What question did this study set out to answer?

To develop an interpretable, graph-based deep learning framework (MIRAGE) that accurately predicts protein–protein binding affinity (ΔG) by modeling cross-dimensional and cross-scale interactions.

August 15, 2026BMC GenomicsOpen Access

Interpretable multilevel interaction modeling for robust protein–protein affinity

View Full Paper

Ask AI

Bookmark

Authors

SWShiwei WuHunan University of Science and Technology HLHaoliang LiuHarbin Engineering University ZHZepeng HuangHarbin Engineering University

Discussion

Member takes

Implication

Machine learning evaluation demonstrates accurate protein–protein binding affinity prediction across multiscale structural features, highlighting utility for antibody and drug design.

Key Points

To develop an interpretable, graph-based deep learning framework (MIRAGE) that accurately predicts protein–protein binding affinity (ΔG) by modeling cross-dimensional and cross-scale interactions.
Designed a dual-module graph framework integrating 1D sequences, 2D contact maps, and 3D structures across both residue- and atom-level scales.
Evaluated performance using public benchmarks, structure-separated cross-validation, and complex structures predicted by AlphaFold3.
Achieved Pearson correlation coefficients of 0.70 and 0.69 on two independent external test sets.
Maintained robust predictive correlation under structure-separated cross-validation and when applied to computationally modeled structures from AlphaFold3.

Cite This Study

Wu et al. (2026) studied this question.

synapsesocial.com/papers/6a8019ce75c2e31742c85ed1 https://doi.org/10.1186/s12864-026-13253-1

View Full Paper

Ask AI

Bookmark

Implication

Machine learning evaluation demonstrates accurate protein–protein binding affinity prediction across multiscale structural features, highlighting utility for antibody and drug design.

Interpretable multilevel interaction modeling for robust protein–protein affinity

Discussion

Member takes

Implication

Key Points

Cite This Study

Interpretable multilevel interaction modeling for robust protein–protein affinity

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider