Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: May 16, 2025

Curation of Computational Chemical Libraries Demonstrated with Alpha-Amino Acids

Curation of Computational Chemical Libraries Demonstrated with Alpha-Amino Acids

Published on: April 13, 2022

MolEM: a unified generative framework for molecular graphs and sequential orders.

Hanwen Zhang^1,2, Deng Xiong³, Xianggen Liu^1,2

¹College of Computer Science, Sichuan University, No.24 South Section 1, Yihuan Road, Chengdu 610065, China.

Briefings in Bioinformatics

|March 31, 2025

Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Policy-Based Active Learning for Efficient Molecular Identification.

Journal of chemical information and modeling·2026

Same author

From target specificity to metabolic efficiency: Design and optimization of etomidate analogues for potential improvement in postoperative outcomes.

Acta pharmaceutica Sinica. B·2026

Same author

Multimodal Deep Learning with Routine Clinical Data for Recurrence Risk Stratification in HR<sup>+</sup>/HER2<sup>-</sup> Early Breast Cancer.

Research (Washington, D.C.)·2026

Same author

A systematic comparison of methodologies for the estimation of the serial interval.

Infectious Disease Modelling·2026

Same author

Learning With Partial and Noisy Correspondence in Graph Matching.

IEEE transactions on pattern analysis and machine intelligence·2026

Same author

Few-shot molecular property optimization <i>via</i> a domain-specialized large language model.

Chemical science·2026

Same journal

K-attention: a biologically informed attention operator for data-efficient sequence-based omics modeling.

Briefings in bioinformatics·2026

Same journal

Accurate prediction of asparagine deamidation in biologics using advanced machine learning models.

Briefings in bioinformatics·2026

Same journal

scImmuneCo: a compendium of cell-type-specific functional modules for decoding immune responses from single-cell RNA-seq data.

Briefings in bioinformatics·2026

Same journal

scGenoByte: a GenoByte embedding transformer with biological priors for cell type annotation.

Briefings in bioinformatics·2026

Same journal

FerroScore: a statistical approach for quantifying tumor-related ferroptosis based on omics data.

Briefings in bioinformatics·2026

Same journal

METEOR: a data-adaptive Mendelian randomization method for powerful detection of shared and specific exposures underlying multiple outcomes.

Briefings in bioinformatics·2026

See all related articles

This summary is machine-generated.

MolEM jointly learns 3D molecular graphs and their sequential orders, improving drug design. This novel framework generates molecules with high binding affinity and realistic structures, outperforming existing methods.

Area of Science:

Computational chemistry
Drug discovery
Machine learning

Background:

Structure-based drug design aims to create high-affinity molecules for protein targets.
Current sequential generative models struggle with the computational intractability of molecular graph ordering.
Existing methods use fixed ordering schemes, leading to suboptimal generation due to loose likelihood bounds.

Purpose of the Study:

To develop a unified generative framework for joint learning of 3D molecular graphs and their sequential orders.
To address the limitations of fixed ordering schemes in molecular graph generation.
To improve the accuracy and flexibility of ligand conformations in drug design.

Main Methods:

Proposed MolEM, a unified generative framework for joint 3D molecular graph and sequential order learning.

Keywords:

molecular graph generation sequential generative model sequential order structure-based drug design variational expectation-maximization

More Related Videos

Modeling an Enzyme Active Site using Molecular Visualization Freeware

Modeling an Enzyme Active Site using Molecular Visualization Freeware

Published on: December 25, 2021

Interactive Molecular Model Assembly with 3D Printing

Interactive Molecular Model Assembly with 3D Printing

Published on: August 13, 2020

Related Experiment Videos

Last Updated: May 16, 2025

Curation of Computational Chemical Libraries Demonstrated with Alpha-Amino Acids

Curation of Computational Chemical Libraries Demonstrated with Alpha-Amino Acids

Published on: April 13, 2022

Modeling an Enzyme Active Site using Molecular Visualization Freeware

Modeling an Enzyme Active Site using Molecular Visualization Freeware

Published on: December 25, 2021

Interactive Molecular Model Assembly with 3D Printing

Interactive Molecular Model Assembly with 3D Printing

Published on: August 13, 2020

Derived and maximized a tight lower bound of the likelihood using a variational expectation-maximization algorithm.

Incorporated QuickVina 2 for molecular docking to optimize binding poses and ligand conformations.

Main Results:

MolEM significantly outperforms baseline models in generating molecules with high binding affinities.
The framework produces molecules with realistic structures.
MolEM efficiently approximates true marginal graph likelihood and identifies chemically relevant orderings.

Conclusions:

MolEM offers a novel approach to learning-based ordering schemes for 3D molecular graph generation.
The joint learning of graphs and orders improves molecular generation quality.
This method advances structure-based drug design by enabling more accurate and flexible ligand conformations.