LaPro-DTA: Latent Dual-View Drug Representations and Salient Protein Feature Extraction for Generalizable Drug--Target Affinity Prediction
This addresses a critical bottleneck in drug discovery by improving generalizability for unseen drugs and targets, though it is an incremental advance over existing methods.
The paper tackled the problem of performance degradation in drug-target affinity prediction under cold-start scenarios by proposing LaPro-DTA, which achieved an 8% MSE reduction on the Davis dataset in the unseen-drug setting.
Drug--target affinity prediction is pivotal for accelerating drug discovery, yet existing methods suffer from significant performance degradation in realistic cold-start scenarios (unseen drugs/targets/pairs), primarily driven by overfitting to training instances and information loss from irrelevant target sequences. In this paper, we propose LaPro-DTA, a framework designed to achieve robust and generalizable DTA prediction. To tackle overfitting, we devise a latent dual-view drug representation mechanism. It synergizes an instance-level view to capture fine-grained substructures with stochastic perturbation and a distribution-level view to distill generalized chemical scaffolds via semantic remapping, thereby enforcing the model to learn transferable structural rules rather than memorizing specific samples. To mitigate information loss, we introduce a salient protein feature extraction strategy using pattern-aware top-$k$ pooling, which effectively filters background noise and isolates high-response bioactive regions. Furthermore, a cross-view multi-head attention mechanism fuses these purified features to model comprehensive interactions. Extensive experiments on benchmark datasets demonstrate that LaPro-DTA significantly outperforms state-of-the-art methods, achieving an 8\% MSE reduction on the Davis dataset in the challenging unseen-drug setting, while offering interpretable insights into binding mechanisms.