EUVIMEDEUVIMED entdecken
Open Access

From Noise to Knowledge: Diffusion Probabilistic Model-Based Neural Inference of Gene Regulatory Networks

Hao Zhu iD, Donna Slonim

Journal of Computational Biology · 01.11.2024 · 31 (11) · 1087-1103

DOI: 10.1089/cmb.2024.0607 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Computational Biology, erschienen am 01.11.2024. Verfasst von Hao Zhu und 1 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Science, Tufts University, Medford, Massachusetts, USA. Hauptthemen laut MEDLINE: Gene Regulatory Networks, Neural Networks, Computer, Models, Statistical. Die Arbeit wurde bisher 15-mal zitiert (Crossref) und nennt 67 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NICHD NIH HHS.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Understanding gene regulatory networks (GRNs) is crucial for elucidating cellular mechanisms and advancing therapeutic interventions. Original methods for GRN inference from bulk expression data often struggled with the high dimensionality and inherent noise in the data. Here we introduce RegDiffusion, a new class of Denoising Diffusion Probabilistic Models focusing on the regulatory effects among feature variables. RegDiffusion introduces Gaussian noise to the input data following a diffusion schedule and uses a neural network with a parameterized adjacency matrix to predict the added noise. We show that using this process, GRNs can be learned effectively with a surprisingly simple model architecture. In our benchmark experiments, RegDiffusion shows superior performance compared to several baseline methods in multiple datasets. We also demonstrate that RegDiffusion can infer biologically meaningful regulatory networks from real-world single-cell data sets with over 15,000 genes in under 5 minutes. This work not only introduces a fresh perspective on GRN inference but also highlights the promising capacity of diffusion-based models in the area of single-cell analysis. The RegDiffusion software package and experiment data are available at https://github.com/TuftsBCB/RegDiffusion.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Gene Regulatory NetworksNeural Networks, ComputerModels, StatisticalHumansComputational BiologyAlgorithmsSoftwareSingle-Cell AnalysisModels, Genetic

Stichwörter der Autor:innen

network inferencesingle cell analysis and regulatory system

Autor:innen und Institutionen

  • Department of Computer Science, Tufts University, Medford, Massachusetts, USA.

ORCID-Profile: Hao Zhu

Förderung

  • NICHD NIH HHS

Zitieren

APA

Zhu, H., & Slonim, D. (2024). From Noise to Knowledge: Diffusion Probabilistic Model-Based Neural Inference of Gene Regulatory Networks. Journal of Computational Biology, 31(11), 1087-1103. https://doi.org/10.1089/cmb.2024.0607

Vancouver

Zhu H, Slonim D. From Noise to Knowledge: Diffusion Probabilistic Model-Based Neural Inference of Gene Regulatory Networks. Journal of Computational Biology. 2024;31(11):1087-1103. doi:10.1089/cmb.2024.0607

BibTeX
@article{Zhu2024,
  author = {Zhu, Hao and Slonim, Donna},
  title = {From Noise to Knowledge: Diffusion Probabilistic Model-Based Neural Inference of Gene Regulatory Networks},
  journal = {Journal of Computational Biology},
  year = {2024},
  volume = {31},
  number = {11},
  pages = {1087--1103},
  doi = {10.1089/cmb.2024.0607}
}
Kostenlos starten →