, Tibetan Plateau

The core problem COSIGT addresses is one of depth. Recent methods such as Locityper have already shown that pangenome references, which store many haplotypes as paths through a graph rather than a single consensus sequence, can dramatically improve targeted genotyping at difficult loci when read data are abundant. Locityper aligns reads to locus-specific haplotypes and selects the best-fitting diploid pair by optimizing alignment accuracy, insert-size concordance, and coverage balance. But that alignment-based likelihood machinery has an Achilles heel: as sequencing coverage falls, the signal-to-noise ratio degrades and genotyping accuracy drops with it. That matters enormously for population-scale studies and biobank cohorts sequenced at variable and often shallow depth, and it matters even more for ancient DNA, where postmortem degradation and microbial contamination routinely push effective coverage below 2X.

COSIGT takes a fundamentally different mathematical route. The pipeline first constructs a local variation graph for the target locus from haplotype-resolved assemblies, rather than querying an entire genome-wide pangenome. This locus-specific design allows parameter tuning, rapid incorporation of new assemblies, and faster genotyping. Reads pre-aligned to the region are extracted, supplemented with unmapped reads rescued by a k-mer filtering tool called kfilt, and then mapped to the local graph. From the graph nodes that reads traverse, the pipeline builds a coverage vector for the sample, in which each element reflects length-normalized, multi-mapping-aware coverage at a node. Every haplotype in the graph is likewise represented as a coverage vector of node traversal counts. COSIGT then enumerates all possible diploid haplotype pairs, sums their vectors to create synthetic genotype profiles, and computes the cosine similarity between each synthetic profile and the observed sample vector. The pair with the highest similarity wins.

The elegance of the approach lies in what cosine similarity measures. Because the metric captures the orientation of a vector rather than its magnitude, it is inherently invariant to overall sequencing depth. A sample sequenced at 1X coverage produces the same relative coverage profile shape as one sequenced at 30X, merely scaled down, and cosine similarity is blind to that scaling. Likelihood-based methods, by contrast, depend on absolute read counts and lose power as counts dwindle. In benchmarks across 326 challenging medically relevant genes and 265 structurally variable regions using short-read data from the 1000 Genomes Project, with pangenome graphs built from assemblies of the Human Pangenome Reference Consortium and the Human Genome Structural Variation Consortium, both COSIGT and Locityper performed well at 5X and 30X coverage, with more than 93 percent high-quality calls. Locityper retained an edge at 30X, reaching 98.2 percent versus 93.9 percent for COSIGT at the medically relevant genes.

At low coverage, the picture reversed dramatically. At 1X, COSIGT delivered 93.4 percent of calls at mid-or-higher quality compared with 84.5 percent for Locityper, and at 2X the gap persisted at 95.8 percent versus 93.4 percent. The advantage became even starker on simulated ancient DNA. Across 48 medically relevant genes enriched for pharmacogenetic and immunogenetic content, including HLAs, CYPs, and mucins, COSIGT maintained roughly 94 percent mid-or-higher quality calls at 1X while Locityper fell to about 46 percent; at 2X the figures were roughly 95 percent versus 60 percent. COSIGT outperformed Locityper at all 48 genes and showed only marginal degradation relative to matched modern DNA at the same coverage. The simulations, generated with a purpose-built simulator called ancestralsim, incorporated realistic ancient DNA damage patterns, short fragment lengths, and contamination levels of 0 or 10 percent, making the result a demanding test.

Robustness to missing references was assessed through leave-all-out benchmarks, in which the true haplotypes of each sample were excluded from the pangenome graph. Even then, COSIGT remained near-optimal, achieving at least 87 percent of the best possible quality at most loci, with 87.6 percent of medically relevant genes and 90.8 percent of structural variant regions falling in the top quintile of achievable quality. An ancestry-mismatched test, in which Peruvian and Colombian samples were genotyped against a graph containing no admixed American assemblies, yielded 87.2 percent high-quality calls at the medically relevant genes, suggesting the method degrades gracefully when the reference panel does not perfectly match the population under study.

The team also demonstrated the tool at genuine population scale. Applying COSIGT to 1,085 whole-genome samples from the Italian Moli-sani cohort, sequenced at roughly 20X coverage, the researchers performed HLA typing across seven classical HLA genes and achieved a mean haplotype-level accuracy of 89.5 percent against types imputed with HLA*IMP:02 from microarray data, comparable to the 90.8 percent achieved by the dedicated HLA genotyping tool T1K. More broadly, the authors report having applied COSIGT to more than 6,000 modern and ancient human genomes, demonstrating population-scalable analysis of complex repeats and multi-copy genes. The pipeline is implemented in Snakemake with containerized deployment, parallelizes across regions and samples, and its per-sample genotyping steps are computationally light, with a median runtime of about 0.02 minutes and median memory use of 52 megabytes for the core genotyping step.

The authors are candid about limitations. Genotyping accuracy ultimately depends on the quality and completeness of the input pangenome: haplotypes absent from the reference panel cannot be called exactly, and novel structural variants will be assigned to the most similar available haplotype. The leave-all-out benchmarks quantify how much accuracy survives this constraint, but they also make clear that expanding pangenome references remains essential. COSIGT currently requires pre-aligned BAM or CRAM files, creating a dependency on reference-based alignment, though future work aims to accept raw FASTQ input directly and to implement haplotype subsampling to keep the method scalable as pangenomes grow. The framework also currently supports diploid genotyping, although the underlying mathematics generalizes to arbitrary ploidy by enumerating k-tuples of haplotypes.

The implications reach well beyond a single software release. Pangenome references are expanding rapidly in size, quality, and taxonomic breadth, and as they do, the bottleneck shifts from reference completeness to scalable genotyping. By making complex-locus genotyping reliable at 1 to 2X coverage, COSIGT opens the door to systematically mining the vast archives of low-coverage sequencing data already sitting in biobanks and legacy datasets, and to bringing archaeological and ancient genomes into population-level biomedical analyses from which their shallow depth previously excluded them. Larger cohorts become affordable at reduced sequencing cost, samples of heterogeneous quality can be combined in a single analysis, and the allelic diversity captured in pangenome references becomes practically accessible across the full spectrum of sequencing depths. The pipeline is compatible with long-read input as well, which could extend its utility to low-coverage long-read datasets too sparse for assembly. For a field that has long treated structurally complex loci as terra incognita in shallow data, the message is that the map, and the compass to read it, may finally have arrived together.

Subject of Research: Pangenome graph-based genotyping of structurally complex genomic loci from low-coverage sequencing data

Article Title: COSIGT: population-scalable genotyping of complex loci from low-coverage sequencing data using pangenome graphs

Article References: Bolognini, D., Guarracino, A., Paleni, C., Dudley, T. S., Iacoviello, L., Raveane, A., Sudmant, P. H., Garrison, E., & Soranzo, N. (2026). COSIGT: population-scalable genotyping of complex loci from low-coverage sequencing data using pangenome graphs. Genome Biology, 27(1), Article 286. https://doi.org/10.1186/s13059-026-04242-4

Image Credits: AI Generated

DOI: 10.1186/s13059-026-04242-4

Keywords: pangenome graphs, genotyping, low-coverage sequencing, ancient DNA, structural variation, copy-number variation, HLA typing, CYP2D6, cosine similarity, population genomics, Genome Biology, bioinformatics