Phi-Former: A Pairwise Hierarchical Approach for Compound-Protein Interactions Prediction

• URL: http://arxiv.org/abs/2602.05479v1
• Date: Thu, 05 Feb 2026 09:39:22 GMT
• Title: Phi-Former: A Pairwise Hierarchical Approach for Compound-Protein Interactions Prediction
• Authors: Zhe Wang, Zijing Liu, Chencheng Xu, Yuan Yao,
• Abstract summary: Drug discovery remains time-consuming, labor-intensive, and expensive.<n>Predicting compound-protein interactions (CPIs) is a critical component in this process.<n>Recent deep learning methods have successfully modeled CPIs at the atomic level.<n>We propose Phi-former, a pairwise hierarchical interaction representation learning method.
• Score: 12.813544613908588
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Drug discovery remains time-consuming, labor-intensive, and expensive, often requiring years and substantial investment per drug candidate. Predicting compound-protein interactions (CPIs) is a critical component in this process, enabling the identification of molecular interactions between drug candidates and target proteins. Recent deep learning methods have successfully modeled CPIs at the atomic level, achieving improved efficiency and accuracy over traditional energy-based approaches. However, these models do not always align with chemical realities, as molecular fragments (motifs or functional groups) typically serve as the primary units of biological recognition and binding. In this paper, we propose Phi-former, a pairwise hierarchical interaction representation learning method that addresses this gap by incorporating the biological role of motifs in CPIs. Phi-former represents compounds and proteins hierarchically and employs a pairwise pre-training framework to model interactions systematically across atom-atom, motif-motif, and atom-motif levels, reflecting how biological systems recognize molecular partners. We design intra-level and inter-level learning pipelines that make different interaction levels mutually beneficial. Experimental results demonstrate that Phi-former achieves superior performance on CPI-related tasks. A case study shows that our method accurately identifies specific atoms or motifs activated in CPIs, providing interpretable model explanations. These insights may guide rational drug design and support precision medicine applications.

Related papers

• ReACT-Drug: Reaction-Template Guided Reinforcement Learning for de novo Drug Design [0.34155322317700576]
  We introduce bfReACT-Drug, a fully integrated, target-agnostic molecular design framework based on Reinforcement Learning.<n>This architecture highlights the potential of integrating structural biology, deep representation learning, and chemical rules to automate and accelerate rational drug design.
  arXiv  Detail & Related papers  (2025-12-24T05:29:35Z)
• Learning Cell-Aware Hierarchical Multi-Modal Representations for Robust Molecular Modeling [74.25438319700929]
  We propose CHMR (Cell-aware Hierarchical Multi-modal Representations), a robust framework that models local-global dependencies between molecules and cellular responses.<n> evaluated on nine public benchmarks spanning 728 tasks, CHMR outperforms state-of-the-art baselines.<n>Results demonstrate the advantage of hierarchy-aware, multimodal learning for reliable and biologically grounded molecular representations.
  arXiv  Detail & Related papers  (2025-11-26T07:15:00Z)
• Towards Interpretable Drug-Drug Interaction Prediction: A Graph-Based Approach with Molecular and Network-Level Explanations [3.6099926707292793]
  Drug-drug interactions (DDIs) represent a critical challenge in pharmacology, often leading to adverse drug reactions with significant implications for patient safety and healthcare outcomes.<n>We propose MolecBioNet, a novel graph-based framework that integrates molecular and biomedical knowledge for robust and interpretable DDI prediction.
  arXiv  Detail & Related papers  (2025-07-12T07:43:19Z)
• PRING: Rethinking Protein-Protein Interaction Prediction from Pairs to Graphs [88.98041407783502]
  PRING is the first benchmark that evaluates protein-protein interaction prediction from a graph-level perspective.<n> PRING curates a high-quality, multi-species PPI network dataset comprising 21,484 proteins and 186,818 interactions.
  arXiv  Detail & Related papers  (2025-07-07T15:21:05Z)
• A Generalist Cross-Domain Molecular Learning Framework for Structure-Based Drug Discovery [32.573496601865465]
  Structure-based drug discovery (SBDD) is a systematic scientific process that develops new drugs by leveraging the detailed physical structure of the target protein.<n>Recent advancements in pre-trained models for biomolecules have demonstrated remarkable success across various biochemical applications.
  arXiv  Detail & Related papers  (2025-03-06T12:04:56Z)
• The Signed Two-Space Proximity Model for Learning Representations in Protein-Protein Interaction Networks [16.396309363020908]
  Accurately predicting complex protein-protein interactions (PPIs) is crucial for decoding biological processes.<n>We present the Signed Two-Space Proximity Model (S2-SPM) for signed PPI networks, which explicitly incorporates both types of interactions.<n>Our approach also enables the identification of archetypes representing extreme protein profiles.
  arXiv  Detail & Related papers  (2025-03-05T21:08:58Z)
• Binding Affinity Prediction: From Conventional to Machine Learning-Based Approaches [48.66541987908136]
  Much work has been devoted to predicting binding affinity over the past decades.<n>We note growing use of both traditional machine learning and deep learning models for predicting binding affinity.<n>With improved predictive performance and the FDA's phasing out of animal testing, AI-driven in silico models, such as AI virtual cells (AIVCs), are poised to advance binding affinity prediction.
  arXiv  Detail & Related papers  (2024-09-30T03:40:49Z)
• Exploiting Hierarchical Interactions for Protein Surface Learning [52.10066114039307]
  Intrinsically, potential function sites in protein surfaces are determined by both geometric and chemical features. In this paper, we present a principled framework based on deep learning techniques, namely Hierarchical Chemical and Geometric Feature Interaction Network (HCGNet) Our method outperforms the prior state-of-the-art method by 2.3% in site prediction task and 3.2% in interaction matching task.
  arXiv  Detail & Related papers  (2024-01-17T14:10:40Z)
• A biologically-inspired evaluation of molecular generative machine learning [17.623886600638716]
  A novel biologically-inspired benchmark for the evaluation of molecular generative models is proposed. We propose a recreation metric, apply drug-target affinity prediction and molecular docking as complementary techniques for the evaluation of generative outputs.
  arXiv  Detail & Related papers  (2022-08-20T11:01:10Z)
• Improved Drug-target Interaction Prediction with Intermolecular Graph Transformer [98.8319016075089]
  We propose a novel approach to model intermolecular information with a three-way Transformer-based architecture. Intermolecular Graph Transformer (IGT) outperforms state-of-the-art approaches by 9.1% and 20.5% over the second best for binding activity and binding pose prediction respectively. IGT exhibits promising drug screening ability against SARS-CoV-2 by identifying 83.1% active drugs that have been validated by wet-lab experiments with near-native predicted binding poses.
  arXiv  Detail & Related papers  (2021-10-14T13:28:02Z)
• Multi-View Self-Attention for Interpretable Drug-Target Interaction Prediction [4.307720252429733]
  In machine learning approaches, the numerical representation of molecules is critical to the performance of the model. We propose a self-attention-based multi-view representation learning approach for modeling drug-target interactions.
  arXiv  Detail & Related papers  (2020-05-01T14:28:17Z)
• Explainable Deep Relational Networks for Predicting Compound-Protein Affinities and Contacts [80.69440684790925]
  DeepRelations is a physics-inspired deep relational network with intrinsically explainable architecture. It shows superior interpretability to the state-of-the-art. It boosts the AUPRC of contact prediction 9.5, 16.9, 19.3 and 5.7-fold for the test, compound-unique, protein-unique, and both-unique sets.
  arXiv  Detail & Related papers  (2019-12-29T00:14:07Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.