Resolving power as a measure of protein stability prediction
- 1Joint Research Centre, European Commission
Abstract
Rank correlation and discriminative resolution are distinct properties of a stability predictor, yet current benchmarks measure only the first. Here we introduce resolving power: the {Delta}{Delta}G separation at which a predictor orders two variants correctly three times in four. Applied to BioEmu across twelve wild-type domains and three supervised regressors on an independent twenty-seven, resolving powers cluster near 1 kcal mol -1 --- comparable to the median variant-pair separation the benchmarks ask these models to rank, so most pairs cannot be ordered. A closed form ties this resolution to the calibration slope ; Spearman correlations account for only 31% of its variance. BioEmu extracts {Delta}G from the ratio of folded and unfolded populations, so what it samples of each becomes a question. Its folded populations reproduce experimental radii of gyration within half an [a]ngstrom, whereas those it labels unfolded are collapsed chains roughly 20 to 40% below a sequence-specific random-coil reference. The ensemble still carries variant differences on panels where the sampled unfolded fraction approaches zero, but the log-odds counting readout runs out of range. Scored by a learned energy, those same conformations recover resolution on ten of twelve panels, with the gain separated from noise on four, without retraining.
Related
- Resolving context-specific protein-protein interactomes forbiological discovery and therapeutic target prioritisation
- A high-throughput genetic screening protocol to measure lipid bilayer stress-induced unfolded protein response in Saccharomyces cerevisiae
- Resolving the mitigation-security-cost trilemma in the power sector through diversified CDR portfolios
- Clinical-Anchored Incremental Survival Learning: A Stability-Constrained Framework for Molecularly Enhanced Survival Prediction in Lung Adenocarcinoma
- Beyond prediction accuracy by integrating GWAASB into wheat stability assessment across environments in Saudi Arabia
BibTeX
@misc{ronco2026resolving,
title = {{Resolving power as a measure of protein stability prediction}},
author = {M. Ronco},
howpublished = {bioRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.18.752669},
url = {https://www.biorxiv.org/content/10.64898/2026.09.18.752669v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Resolving power as a measure of protein stability prediction
AU - M. Ronco
PY - 2026
DA - 2026/09/18
PB - bioRxiv
DO - 10.64898/2026.09.18.752669
UR - https://www.biorxiv.org/content/10.64898/2026.09.18.752669v1
AB - Rank correlation and discriminative resolution are distinct properties of a stability predictor, yet current benchmarks measure only the first. Here we introduce resolving power: the {Delta}{Delta}G separation at which a predictor orders two variants correctly three times in four. Applied to BioEmu across twelve wild-type domains and three supervised regressors on an independent twenty-seven, resolving powers cluster near 1 kcal mol -1 --- comparable to the median variant-pair separation the benchmarks ask these models to rank, so most pairs cannot be ordered. A closed form ties this resolution to the calibration slope ; Spearman correlations account for only 31% of its variance. BioEmu extracts {Delta}G from the ratio of folded and unfolded populations, so what it samples of each becomes a question. Its folded populations reproduce experimental radii of gyration within half an [a]ngstrom, whereas those it labels unfolded are collapsed chains roughly 20 to 40% below a sequence-specific random-coil reference. The ensemble still carries variant differences on panels where the sampled unfolded fraction approaches zero, but the log-odds counting readout runs out of range. Scored by a learned energy, those same conformations recover resolution on ten of twelve panels, with the gain separated from noise on four, without retraining.
ER -