# DyVarMap: Integrating Conformational Dynamics and Interpretable Machine Learning for Cancer-Associated Missense Variant Classification in FGFR2

## TL;DR

DyVarMap combines structural dynamics and machine learning to better classify cancer-related genetic variants in FGFR2, offering clearer explanations for their impact.

## Contribution

DyVarMap introduces a novel framework integrating conformational dynamics and interpretable machine learning for variant classification in FGFR2.

## Key findings

- DyVarMap achieves an AUROC of 0.77 with better calibration than existing tools like PolyPhen-2 and AlphaMissense.
- Key predictors include K659–E565 salt-bridge distance and DFG motif dihedral angles, linking predictions to activation mechanisms.
- Case studies show DyVarMap provides structurally coherent explanations for borderline variants like A628T, E608K, and L618F.

## Abstract

Accurate interpretation of missense variants in cancer-associated genes remains a critical challenge in precision oncology, as most sequence-based predictors lack mechanistic explanations. Receptor tyrosine kinases like FGFR2 exemplify this problem: their function depends on conformational dynamics, yet most variants remain classified as variants of uncertain significance (VUS). In this paper we present DyVarMap, an interpretable structural-learning framework that integrates AlphaFold2-based ensemble generation with physics-driven refinement, manifold learning, and supervised classification using five biophysically motivated geometric features. Applied to FGFR2, the framework generates diverse conformational ensembles, identifies metastable states through nonlinear dimensionality reduction, and classifies pathogenicity while providing mechanistic attributions via SHAP analysis. External validation on ten kinase-domain variants yields an AUROC of 0.77 with superior calibration (Brier score = 0.108) compared to PolyPhen-2 (0.125) and AlphaMissense (0.132). Feature importance analysis consistently identifies K659–E565 salt-bridge distance and DFG motif dihedral angles as top predictors, directly linking predictions to known activation mechanisms. Case studies of borderline variants (A628T, E608K, L618F) demonstrate the framework’s ability to provide structurally coherent mechanistic explanations. DyVarMap bridges the gap between static structure prediction and dynamics-aware functional assessment, generating testable hypotheses for experimental validation and demonstrating the value of incorporating conformational dynamics into variant effect prediction for precision oncology.