Towards molecular structure discovery from cryo-ET density volumes via modelling auxiliary semantic prototypes
Briefings in Bioinformatics · 22.11.2024 · 26 (1)
DOI: 10.1093/bib/bbae570 ↗Kurz gefasst
Fachartikel in der Zeitschrift Briefings in Bioinformatics, online erschienen am 18.01.2025. Verfasst von Ashwin Nair und 8 weiteren Autor:innen. Beteiligte Einrichtung: Department of Data Science, Indian Institute of Science Education and Research, Vithura, 695551, Kerela, India. Hauptthemen laut MEDLINE: Cryoelectron Microscopy, Semantics, Electron Microscope Tomography, Imaging, Three-Dimensional. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 47 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: NIGMS NIH HHS.
Abstract (Originalsprache)
Cryo-electron tomography (cryo-ET) is confronted with the intricate task of unveiling novel structures. General class discovery (GCD) seeks to identify new classes by learning a model that can pseudo-label unannotated (novel) instances solely using supervision from labeled (base) classes. While 2D GCD for image data has made strides, its 3D counterpart remains unexplored. Traditional methods encounter challenges due to model bias and limited feature transferability when clustering unlabeled 2D images into known and potentially novel categories based on labeled data. To address this limitation and extend GCD to 3D structures, we propose an innovative approach that harnesses a pretrained 2D transformer, enriched by an effective weight inflation strategy tailored for 3D adaptation, followed by a decoupled prototypical network. Incorporating the power of pretrained weight-inflated Transformers, we further integrate CLIP, a vision-language model to incorporate textual information. Our method synergizes a graph convolutional network with CLIP’s frozen text encoder, preserving class neighborhood structure. In order to effectively represent unlabeled samples, we devise semantic distance distributions, by formulating a bipartite matching problem for category prototypes using a decoupled prototypical network. Empirical results unequivocally highlight our method’s potential in unveiling hitherto unknown structures in cryo-ET. By bridging the gap between 2D GCD and the distinctive challenges of 3D cryo-ET data, our approach paves novel avenues for exploration and discovery in this domain.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Data Science, Indian Institute of Science Education and Research, Vithura, 695551, Kerela, India.
- Computational Biology Department, Carnegie Mellon University, Pittsburgh, PA, 15213, USA.
- Machine Learning and Visual Computing Lab, Indian Institute of Technology Bombay, Powai, 400076, Maharashtra, India.
- Computer Science and Automation, Indian Institute of Science, CV Raman Rd, 560012, Karnataka, India.
- Computer Science and Engineering, LNM Institute of Information Technology, Jaipur, 302031, Rajasthan, India.
Förderung
- NIGMS NIH HHS
Zitieren
APA
Nair, A., Li, X., Solanki, B., Mukhopadhyay, S., Jha, A., Rafid Uddin, M., Singha, M., Banerjee, B., & Xu, M. (2024). Towards molecular structure discovery from cryo-ET density volumes via modelling auxiliary semantic prototypes. Briefings in Bioinformatics, 26(1). https://doi.org/10.1093/bib/bbae570
Vancouver
Nair A, Li X, Solanki B, Mukhopadhyay S, Jha A, Rafid Uddin M, et al. Towards molecular structure discovery from cryo-ET density volumes via modelling auxiliary semantic prototypes. Briefings in Bioinformatics. 2024;26(1). doi:10.1093/bib/bbae570
BibTeX
@article{Nair2024,
author = {Nair, Ashwin and Li, Xingjian and Solanki, Bhupendra and Mukhopadhyay, Souradeep and Jha, Ankit and Rafid Uddin, Mostofa and Singha, Mainak and Banerjee, Biplab and Xu, Min},
title = {Towards molecular structure discovery from cryo-ET density volumes via modelling auxiliary semantic prototypes},
journal = {Briefings in Bioinformatics},
year = {2024},
volume = {26},
number = {1},
doi = {10.1093/bib/bbae570}
}
EUVIMED