Structure-conditioned self-supervised learning of residue interaction constraints in protein kinases for variant interpretation.

Abstract

Protein kinases are key regulators of cellular signaling and are frequently implicated in human diseases. Although kinase domains are structurally conserved, predicting the effects of amino acid substitutions remains challenging as mutations often introduce subtle structural perturbations that are not captured by sequence-based or evolutionary methods. Existing supervised approaches further rely on pathogenicity annotations that are inconsistent across databases, thereby motivating the development of structure-based, label-independent frameworks for mutation effect prediction. We present a structure-based method using SE(3)-transformers to learn residue compatibility with the local structural environment from experimentally resolved kinase 3D structures. Proteins are represented as atom-level graphs with physicochemical descriptors derived from the CHARMM force field and spatial connectivity. The model is trained on two self-supervised tasks given local structural context: masked residue atom reconstruction and masked residue classification. This formulation enables learning of geometric and physicochemical constraints without relying on pathogenicity labels. Evaluation using reconstruction loss, residue prediction accuracy, and comparison with BLOSUM substitution patterns indicate that the model captures biologically meaningful relationships between residue identity and 3D structural context. We interpret the scores assigned to alternative amino acids as measures of structural fitness, where low-scoring residues are hypothesized to be less compatible with the local environment and more likely to induce deleterious effects on protein structure and activity. https://zenodo.org/records/20393799.