Hierarchical Graph Representation Learning From a Statistical Perspective for Generalizable and Interpretable Protein-Ligand Binding Affinity Prediction.
Jul 2026· IEEE transactions on computational biology and bioinformatics· Vol PP· 0 citations
Medicine
Abstract
Protein-ligand binding affinity (PLA) prediction aims to guide rational drug design by estimating the strength of interaction. The effectiveness of the representation learning of protein and ligand is key to successful PLA prediction. To this end, attention mechanism, as a powerful architectural paradigm, has been introduced and gradually emerged as the prevailing approach. However, intuitively, the classical attention paradigm based on similarity does not fit the biological mechanisms relevant for binding. Worse still, the cooperative and antagonistic effects among multiple atoms are deliberately disregarded in the classical formulation of attention mechanisms. Consequently, the rigid transplantation of classical architectures substantially undermines the PLA prediction performance. To address these challenges, we employ a hierarchical statistical attention model (HISA). Specifically, HISA employs a statistical attention mechanism (SAM) based on non-similarity computation to fit the biological prior and perceive the relationship of multiple atoms. In addition, we optimize HISA by employing clustering, enabling hierarchical representations of biomolecules. Extensive experiments demonstrate that HISA achieves state-of-the-art performance on multiple PLA benchmarks while simultaneously exhibiting generalizability and interpretability.
MOTIVATION
To enable real-world protein-ligand affinity prediction, not only out-of-distribution generalization but also robustness to variable structural availability and quality should be considered in model design.
RESULTS
We present AlignNet, a hierarchical representation alignment framework that mitigates intra- and inter-molecular heterogeneity to learn robust protein-ligand embeddings for generalizable affinity prediction, even from sequence-level inputs. Its intra-molecular module projects unimodal and multimodal features into a unified space, aligning augmented multimodal views for feature fusion and unimodal with multimodal embeddings to distill multimodal priors for structure-agnostic inference. Its inter-molecular module aligns protein and ligand embeddings for cross-molecular integration. Extensive experiments show that AlignNet (i) achieves highly competitive performance, with up to a 20.4% gain in SCC on the challenging LBA 30% split under sequence-only settings, suggesting improved out-of-distribution generalization; and (ii) learns well-separated affinity-related clusters, supporting reliable structure-independent prediction.
AVAILABILITY AND IMPLEMENTATION
AlignNet is available at https://github.com/altriavin/AlignNet.
SUPPLEMENTARY INFORMATION
Supplementary data are available at Bioinformatics online.
Xiaowen Hu, Hongyi Huang, Hao Sun et al.· Bioinformatics· 0 citations
Quantifying protein–protein binding affinity is essential for understanding molecular recognition and guiding antibody and inhibitor design. However, binding affinity is governed by tightly coupled sequence, structural, and chemical determinants. Existing models often encode these factors in isolation, limiting their ability to capture the multi-level dependencies underlying binding affinity
$$\left(\Delta\text{G}\right)$$
.
We propose MIRAGE, a graph-based framework for direct
$$\Delta \text{G}$$
prediction that explicitly models interactions across multidimensional (1D sequences, 2D contact maps, 3D structures) and multi-scale (residue-level, atom-level) features. MIRAGE integrates two complementary modules to capture cross-dimensional and cross-scale dependencies, enabling unified residue–atom representation learning. Across public benchmarks, MIRAGE demonstrated strong generalization, achieving Pearson correlations of 0.70 and 0.69 on two independent external test sets and retaining predictive effectiveness under structure-separated cross-validation designed to reduce structural information sharing. In a supplementary analysis with AlphaFold3-predicted complex structures, MIRAGE also preserved significant predictive correlations when experimentally resolved structures were unavailable. Ablation studies confirm the contributions of each module. Interpretability analyses further show that the model focuses on biophysically meaningful interface regions. The source code of MIRAGE is available from
https://github.com/ShiweiWu-545/MIRAGE
.
These results indicate that explicitly modeling multi-level interactions is important for accurately capturing the determinants of binding affinity. MIRAGE provides an interpretable and robust framework for structure-aware
$$\Delta \text{G}$$
prediction, with potential applications in protein engineering and drug design.
The resulting model, HydrAffinity, is an interaction-free, dynamic sparse model that uses pre-trained encoders and MoE for parameter-efficient learning and outperforms all interaction-free methods and matches state-of-the-art interaction-based methods on CASF-2016.
Despite challenges related to data sparsity and conformational variability, ViTs show strong performance and high robustness in structure-based affinity prediction tasks, underscore their effectiveness in learning spatial patterns and suggest broader applicability to related tasks, such as protein-protein or protein-nucleic acid interaction modeling.
Jakub Poziemski, Paweł Siedlecki· Scientific Reports· 0 citations
Structure-based drug design (SBDD) models are central to modern pharmaceutical research, enabling the rational exploration of protein-ligand interactions at atomic resolution. However, most existing approaches frame molecular generation as an isolated optimization or a one-to-one matching task, overlooking the shared binding patterns and intrinsic similarities among protein-ligand complexes. This fragmented perspective constrains their ability to capture the fundamental principles governing molecular recognition and binding specificity. Moreover, the limited availability of high-quality experimental data further hampers model generalization and real-world applicability. To address these challenges, we present READ, a retrieval-alignment molecular generation framework that conditions the generative process on small molecules targeting homologous proteins. Retrieved ligands are aligned with a diffusion model across multiple representational spaces and integrated as conditional guidance throughout successive stages of generation. Under a standardized docking-based evaluation protocol, READ achieves consistently strong performance against state-of-the-art SBDD methods. More importantly, it introduces a retrieval-alignment paradigm for structure-based molecular generation, offering a practical framework for early-stage computational hit generation while leaving prospective experimental validation as future work.
Dong Xu, Zhangfan Yang, Junchuang Cai et al.· IEEE transactions on computa...· 1 citation
An Algebraic Graph Neural Network model designed to encode molecular structures into a low-dimensional graph representation while preserving critical biochemical interactions is introduced, demonstrating superior performance in binding affinity prediction compared to state-of-the-art scoring functions.
Augustine Ouru, Xi Chen, Cameron Yeagle et al.· Computational and Mathematic...· 0 citations