Multi-omics analysis

Multi-omics identification of blood biomarkers for transition cow disease and associated quantitative trait loci

---

Abstract

The transition period (three weeks before to three weeks after calving) is the most critical stage in a dairy cow's lactation cycle. Major metabolic adaptations during this period predispose cows to diseases including mastitis, metritis, retained fetal membranes, milk fever, and ketosis, that reduce productivity, increase treatment costs, and contribute to early culling, yet the molecular mechanism and genetic basis underlying susceptibility remain poorly defined. This study integrated lipidomics, metabolomics, and transcriptomics with genome-wide association studies (GWAS) to identify serum biomarkers and associated quantitative trait loci (QTL) for transition cow diseases.In a case-control study, blood samples were collected from 200 clinically diseased and 331 healthy transition cows within 70 days after calving across 15 pasture-based farms in Victoria, Australia. Technical and biological confounders (farm, sampling date, days in milk, breed, parity, haemolysis, and batch effect for metabolomics) were corrected prior to analysis. Biomarkers were identified using a consensus approach integrating differential abundance (DA) analysis, machine learning (ML) predictive models (Random Forest and Elastic Net), and receiver operating characteristic curve analysis. Features significant in DA and selected by both ML were prioritised as high-confidence biomarkers. GWAS mapped QTLs, and heritability were estimated using genomic REML (GREML) implemented in GCTA.Multi-omics profiling revealed extensive molecular changes in diseased compared to healthy cows. Lipidomics identified 163 DA lipids (142 decreased, 21 increased; FDRConsensus biomarker panels were established across all platforms: 40 lipids, 18 metabolites, and 39 genes, with transcriptomic markers demonstrating the highest discriminatory performance for GWAS features (AUC 0.80). Heritability estimates for lipids ranged from 0 to 0.79, indicating variable genetic control. GWAS revealed shared genetic architecture across omics layers: lipid QTLs mapped to chromosomes 11 and 19, metabolite QTLs to chromosomes 10, 11, 16, 18, 23, and 25, and gene expression QTLs overlapped with lipid QTLs on chromosome 19 and metabolite QTLs on chromosomes 10 and 16. This co-localisation suggests transcriptional regulation may mediate metabolic dysregulation. Eleven GWAS associated lipids and 775 genes were also differentially abundant in diseased animals, linking genetic variation to disease-associated molecular phenotypes.This multi-omics approach demonstrates that lipidomic and metabolomic signatures reflect key physiological disruptions during transition cow diseases. Integration of molecular biomarkers with genomic markers enables discovery of disease indicators and genetic variants, supporting development of genomic breeding values for improved transition health and enhancing animal welfare and farm profitability.

Keywords: 2026

How to Cite:

Kudriashova, A., Chamberlain, A., Tahir, M., Liu, Z., Rochfort, S., Mason, B., Elkins, A., Ram, D., Hemsworth, J., Reich, C. & Goddard, M., (2026) “Multi-omics identification of blood biomarkers for transition cow disease and associated quantitative trait loci”, World Congress on Genetics Applied to Livestock Production Digital Archive 2026(1): 2286256. doi: https://doi.org/10.31274/wcgalp.23897

Rights: 1