Navigating high-order protein fitness landscapes via deep learning on directed evolution trajectories

Abstract

Accurately predicting the fitness effects of high-order mutations is a grand challenge in understanding and engineering proteins. Existing models, including pretrained protein language models, struggle to capture the multiresidue interactions that govern these effects. Here, we introduce DENet, a deep learning framework that harnesses the rich comutation information within directed evolution (DE) trajectories to reconstruct high-resolution fitness landscapes for deciphering and engineering of complex protein variants. Applied to the cancer target KRAS, DENet-guided screening systematically identified high-order mutants with potent activities and uncovered hidden allosteric mechanisms. For MEK1, DENet nominated complex variants with >1,000-fold increased drug resistance, revealed synergistic tail mutations, and retrospectively identified over 75% of known clinical mutations, largely outperforming existing models. To broaden the framework’s applicability, we developed an in silico strategy that simulates directed evolution to infer comutation information from widely available single-mutant datasets. DENet provides a quantitative framework for navigating complex fitness landscapes, uniting the rational engineering of multimutation proteins with the elucidation of their mechanisms and clinical implications.