GENOMIC AND MICROBIOME ANALYSIS TO IMPROVE FILLET YIELD AND QUALITY TRAITS IN RAINBOW TROUT

Abstract

Rainbow trout is the most widely farmed cool- and cold-water freshwater fish species in the United States. It is primarily cultivated for its fillets, rich in protein and essential fatty acids. As interest grows in enhancing the productive efficiency of rainbow trout, most breeding programs still rely heavily on traditional, pedigree-based selection. Incorporating genomic information holds promise for accelerating genetic gains, particularly for complex or lethally measured traits such as fillet yield and quality. However, even with the availability of genomic data, a substantial portion of phenotypic variation remains unexplained—an issue commonly referred to as “missing heritability.” Recently, the gut microbiome has emerged as a potential contributor to phenotypic variation through host–microbiome interactions. Some microbiome features may be heritable, which could help account for portions of the missing heritability. In this study, we investigated the genomic and microbiome architecture of rainbow trout families from USDA breeding programs, aiming to improve fillet yield and quality traits. Fillet color, a key quality attribute influencing consumer preferences, was a major focus. Using genome-wide association studies (GWAS) and RNA-Seq, we characterized the genetic basis of variation in fillet yield and quality traits. GWAS identified SNPs within genes involved in carotenoid metabolism (e.g., β,β-carotene 15,15-dioxygenase, retinol dehydrogenase), myoglobin regulation (ATP5F1B), and structural maintenance of muscle tissue (klh41b, COL28A1, CTSK). RNA-Seq further highlighted genes involved in lipid/carotenoid metabolism, ribosomal function, mitochondrial activity, and stress response as contributors to fillet color variation. We also integrated expression quantitative trait loci (eQTL) mapping to connect genetic variants with gene expression and phenotypic outcomes. A total of 6,275 cis-eQTLs were identified near promoter regions influencing gene expression. Notable candidate genes included GABRR1-like, AZIN1, and ATP6V1 for body weight; TNNI2, PDLIM3, CDK5, and CYP3A27 for tissue robustness; and GATD3A, PPP2R1BB, and USP6NL for muscle development. A specific A-A-C eQTL haplotype was strongly associated with elevated expression of ATP6V1. These eQTLs overlapped with regions enriched for active enhancers, making them strong candidates for inclusion in genomic selection strategies. To assess the role of the gut microbiome, we employed both 16S rRNA gene sequencing and shotgun metagenomics to identify taxa and pathways predictive of fillet color. Fish with red fillets were enriched (LDA score > 1.5) for Leuconostoc lactis, Corynebacterium variabile, Jeotgalicoccus halotolerans, and Leucobacter chromiireducens—bacterial species with known probiotic functions and carotenoid biosynthesis capabilities. These fish also exhibited enrichment of the Methylerythritol Phosphate (MEP) pathway, which is involved in carotenoid production. Microbiome-wide association studies (MWAS) revealed microbial taxa and metabolic pathways associated with growth traits. These included pathways such as fatty acid elongation, fatty acid oxidation II, pyruvate fermentation to isobutanol, glyoxylate cycle, mixed acid fermentation, superpathway of branched chain amino acid biosynthesis, and peptidoglycan biosynthesis II —all of which may influence growth performance in rainbow trout. Lastly, we demonstrated the feasibility of combining low-coverage whole-genome sequencing (1x WGS) with genotype imputation and microbiome data for genomic prediction. Our findings show that combining 1x WGS with pedigree data provides predictive power that outperforms the pedigree-only-based models. Moreover, incorporating microbiome data improved prediction accuracy for body weight and muscle yield by 7.2%, and 8.6%, respectively, offering potential for substantial cumulative gains across generations. In conclusion, this study identifies key genetic and microbiome contributors to variation in fish growth, fillet yield and quality traits in rainbow trout. Our results highlight the potential of integrating genomic and microbiome information to enhance selective breeding strategies, promote sustainable aquaculture, improve protein security, and deliver economic benefits to the U.S. aquaculture industry.