StarBASE-GP IEEE TEVC 2025

We present the Star-Based Automated Single-locus and Epistasis analysis tool \u2013 Genetic Programming (StarBASE-GP), an automated framework for discovering biologically meaningful genetic variants associated with phenotypic variation in large-scale genomic datasets. StarBASE-GP uses a genetic programming\u2013based multi-objective optimization strategy to evolve machine learning pipelines that simultaneously maximize explanatory power ($r^2$) and minimize pipeline complexity. Biological domain knowledge is integrated at multiple stages, including the use of nine inheritance encoding strategies to model deviations from additivity, a custom linkage disequilibrium pruning node that reduces redundancy among features, and a dynamic variant recommendation system that prioritizes informative candidates for pipeline inclusion. We evaluate StarBASE-GP on a cohort of \\textit{Rattus norvegicus} (brown rat) to identify variants associated with body mass index, benchmarking its performance against a random baseline and a biologically na\u00efve version of the tool. StarBASE-GP consistently evolves Pareto fronts with superior performance, yielding higher precision in identifying both ground truth and novel quantitative trait loci, highlighting relevant targets for future validation. By incorporating evolutionary search and relevant biological theory into a flexible automated machine learning framework, StarBASE-GP demonstrates robust potential for advancing variant discovery in complex traits.