Integrating Genomic and Environmental Data Using Machine Learning for Vernalization Response Prediction

Abstract

This research investigates the integration of genomic and environmental data using Random Forests to predict vernalization response in barley. Vernalization, the requirement of a prolonged period of cold to induce flowering, is a critical adaptive trait for temperate cereal crops. The study compiles a comprehensive dataset of barley genotypes, gene expression levels related to vernalization (e.g., VRN1, VRN2, and FT1 genes), and detailed environmental variables including temperature, photoperiod, soil moisture, and humidity. By employing a Random Forest algorithm, the research identifies key genetic and environmental factors that influence vernalization. The findings suggest that this machine learning approach effectively models the complex interactions between genotype and environment, providing insights for breeding climate-resilient barley varieties. This integrative approach not only enhances our understanding of the genetic basis of vernalization but also aids in the development of barley varieties with optimized flowering times for diverse climatic conditions.