HiF-DTA: Hierarchical Feature Learning Network for Drug-Target Affinity Prediction

TL;DR

HiF-DTA introduces a hierarchical deep learning model that captures both global and local features of drugs and proteins at multiple scales, significantly improving drug-target affinity prediction accuracy.

Contribution

The paper presents a novel hierarchical network with dual pathways and multi-scale fusion for enhanced feature extraction in DTA prediction.

Findings

Outperforms state-of-the-art methods on Davis, KIBA, and Metz datasets.

Global-local feature extraction is crucial for accurate DTA prediction.

Multi-scale fusion improves the representation of drugs and proteins.

Abstract

Accurate prediction of Drug-Target Affinity (DTA) is crucial for reducing experimental costs and accelerating early screening in computational drug discovery. While sequence-based deep learning methods avoid reliance on costly 3D structures, they still overlook simultaneous modeling of global sequence semantic features and local topological structural features within drugs and proteins, and represent drugs as flat sequences without atomic-level, substructural-level, and molecular-level multi-scale features. We propose HiF-DTA, a hierarchical network that adopts a dual-pathway strategy to extract both global sequence semantic and local topological features from drug and protein sequences, and models drugs multi-scale to learn atomic, substructural, and molecular representations fused via a multi-scale bilinear attention module. Experiments on Davis, KIBA, and Metz datasets show HiF-DTA outperforms state-of-the-art baselines, with ablations confirming the importance of global-local extraction and multi-scale fusion.