# PLXFPred: interpretable cross-attention networks with hierarchical fusion of multi-modal features for predicting protein–ligand interactions and affinities

## TL;DR

PLXFPred is a new model that improves predictions of how proteins and drugs interact, using advanced deep learning techniques and offering better accuracy and interpretability.

## Contribution

The novel contribution is the cross-modal fusion and hierarchical integration of multi-modal features using interpretable deep learning for protein–ligand interaction prediction.

## Key findings

- PLXFPred reduces prediction errors by over 50% compared to existing models.
- The model provides interpretable insights through attention weights and SHAP analysis.

## Abstract

Accurately predicting protein–ligand interactions and binding affinities is essential for advancing structural biology. Despite recent advancements in deep learning, achieving rapid and precise predictions remains a challenging task. Our approach, Protein–Ligand Cross-Modal Fusion Predictor (PLXFPred), extracts physicochemical properties from amino acid sequences and SMILES. Additionally, it leverages pre-trained models to derive high-dimensional features. GATv2 and BILSTM were used to process the structural and sequence features, respectively. The model’s core involves fusing sequence and graph features via a cross-modal cross-attention mechanism, followed by a multi-modal hierarchical fusion strategy that integrates high-level graph, early fusion, and cross-fusion features. Residual connections and conditional domain adversarial learning improve generalization to previously unseen protein–ligand pairs. Compared to state-of-the-art models, PLXFPred demonstrates superior performance, reducing errors (RMSD, MAE, SD) by over 50%, while providing interpretable biological insights through attention weight visualization and SHAP analysis.

The resource codes are available at https://github.com/xiyuyangtuo/PLXFPred/.