Sequence-Informed Geometric Evaluation of RNA 3D Structures

Abstract

Computational RNA structure pipelines generate many candidate conformations for the same sequence. Reliable evaluation therefore requires more than recognising plausible geometry, it requires determining whether that geometry is compatible with the sequence. We introduce SIRGE, a sequence-informed geometric evaluator that conditions structural representations on nucleotide embeddings from a pretrained RNA language model. Early results show that SIRGE outperforms established evaluators in Kendall--τ alignment, Top-1 selection, and Top-3 ranking. Controlled comparisons further show that sequence conditioning corrects errors made by an otherwise matched geometric model and improves target-level rank structure. These findings provide initial evidence that pretrained sequence representations supply ranking information that complements geometric reasoning.

Evolutionary Informed Translation from Protein to Nucleic Acid Sequence
Anna Hart, Mohammed El-Kebir, Huimin Zhao +1
arXiv15 Oct 2026PreprintarXiv licenceFull textBiophysicsarXiv
Predicting protein structures at any folding stage from sequence
Lisheng Zhang, Qingsong Li, Zinan Xie +7
arXiv3 Nov 2026PreprintCC BY 4.0Full textBiophysicsarXiv
GeoEPred: A multimodal structure-aware geometric deep learning framework for Gram-negative bacterial secreted effector prediction with sequence semantics
Shouzhen Song, Hua Shi, Hongfeng Wu +5
PLOS Computational Biology30 Sep 2026Peer-reviewedCC BY 4.0Full textMolecular biology & gene regulation Genetics & genomicsPLOS
Constructed Negatives Shape the Evaluation of Sequence-Based Protein–Protein Interaction Models
Enzo Durand, Sergei Grudinin, Béatrice Durand +1
HAL8 Oct 2026PreprintCC BY 4.0Full textGeneral biochemistry Bioinformatics & computational biologyHALbioRxiv
Virtual Screening–Derived Computational Lead Candidates for MAO-B Inhibition in Parkinson’s Disease: Pathway-Informed Docking, Molecular Dynamics, DFT and ADMET Evaluation
Ibrahim Khater, Seifeldin Elabed, Modather F. Hussein +4
International Journal of Molecular Sciences30 Oct 2026Peer-reviewedCC BY 4.0General biochemistryCrossref

Cite

BibTeX RIS

BibTeX

@misc{zerio2026sequenceinformed,
  title = {{Sequence-Informed Geometric Evaluation of RNA 3D Structures}},
  author = {Andrea Zerio and Yighua Yao and Alessandro Micheli and Roland G. Huber and Mile Sikic and Samir Bhatt and Andres R. Masegosa and Yuangang Pan},
  howpublished = {arXiv},
  year = {2026},
  month = sep,
  doi = {10.48550/arxiv.2609.10644},
  eprint = {2609.10644},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2609.10644},
  note = {CC BY 4.0},
}

RIS

TY  - UNPB
TI  - Sequence-Informed Geometric Evaluation of RNA 3D Structures
AU  - Andrea Zerio
AU  - Yighua Yao
AU  - Alessandro Micheli
AU  - Roland G. Huber
AU  - Mile Sikic
AU  - Samir Bhatt
AU  - Andres R. Masegosa
AU  - Yuangang Pan
PY  - 2026
DA  - 2026/09/09
PB  - arXiv
DO  - 10.48550/arxiv.2609.10644
UR  - https://arxiv.org/abs/2609.10644
AB  - Computational RNA structure pipelines generate many candidate conformations for the same sequence. Reliable evaluation therefore requires more than recognising plausible geometry, it requires determining whether that geometry is compatible with the sequence. We introduce SIRGE, a sequence-informed geometric evaluator that conditions structural representations on nucleotide embeddings from a pretrained RNA language model. Early results show that SIRGE outperforms established evaluators in Kendall--τ alignment, Top-1 selection, and Top-3 ranking. Controlled comparisons further show that sequence conditioning corrects errors made by an otherwise matched geometric model and improves target-level rank structure. These findings provide initial evidence that pretrained sequence representations supply ranking information that complements geometric reasoning.
ER  -

Abstract

Related

Cite