EUVIMEDEUVIMED entdecken
Crossref

MGDM: Molecular generation using a multinomial diffusion model

Sisi Yuan, Chen Zhao, Lin Liu, Guifei Zhou

Methods · 2025-07 · 239 · 1-9

DOI: 10.1016/j.ymeth.2025.03.001 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Methods, 2025-07. Verfasst von Sisi Yuan und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Bioinformatics and Genomics, the University of North Carolina at Charlotte, Charlotte, NC, USA. Electronic address: syuan4@charlotte.ed…. Hauptthemen laut MEDLINE: Drug Design. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 80 Literaturangaben. Gefördert von: Yunnan Provincial Science and Technology Department.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate analysis of molecular structures and the rapid generation of valid molecules remain significant challenges in De Novo drug design. In this study, we propose the Multinomial Generated Diffusion Model (MGDM) for molecular generation. This model leverages a multinomial diffusion framework to process discrete data, with a focus on learning the multinomial distribution inherent in the dataset. During the generation process, the model progressively denoises molecules, transitioning from a uniform noise distribution to ultimately produce valid molecular structures. Initially, we generate molecules unconditionally to expand the compound library. In the next phase, we focus on generating molecules with specific properties to assess the model's capacity for conditional generation. For this, we implement a classifier-free guidance strategy, which directs the diffusion model's task without the need for training separate classifier models. To validate the effectiveness of our framework, we conducted experiments using the Molecular Sets (MOSES) dataset. The results demonstrate that, compared to several state-of-the-art methods, MGDM generates valid molecules while achieving superior or comparable performance in terms of novelty and diversity.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Drug DesignDiffusionAlgorithmsMolecular Structure

Stichwörter der Autor:innen

DenoiseGumbel-Max samplingKullback-Leibler divergenceMolecular generationMultinomial diffusion

Autor:innen und Institutionen

  • Department of Bioinformatics and Genomics, the University of North Carolina at Charlotte, Charlotte, NC, USA. Electronic address: syuan4@charlotte.edu.
  • College of Information Science and Engineering, Hunan University, Changsha, Hunan 410086, PR China. Electronic address: 15536868641@163.com.
  • School of Information Science and Technology, Yunnan Normal University, Kunming 650500, PR China. Electronic address: liulin@ynnu.edu.cn.
  • School of Information Science and Technology, Yunnan Normal University, Kunming 650500, PR China. Electronic address: zhouguifei@ynnu.edu.cn.

Förderung

  • Yunnan Provincial Science and Technology Department

Zitieren

APA

Yuan, S., Zhao, C., Liu, L., & Zhou, G. (2025). MGDM: Molecular generation using a multinomial diffusion model. Methods, 239, 1-9. https://doi.org/10.1016/j.ymeth.2025.03.001

Vancouver

Yuan S, Zhao C, Liu L, Zhou G. MGDM: Molecular generation using a multinomial diffusion model. Methods. 2025;239:1-9. doi:10.1016/j.ymeth.2025.03.001

BibTeX
@article{Yuan2025,
  author = {Yuan, Sisi and Zhao, Chen and Liu, Lin and Zhou, Guifei},
  title = {MGDM: Molecular generation using a multinomial diffusion model},
  journal = {Methods},
  year = {2025},
  volume = {239},
  pages = {1--9},
  doi = {10.1016/j.ymeth.2025.03.001}
}
Kostenlos starten →