BioBO: Biology-informed Bayesian Optimization for Perturbation Design

Efficient design of genomic perturbation experiments is crucial for accelerating drug discovery and therapeutic target identification, yet exhaustive perturbation of the human genome remains infeasible. Bayesian optimization (BO) has recently emerged as a powerful framework for selecting informative interventions, but existing approaches often fail to exploit domain-specific biological prior knowledge. We propose Biology-Informed Bayesian Optimization (BioBO), a method that integrates Bayesian optimization with multimodal gene embeddings and enrichment analysis, a widely used tool for gene prioritization in biology, to enhance surrogate modeling and acquisition strategies. BioBO leverages biologically grounded priors within the $\pi$BO framework to balance exploration and exploitation. Through experiments on the established public benchmark, we demonstrate that BioBO improves labeling efficiency, consistently outperforms conventional BO, and identifies top-performing perturbations more effectively. These results highlight the potential of incorporating structured biological knowledge into BO frameworks for more efficient and interpretable genomic experimental design.