Rory Coffey

Systematic feature and architecture evaluation reveals tokenized learned embeddings enhance siRNA efficacy prediction

This study evaluates features and architectures for siRNA efficacy prediction, finding that tokenized learned embeddings improve accuracy and generalizability over complex models.

Reinforced molecular dynamics: Physics-infused generative machine learning model explores CRBN activation process

The paper introduces a machine learning model called reinforced molecular dynamics (rMD) that uses MD and free-energy data to efficiently explore protein conformational space, applied to study CRBN activation.