Integrative multi-omics analysis of host genetics, rumen transcriptome and microbiome prioritizes candidate features associated with body weight in Hanwoo steers

Hanwoo (Korean cattle) is an indigenous breed of high economic value in Korea, where improving body weight (BW) remains a priority for both productivity and farm profitability. Because growth traits are shaped by multiple biological layers, including host genetics, rumen microbial ecology, and transcriptomic regulation, a single-omics perspective is often insufficient. To address this complexity, we applied an integrative multi-omics framework, DIABLO (Data Integration Analysis for Biomarker discovery using Latent cOmponents), combining genomic (42,091 SNPs), rumen transcriptomic (RNA-seq), and microbiome (16S rRNA) data from 58 steers. After adjusting BW for age, animals were classified into High- and Low-BW groups. Feature pre-selection yielded 100 discriminant SNPs, 161 large-effect candidate expression features, and 20 candidate microbial features showing nominal abundance differences between groups. Integration of these datasets highlighted exploratory SNP–gene–microbiome associations, including three SNPs, ARS-BFGL-NGS-104362, ARS-BFGL-NGS-43924 and ARS-BFGL-NGS-24293, that remained significantly associated with BW. These variants were linked to epithelial and metabolic genes such as FUT1 , KRTDAP , and TCN1 , which in turn showed strong correlations with Anaerobutyricum , a butyrate-producing genus identified as an exploratory microbial candidate. Taken together, these results suggest candidate host epithelial and rumen microbial axes associated with BW variation, rather than establishing causal mechanisms. By applying a supervised multi-omics integration strategy, this study identifies exploratory candidate multi-omics features associated with BW variation in Hanwoo steers. Because of the modest sample size and the supervised nature of the integration framework, the resulting signature should be interpreted as hypothesis-generating and requiring validation in larger independent cohorts.