Award Talk: Integrating Multi-OMICs Data into Practical Swine Production

Over the past decade, advances in sequencing and high throughput profiling technologies have revolutionized our ability to characterize genes, proteins and metabolites with unprecedented depth and detail in biological samples. These tools offer exciting opportunities to understand the complex interactions between diet, environment, animals, their gut microbes, and physiological performance. However, despite rapid technological progress, a persistent challenge remains: how do we translate these high-dimensional datasets into actionable insights for production management? Researchers and industry professionals alike often ask the same questions: how specific genes, microorganisms or metabolites converge to impact animal production systems? What are the practical steps to move from data to decisions? For example, gut microbiome influences swine production far beyond its contribution to nutrient utilization, by playing indispensable roles in immune system function, direct defense against pathogens, intestinal integrity, energetic turnover, bile acids metabolism, and even neuroactivity. These arms converge to impact disease resistance, stress responses, feed efficiency, and overall animal behavior and welfare. Understanding these complex interactions requires approaches that capture both nutritional and broader physiological consequences of microbiome composition and function. Traditional experimental designs in animal nutrition effectively predict physiological outcomes based on low-dimensional datasets. Yet, when the number of predictors and response variables exceeds sample sizes, standard statistical approaches fall short in capturing full biological variation. This is where dimensionality reduction techniques such as principal component analysis (PCA), sparse partial least squares discriminant analysis (SPLS-DA), canonical correlation analysis (CCA), multi-omics integration methods coupled with Biomarker discovery via Latent cOmponents and sparse integrative discriminant analysis (SIDA), among other methods, become essential for bridging the gap between complex biological data and practical application. Current technological developments in user-friendly bioinformatics and statistical pipelines allow nutritionists and swine production scientists alike to incorporate multi-omics information into practical applications without requiring advanced computational expertise. As the swine industry increasingly incorporates economical alternative feedstuffs (e.g. high-fiber co-products) in swine nutrition, integrating multi-omics data to better understand energy utilization, health, and immune function has become a critical need. Such knowledge can be applied to optimize diet formulation, save production costs and ultimately improve the sustainability and profitability of swine production systems.