Deep generative models for 3D molecular structure

被引:19
作者
Baillif, Benoit [1 ]
Cole, Jason [2 ]
McCabe, Patrick [2 ]
Bender, Andreas [1 ]
机构
[1] Univ Cambridge, Yusuf Hamied Dept Chem, Lensfield Rd, Cambridge CB2 1EW, England
[2] Cambridge Crystallog Data Ctr, 12 Union Rd, Cambridge CB2 1EZ, England
关键词
DESIGN; LANGUAGE; DOCKING;
D O I
10.1016/j.sbi.2023.102566
中图分类号
Q5 [生物化学]; Q7 [分子生物学];
学科分类号
071010 ; 081704 ;
摘要
Deep generative models have gained recent popularity for chemical design. Many of these models have historically operated in 2D space; however, more recently explicit 3D molecular generative models have become of interest, which are the topic of this article. Dozens of published models have been developed in the last few years to generate molecules directly in 3D, outputting both the atom types and coordinates, either in oneshot or adding atoms or fragments step-by-step. These 3D generative models can also be guided by structural information such as a binding pocket representation to successfully generate molecules with docking score ranges similar to known actives, but still showing lower computational efficiency and generation throughput than 1D/2D generative models and sometimes producing unrealistic conformations. We advocate for a unified benchmark of metrics to evaluate generation and propose perspectives to be addressed in next implementations.
引用
收藏
页数:10
相关论文
共 77 条
[1]  
Arcidiacono M, 2021, arXiv, DOI 10.48550/arXiv.2109.15308
[2]   GEOM, energy-annotated molecular conformations for property prediction and molecular generation [J].
Axelrod, Simon ;
Gomez-Bombarelli, Rafael .
SCIENTIFIC DATA, 2022, 9 (01)
[3]  
Belgrave Danielle, 2022, ADV NEUR IN
[4]   REINVENT 2.0: An AI Tool for De Novo Drug Design [J].
Blaschke, Thomas ;
Arus-Pous, Josep ;
Chen, Hongming ;
Margreitter, Christian ;
Tyrchan, Christian ;
Engkvist, Ola ;
Papadopoulos, Kostas ;
Patronov, Atanas .
JOURNAL OF CHEMICAL INFORMATION AND MODELING, 2020, 60 (12) :5918-5922
[5]   ProLIF: a library to encode molecular interactions as fingerprints [J].
Bouysset, Cedric ;
Fiorucci, Sebastien .
JOURNAL OF CHEMINFORMATICS, 2021, 13 (01)
[6]   GuacaMol: Benchmarking Models for de Novo Molecular Design [J].
Brown, Nathan ;
Fiscato, Marco ;
Segler, Marwin H. S. ;
Vaucher, Alain C. .
JOURNAL OF CHEMICAL INFORMATION AND MODELING, 2019, 59 (03) :1096-1108
[7]  
Chan L, 2022, ARXIV, DOI DOI 10.48550/ARXIV.2204.10663
[8]   Knowledge-Based Conformer Generation Using the Cambridge Structural Database [J].
Cole, Jason C. ;
Korb, Oliver ;
McCabe, Patrick ;
Read, Murray G. ;
Taylor, Robin .
JOURNAL OF CHEMICAL INFORMATION AND MODELING, 2018, 58 (03) :615-629
[9]  
Cook S, 2013, CUDA PROGRAMMING: A DEVELOPER'S GUIDE TO PARALLEL COMPUTING WITH GPUS, P1
[10]   3-D Inorganic Crystal Structure Generation and Property Prediction via Representation Learning [J].
Court, Callum J. ;
Yildirim, Batuhan ;
Jain, Apoorv ;
Cole, Jacqueline M. .
JOURNAL OF CHEMICAL INFORMATION AND MODELING, 2020, 60 (10) :4518-4535