A machine learning approach to define antimalarial drug action from heterogeneous cell-based screens
Abstract
Antimalarial drug-resistance is driving calls for discovery of drugs with new mechanisms of action. Current cell-based drug screens are restricted to binary live/dead readouts with no provision for mechanism prediction. Machine learning methods are increasingly being used to improve information extraction from imaging data. Such methods, however, work poorly with heterogeneous phenotypes and generally require time-consuming human-led training. We have developed a semi-supervised machine learning approach, combining human- and machine-labeled training data from mixed human malaria parasite cultures. Designed for high-throughput and high-resolution screening, our semi-supervised approach is robust to natural parasite morphological heterogeneity and reproducibly orders parasite developmental stages. Our approach also reproducibly detects and clusters drug-induced morphological outliers by mechanism of action, demonstrating the potential power of machine learning for accelerating cell-based drug discovery.