Data Imbalance in Drug Response Prediction - Multi-Objective Optimization Approach in Deep Learning Setting

Drug response prediction (DRP) methods tackle the complex task of associating the effectiveness of small molecules with the specific genetic makeup of the patient. Anti-cancer DRP is a particularly challenging task requiring costly exper-iments as underlying pathogenic mechanisms are broad and associated with multiple genomic pathways. The scientific community has exerted significant efforts to generate public drug screening datasets, giving a path to various machine learning (ML) models that attempt to reason over complex data space of small compounds and biological characteris-tics of tumors. However, the data depth is still lacking compared to computer vision or natural language processing domains, limiting current learning capabilities. To combat this issue and increase the generalizability of the DRP mod-els, we are exploring strategies that explicitly address the imbalance in the DRP datasets. We reframe the problem as a multi-objective optimization across multiple drugs to maximize deep learning model performance. We implement this approach by constructing Multi-Objective Optimization Regularized by Loss Entropy (MOORLE) loss function and plug-ging it into a Deep Learning model. We demonstrate the utility of proposed drug discovery methods and make sugges-tions for further potential application of the work to promote equitable outcomes in the healthcare field. Availability: https://github.com/AlexandrNP/MOORLE Contact: onarykov@anl.gov