Evidence-resolved benchmarks separate biological context recognition from direct-partner specificity in protein interaction models
- 1Peking University
Abstract
Sequence-based models enable scalable protein-protein interaction (PPI) prediction, but whether high scores identify direct physical partners or broader biological compatibility remains unclear. Here we show that biological-context recognition and direct-partner specificity are separable prediction capabilities. We introduce ProtNegDB (http://protnegdb.bjmu.edu.cn/), an evidence-resolved resource distinguishing experimentally supported non-binding from coordinate-verified noncontact within PDB biological assemblies. Using context-matched benchmarks, we find that five sequence-based predictors show limited ability to distinguish direct partners from noncontacting members of the same assembly, whereas PLM-Interact readily ranks both above unrelated proteins. Representation analyses reveal that adaptation to PPI prediction can strengthen biological-context encoding without yielding transferable direct-contact specificity. AlphaFold3 shows greater, though still incomplete, direct-partner resolution, suggesting that explicit multimer modeling can help refine partner specificity. In PLM-Interact, negatives defined by reconstituted direct-binding, cellular binary/proximity and co-complex assays occupy distinct score regimes. Across independent datasets and multiple models, experimental direct non-binding and assembly noncontact further show reproducible, model-dependent scoring patterns. High sequence-based PPI scores should not be interpreted as direct-interface probabilities. Mechanistically interpretable evaluation instead requires explicit consideration of interaction-evidence semantics and partner specificity.
Related
- Joint protein-DNA language modeling enables prediction and engineering of DNA recognition specificity
- Bridging the gap between omics and structural data: A framework for interpreting protein-RNA interaction specificity
- Anterior temporal lobes show route-dependent sensitivity to conceptual specificity: Evidence from MEG and EEG
- Antigen recognition licenses therapeutic CD8+ T cells to organize a multicellular myeloid interaction network in the tumor microenvironment
- Processing abstract concepts through interaction: effects on social recognition
BibTeX
@misc{cheng2026evidenceresolved,
title = {{Evidence-resolved benchmarks separate biological context recognition from direct-partner specificity in protein interaction models}},
author = {H. Cheng and C. Sun and C. Ye and C. Wu and D. Zhao},
howpublished = {bioRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.07.757219},
url = {https://www.biorxiv.org/content/10.64898/2026.10.07.757219v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - Evidence-resolved benchmarks separate biological context recognition from direct-partner specificity in protein interaction models
AU - H. Cheng
AU - C. Sun
AU - C. Ye
AU - C. Wu
AU - D. Zhao
PY - 2026
DA - 2026/10/07
PB - bioRxiv
DO - 10.64898/2026.10.07.757219
UR - https://www.biorxiv.org/content/10.64898/2026.10.07.757219v1
AB - Sequence-based models enable scalable protein-protein interaction (PPI) prediction, but whether high scores identify direct physical partners or broader biological compatibility remains unclear. Here we show that biological-context recognition and direct-partner specificity are separable prediction capabilities. We introduce ProtNegDB (http://protnegdb.bjmu.edu.cn/), an evidence-resolved resource distinguishing experimentally supported non-binding from coordinate-verified noncontact within PDB biological assemblies. Using context-matched benchmarks, we find that five sequence-based predictors show limited ability to distinguish direct partners from noncontacting members of the same assembly, whereas PLM-Interact readily ranks both above unrelated proteins. Representation analyses reveal that adaptation to PPI prediction can strengthen biological-context encoding without yielding transferable direct-contact specificity. AlphaFold3 shows greater, though still incomplete, direct-partner resolution, suggesting that explicit multimer modeling can help refine partner specificity. In PLM-Interact, negatives defined by reconstituted direct-binding, cellular binary/proximity and co-complex assays occupy distinct score regimes. Across independent datasets and multiple models, experimental direct non-binding and assembly noncontact further show reproducible, model-dependent scoring patterns. High sequence-based PPI scores should not be interpreted as direct-interface probabilities. Mechanistically interpretable evaluation instead requires explicit consideration of interaction-evidence semantics and partner specificity.
ER -