OmniScore: Universal Scoring of Diverse Biomolecular Complexes via Equivariant Geometry-Aware Discrete Representation Learning
- 1Seoul National University
Abstract
Biomolecular structures provide a common physical basis for understanding molecular function and interactions. However, computational models typically specialize in individual molecular classes or tasks, hindering a unified representation of structural information across biological systems. Here we introduce OmniScore, a universal geometric encoder that learns shared representations of proteins, nucleic acids, small molecules and their complexes. Pretraining integrates geometric graph encoding, sequential modeling and tokenization with structural and physicochemical supervision. The frozen pretrained encoder supports complex-level model quality assessment and interaction prediction, as well as residue-level prediction, through task-specific heads. OmniScore outperforms the evaluated antibody--antigen quality-assessment baselines and also improves RNA structure ranking over baselines. For peptide-major histocompatibility complex interaction prediction, the OmniScore embeddings complement structure-prediction confidence metrics to improve predictive performance. Together, these findings support OmniScore as a shared geometric foundation for biomolecular representation and scoring across molecular classes and prediction scales, which may facilitate multi-component therapeutics development.
Related
- PDI-Spot: A Robust Representation Learning Framework for Identifying Energetic Hotspots in Protein–DNA Complexes
- SARS-CoV-2 Nucleocapsid Protein forms biomolecular condensates and interacts with ALS-associated RNA binding proteins in vitro
- Translational activation: An unforeseen function of RNP biomolecular condensates
- Learning interpretable kinetic models for biomolecular interaction networks
- Astrolabe: Programmable all-atom biomolecular design through cross-context coordination and structural intervention
BibTeX
@misc{bui2026omniscore,
title = {{OmniScore: Universal Scoring of Diverse Biomolecular Complexes via Equivariant Geometry-Aware Discrete Representation Learning}},
author = {T.-C. Bui and Y. Hong and J. Song and Y. Tao and I. Chung and J. G. Park and J. Ko and J. Lee},
howpublished = {bioRxiv},
year = {2026},
month = aug,
doi = {10.64898/2026.08.28.747942},
url = {https://www.biorxiv.org/content/10.64898/2026.08.28.747942v2},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - OmniScore: Universal Scoring of Diverse Biomolecular Complexes via Equivariant Geometry-Aware Discrete Representation Learning
AU - T.-C. Bui
AU - Y. Hong
AU - J. Song
AU - Y. Tao
AU - I. Chung
AU - J. G. Park
AU - J. Ko
AU - J. Lee
PY - 2026
DA - 2026/08/28
PB - bioRxiv
DO - 10.64898/2026.08.28.747942
UR - https://www.biorxiv.org/content/10.64898/2026.08.28.747942v2
AB - Biomolecular structures provide a common physical basis for understanding molecular function and interactions. However, computational models typically specialize in individual molecular classes or tasks, hindering a unified representation of structural information across biological systems. Here we introduce OmniScore, a universal geometric encoder that learns shared representations of proteins, nucleic acids, small molecules and their complexes. Pretraining integrates geometric graph encoding, sequential modeling and tokenization with structural and physicochemical supervision. The frozen pretrained encoder supports complex-level model quality assessment and interaction prediction, as well as residue-level prediction, through task-specific heads. OmniScore outperforms the evaluated antibody--antigen quality-assessment baselines and also improves RNA structure ranking over baselines. For peptide-major histocompatibility complex interaction prediction, the OmniScore embeddings complement structure-prediction confidence metrics to improve predictive performance. Together, these findings support OmniScore as a shared geometric foundation for biomolecular representation and scoring across molecular classes and prediction scales, which may facilitate multi-component therapeutics development.
ER -