Integrating structural and biological evidence to rerank ESMFold2 protein-protein interactions
- 1State Key Laboratory of Membrane Biology
Abstract
Large-scale protein structure prediction enables proteome-wide protein-protein interaction (PPI) screening, but distinguishing biologically meaningful interactions from spurious interfaces remains challenging. Here, we develop a scalable framework combining fast, MSA-free ESMFold2 prediction with PAE-guided domain parsing and evidence-based reranking. Three-recycle ESMFold2-Fast achieved 57% acceptable-or-better DockQ scores on FoldBench, comparable to AlphaFold2-Multimer while substantially reducing computation. PAE-guided parsing preserved 98.1% of XL-MS cross-links within parsed domain pairs. We then developed the Structure Prediction and Omics informed Classifier (SPOC)-ESMFold2, which integrates structural features with independent biological evidence to prioritize predicted interactions. SPOC-ESMFold2 achieved an AUROC of 0.93 and AUPR of 0.90, compared with 0.87 and 0.79 for a structural-only classifier. Under a 1:128 positive-to-negative ratio, SPOC-ESMFold2 achieved 17.2% recall at 5% false-discovery rate, substantially outperforming structural confidence metrics alone. This framework enables scalable PPI screening by integrating structural plausibility with orthogonal biological evidence to prioritize candidates for experimental investigation.
Related
- Structured biological evidence enables general-purpose language models to predict bacteriophage-host interactions
- Plant–Insect Interactions in Biological Invasions: From Classical Hypotheses to Multitrophic Networks and Biological Control
- Integrating complementary biological information for multi-objective enzyme engineering
- Converging Paths: Artificial Intelligence for Integrating Biological and Digital Biomarkers in Personalized Longevity Medicine
- Proposed Interactions and Crosstalk Among Biological Aging Biomarkers (Telomere Length, Telomerase, Sirtuins, mtDNA Copy Number and mtDNA Methylation Levels) and Their Potential Role in Colorectal Cancer Risk: A Review
BibTeX
@misc{xie2026integrating,
title = {{Integrating structural and biological evidence to rerank ESMFold2 protein-protein interactions}},
author = {J. Xie and M. Li and Y. Chai and G. Ou and W. Li and Z. Guo},
howpublished = {bioRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.16.751194},
url = {https://www.biorxiv.org/content/10.64898/2026.09.16.751194v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Integrating structural and biological evidence to rerank ESMFold2 protein-protein interactions
AU - J. Xie
AU - M. Li
AU - Y. Chai
AU - G. Ou
AU - W. Li
AU - Z. Guo
PY - 2026
DA - 2026/09/16
PB - bioRxiv
DO - 10.64898/2026.09.16.751194
UR - https://www.biorxiv.org/content/10.64898/2026.09.16.751194v1
AB - Large-scale protein structure prediction enables proteome-wide protein-protein interaction (PPI) screening, but distinguishing biologically meaningful interactions from spurious interfaces remains challenging. Here, we develop a scalable framework combining fast, MSA-free ESMFold2 prediction with PAE-guided domain parsing and evidence-based reranking. Three-recycle ESMFold2-Fast achieved 57% acceptable-or-better DockQ scores on FoldBench, comparable to AlphaFold2-Multimer while substantially reducing computation. PAE-guided parsing preserved 98.1% of XL-MS cross-links within parsed domain pairs. We then developed the Structure Prediction and Omics informed Classifier (SPOC)-ESMFold2, which integrates structural features with independent biological evidence to prioritize predicted interactions. SPOC-ESMFold2 achieved an AUROC of 0.93 and AUPR of 0.90, compared with 0.87 and 0.79 for a structural-only classifier. Under a 1:128 positive-to-negative ratio, SPOC-ESMFold2 achieved 17.2% recall at 5% false-discovery rate, substantially outperforming structural confidence metrics alone. This framework enables scalable PPI screening by integrating structural plausibility with orthogonal biological evidence to prioritize candidates for experimental investigation.
ER -