Learning to Align Molecules and Proteins: A Geometry-Aware Approach to Binding Affinity
This work addresses the challenge of accelerating drug discovery by providing more robust affinity predictions, though it appears incremental as it builds on existing deep learning methods with specific enhancements.
The paper tackles the problem of predicting drug-target binding affinity by introducing FIRM-DTI, a lightweight framework that uses feature-wise linear modulation and metric learning to improve generalization, achieving state-of-the-art performance on the Therapeutics Data Commons DTI-DG benchmark.
Accurate prediction of drug-target binding affinity can accelerate drug discovery by prioritizing promising compounds before costly wet-lab screening. While deep learning has advanced this task, most models fuse ligand and protein representations via simple concatenation and lack explicit geometric regularization, resulting in poor generalization across chemical space and time. We introduce FIRM-DTI, a lightweight framework that conditions molecular embeddings on protein embeddings through a feature-wise linear modulation (FiLM) layer and enforces metric structure with a triplet loss. An RBF regression head operating on embedding distances yields smooth, interpretable affinity predictions. Despite its modest size, FIRM-DTI achieves state-of-the-art performance on the Therapeutics Data Commons DTI-DG benchmark, as demonstrated by an extensive ablation study and out-of-domain evaluation. Our results underscore the value of conditioning and metric learning for robust drug-target affinity prediction.