Representation-aware calibration enables statistically reliable protein structure search
- 1University of Wisconsin - Madison
Abstract
Searching large databases of predicted protein structures can reveal remote homologs, but reliable discovery requires accurate significance estimates. We show that native E-values reported by protein structure search tools such as Foldseek and Reseek can be unreliable, leading to false discoveries or missed homologs. To address this problem, we introduce Struct-Caliper, which provides reliable query-specific p-values for protein structure search. The method uses representation-aware query randomization to generate background queries and compares observed alignment scores with the highest scores from randomized searches of the same database. In simulations, SG p-values generally kept reported false positive at or below the chosen cutoff under the evaluated randomization procedures. In searches of 15,177 ASTRAL40 structures, SG p-values increased Foldseek homolog recovery by 18.2% with slightly fewer false discoveries compared with native E-values at a cutoff of 0.01. For Reseek, false discoveries even decreased by 99.5%. Across 73 queries searched against more than two million AlphaFoldDB structures, SG p-values increased the mean recovery of DALI-supported matches by 23.5% at the same cutoff. These results show that Struct-Caliper can improve the reliability of structure searches and help prioritize discoveries for further biological analysis.
Related
- Integrating Narrow-Window DIA with AI-Powered Search Enables Deep and Reliable Functional Proteomics
- Representation-aware conformal prediction for reliable uncertainty quantification in drug-target affinity
- Toward Reliable Patient-Specific Aortic Strain Mapping from 4D CTA: Validation, Spectral Structure, and Clinical Potential
- Abstract representation in the hippocampus predicts spontaneously adopted structure-based behavioral strategy
- CUBE: Multimodal Representation Learning Reveals Biological Structure Across Histomorphology, Spatial Protein Phenotypes, and Transcriptome-Associated Signals
BibTeX
@misc{shi2026representationaware,
title = {{Representation-aware calibration enables statistically reliable protein structure search}},
author = {L. Shi and H. Zhou and Y. Yang and Y. Lu},
howpublished = {bioRxiv},
year = {2026},
month = nov,
doi = {10.64898/2026.11.03.764110},
url = {https://www.biorxiv.org/content/10.64898/2026.11.03.764110v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Representation-aware calibration enables statistically reliable protein structure search
AU - L. Shi
AU - H. Zhou
AU - Y. Yang
AU - Y. Lu
PY - 2026
DA - 2026/11/03
PB - bioRxiv
DO - 10.64898/2026.11.03.764110
UR - https://www.biorxiv.org/content/10.64898/2026.11.03.764110v1
AB - Searching large databases of predicted protein structures can reveal remote homologs, but reliable discovery requires accurate significance estimates. We show that native E-values reported by protein structure search tools such as Foldseek and Reseek can be unreliable, leading to false discoveries or missed homologs. To address this problem, we introduce Struct-Caliper, which provides reliable query-specific p-values for protein structure search. The method uses representation-aware query randomization to generate background queries and compares observed alignment scores with the highest scores from randomized searches of the same database. In simulations, SG p-values generally kept reported false positive at or below the chosen cutoff under the evaluated randomization procedures. In searches of 15,177 ASTRAL40 structures, SG p-values increased Foldseek homolog recovery by 18.2% with slightly fewer false discoveries compared with native E-values at a cutoff of 0.01. For Reseek, false discoveries even decreased by 99.5%. Across 73 queries searched against more than two million AlphaFoldDB structures, SG p-values increased the mean recovery of DALI-supported matches by 23.5% at the same cutoff. These results show that Struct-Caliper can improve the reliability of structure searches and help prioritize discoveries for further biological analysis.
ER -