Inferring dynamic information from protein structures by Gaussian integrals and deep learning
Bioinformatics · 2026
DOI: 10.1093/bioinformatics/btag446 ↗Kurz gefasst
Fachartikel in der Zeitschrift Bioinformatics, 2026. Verfasst von Felipe Vilicich und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Systems and Computational Biology, Albert Einstein College of Medicine, Bronx, NY 10461, United States. Hauptthemen laut MEDLINE: Deep Learning, Proteins, Molecular Dynamics Simulation, Computational Biology. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NIH HHS, NIGMS NIH HHS, Albert Einstein College of Medicine.
Abstract (Originalsprache)
Motivation Protein dynamics are central to function, but experiments and molecular dynamics (MD) simulations remain costly, low-throughput, and difficult to compare across protocols. Scalable structure-based methods are needed to infer dynamics from static protein structures. Results We present a deep learning framework that predicts protein dynamics from 30-dimensional Gaussian integral (GI) descriptors of Cα backbone topology. Using 1374 ATLAS protein chains with MD-derived RMSF, GI stratified proteins into fold-relevant clusters enriched for secondary structure, sequence homology, and ECOD families. An attention-based 1D-CNN classified flexible versus non-flexible proteins with test AUC = 0.772 and separated slow-mode– from fast-mode–dominated dynamics with AUC = 0.91. Regression models recovered mean RMSF (Pearson r = 0.72; R² = 0.46) and slow-mode RMSF more accurately (Pearson r = 0.83; R² = 0.62), supporting rapid inference of flexibility and collective-motion bias. Availability and implementation Code and data are available on GitHub at: https://github.com/fvilicich/gaussian_integral/blob/main/gaussian_integral_classification.ipynb.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- Department of Systems and Computational Biology, Albert Einstein College of Medicine, Bronx, NY 10461, United States.
- Departamento de Ciencias Aplicadas, Universidad Siglo 21, Córdoba X5147, Argentina.
- Data Science Institute, Vanderbilt University, Nashville, TN 37212, United States.
- Department of Pathology, Albert Einstein College of Medicine, Bronx, NY 10461, United States.
ORCID-Profile: Shanye Yin, Yinghao Wu
Förderung
- NIH HHS
- NIGMS NIH HHS
- Albert Einstein College of Medicine
- National Institutes of Health
Zitieren
APA
Vilicich, F., Bottino, N., Su, Z., Yin, S., & Wu, Y. (2026). Inferring dynamic information from protein structures by Gaussian integrals and deep learning. Bioinformatics. https://doi.org/10.1093/bioinformatics/btag446
Vancouver
Vilicich F, Bottino N, Su Z, Yin S, Wu Y. Inferring dynamic information from protein structures by Gaussian integrals and deep learning. Bioinformatics. 2026. doi:10.1093/bioinformatics/btag446
BibTeX
@article{Vilicich2026,
author = {Vilicich, Felipe and Bottino, Nicolás and Su, Zhaoqian and Yin, Shanye and Wu, Yinghao},
title = {Inferring dynamic information from protein structures by Gaussian integrals and deep learning},
journal = {Bioinformatics},
year = {2026},
doi = {10.1093/bioinformatics/btag446}
}
EUVIMED