Unified modeling of mutation-induced protein stability responses with OmegaTherm
- 1Central South University
Abstract
Accurate prediction of mutation-induced protein stability responses, including changes in folding free energy and melting temperature, is critical for protein engineering. Existing models represent these related but non-equivalent measurements separately and often focus on single mutations. Here we present OmegaTherm, a sequence-based framework that jointly learns both stability responses for single and multiple mutations. OmegaTherm introduces a shared protein language model to capture transferable mutation context, measurement-specific encoders to preserve measurement identity, and paired-measurement-informed distillation to learn complementary knowledge from heterogeneous data. An antisymmetric prediction architecture further enforces physical consistency between forward and reverse mutations. Across independent benchmarks, OmegaTherm achieves state-of-the-art performance in quantitative prediction, mutation classification and protein-level ranking. Beyond benchmark prediction, OmegaTherm-derived stability landscapes correlate strongly with fitness landscapes from deep mutational scanning, enrich favorable variants and reveal function-associated mutation-sensitive regions. These results establish unified representation learning as an effective strategy for integrating related but non-equivalent biochemical measurements.
Related
- A premature ovarian insufficiency-associated LMD-3 point mutation impairs protein stability and causes autophagic-lysosomal dysfunction in Caenorhabditis elegans
- Connectome-Based Modeling of Mutation-Specific Amyloid-β Aggregation in Familial Alzheimer's Disease
- UniHAR: A unified hierarchical transformer with multi-scale temporal modeling for wearable human activity recognition
- LifeLong Digital Twin: A Unified Modeling Paradigm and Agent Harness for Event-Driven Lifelong Health State Trajectories
- A context-aware temporal modeling through unified multi-scale temporal encoding and hierarchical sequence learning for single-channel EEG sleep staging
BibTeX
@misc{wang2026unified,
title = {{Unified modeling of mutation-induced protein stability responses with OmegaTherm}},
author = {W. Wang and Y. Zhou and W. Yu and Y. Wu and Q. Yang and R. Zheng and X. Huang and M. Li and Y. Zhang},
howpublished = {bioRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.23.753709},
url = {https://www.biorxiv.org/content/10.64898/2026.09.23.753709v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - Unified modeling of mutation-induced protein stability responses with OmegaTherm
AU - W. Wang
AU - Y. Zhou
AU - W. Yu
AU - Y. Wu
AU - Q. Yang
AU - R. Zheng
AU - X. Huang
AU - M. Li
AU - Y. Zhang
PY - 2026
DA - 2026/09/23
PB - bioRxiv
DO - 10.64898/2026.09.23.753709
UR - https://www.biorxiv.org/content/10.64898/2026.09.23.753709v1
AB - Accurate prediction of mutation-induced protein stability responses, including changes in folding free energy and melting temperature, is critical for protein engineering. Existing models represent these related but non-equivalent measurements separately and often focus on single mutations. Here we present OmegaTherm, a sequence-based framework that jointly learns both stability responses for single and multiple mutations. OmegaTherm introduces a shared protein language model to capture transferable mutation context, measurement-specific encoders to preserve measurement identity, and paired-measurement-informed distillation to learn complementary knowledge from heterogeneous data. An antisymmetric prediction architecture further enforces physical consistency between forward and reverse mutations. Across independent benchmarks, OmegaTherm achieves state-of-the-art performance in quantitative prediction, mutation classification and protein-level ranking. Beyond benchmark prediction, OmegaTherm-derived stability landscapes correlate strongly with fitness landscapes from deep mutational scanning, enrich favorable variants and reveal function-associated mutation-sensitive regions. These results establish unified representation learning as an effective strategy for integrating related but non-equivalent biochemical measurements.
ER -