A comprehensive exploration of the druggable conformational space of protein kinases using AI-predicted structures

Protein kinase function and interactions with drugs are controlled in part by the movement of the DFG and alpha C-Helix motifs that are related to the catalytic activity of the kinase. Small molecule ligands elicit therapeutic effects with distinct selectivity profiles and residence times that often depend on the active or inactive kinase conformation(s) they bind. Modern AI-based structural modeling methods have the potential to expand upon the limited availability of experimentally determined kinase structures in inactive states. Here, we first explored the conformational space of kinases in the PDB and models generated by AlphaFold2 (AF2) and ESMFold, two prominent AI-based protein structure prediction methods. Our investigation of AF2's ability to explore the conformational diversity of the kinome at various multiple sequence alignment (MSA) depths showed a bias within the predicted structures of kinases in DFG-in conformations, particularly those controlled by the DFG motif, based on their overabundance in the PDB. We demonstrate that predicting kinase structures using AF2 at lower MSA depths explored these alternative conformations more extensively, including identifying previously unobserved conformations for 398 kinases. Ligand enrichment analyses for 23 kinases showed that, on average, docked models distinguished between active molecules and decoys better than random (average AUC (avgAUC) of 64.58), but select models perform well (e.g., avgAUCs for PTK2 and JAK2 were 79.28 and 80.16, respectively). Further analysis explained the ligand enrichment discrepancy between low- and high-performing kinase models as binding site occlusions that would preclude docking. The overall results of our analyses suggested that, although AF2 explored previously uncharted regions of the kinase conformational space and select models exhibited enrichment scores suitable for rational drug discovery, rigorous refinement of AF2 models is likely still necessary for drug discovery campaigns. Greater abundance of kinase structural data in inactive conformations, currently lacking in structural databases, would improve our understanding of how protein kinases function, and expand drug discovery and development for this important family of therapeutic targets. Modern approaches utilizing artificial intelligence and machine learning, like AlphaFold2 and ESMFold, have potential for efficiently capturing novel protein conformations. We provide evidence for a bias within AlphaFold2 and ESMFold to predict structures of kinases in their active states, similar to their overrepresentation in the PDB. We show that lowering the multiple sequence alignment depth used in the AlphaFold2 algorithm can help explore the kinase conformational space more broadly. Through a series of quantitative and visual analyses of the models, we also offer a critique of AlphaFold2-generated models, highlighting their potential utility in drug discovery, but also underscoring the need for further refinement to enhance their suitability for rational drug design. Furthermore, many of the high-quality and high-enriching models we make available may represent starting points for novel drug discovery campaigns.