Predicting locations of cryptic pockets from single protein structures using the PocketMiner graph neural network.
Predicting locations of cryptic pockets from single protein structures using the PocketMiner graph neural network.
复制标题
DOI：
10.1038/s41467-023-36699-3
复制
发表时间：
2023-03-01
影响因子：
16.6
通讯作者：
Bowman, Gregory R.
中科院分区：
文献类型：
--
作者：
Meller, Artur;Ward, Michael;Borowsky, Jonathan;Kshirsagar, Meghana;Lotthammer, Jeffrey M.;Oviedo, Felipe;Ferres, Juan Lavista;Bowman, Gregory R.
隐蔽的口袋扩大了药物发现的范围，因为目前被认为无法下药的蛋白质在其基态结构中没有口袋，因此能够进行靶向治疗。然而，识别隐秘的口袋是劳动密集型的，而且速度很慢。准确而快速地预测结构是否以及在哪里可能形成隐蔽的口袋的能力将极大地加快寻找可吸毒的口袋的速度。在这里，我们介绍PocketMiner，这是一种图形神经网络，经过训练后可以在分子动力学模拟中预测口袋可能打开的位置。将PocketMiner应用于一个由39个经过实验确认的密袋组成的新数据集中的单个结构，表明它可以准确地识别密袋(ROC-AUC：0.87)>；，速度是现有方法的1000倍。我们将PocketMiner应用于人类蛋白质组，并在模拟中显示预测到的口袋打开，这表明根据现有结构被认为没有口袋的蛋白质中，超过一半可能包含隐藏的口袋，从而极大地扩大了潜在的可下药蛋白质组。隐蔽的口袋可以靶向目前被认为无法下药的蛋白质，因为它们的基态结构中没有口袋。在这里，作者开发了一种图形神经网络，它只使用分子模拟数据进行训练，就可以准确地预测静态结构中的隐蔽口袋。
Cryptic pockets expand the scope of drug discovery by enabling targeting of proteins currently considered undruggable because they lack pockets in their ground state structures. However, identifying cryptic pockets is labor-intensive and slow. The ability to accurately and rapidly predict if and where cryptic pockets are likely to form from a structure would greatly accelerate the search for druggable pockets. Here, we present PocketMiner, a graph neural network trained to predict where pockets are likely to open in molecular dynamics simulations. Applying PocketMiner to single structures from a newly curated dataset of 39 experimentally confirmed cryptic pockets demonstrates that it accurately identifies cryptic pockets (ROC-AUC: 0.87) >1,000-fold faster than existing methods. We apply PocketMiner across the human proteome and show that predicted pockets open in simulations, suggesting that over half of proteins thought to lack pockets based on available structures likely contain cryptic pockets, vastly expanding the potentially druggable proteome. Cryptic pockets enable targeting of proteins currently considered undruggable because they lack pockets in their ground state structures. Here, the authors develop a graph neural network that accurately predicts cryptic pockets in static structures by training using molecular simulation data alone.
登录
查看更多内容
Biopython：用于计算分子生物学和生物信息学的免费 Python 工具。
DOI：
10.1093/bioinformatics/btp163
发表时间：
2009-06-01
期刊：
Bioinformatics (Oxford, England)
影响因子：
--
作者：
Cock PJ;Antao T;Chang JT;Chapman BA;Cox CJ;Dalke A;Friedberg I;Hamelryck T;Kauff F;Wilczynski B;de Hoon MJ
通讯作者：
de Hoon MJ
影响因子：
3
作者：
Duan, Y;Wu, C;Kollman, P
通讯作者：
Kollman, P
DOI：
10.1073/pnas.1417811112
发表时间：
2015-03-03
影响因子：
11.1
作者：
Bowman, Gregory R.;Bolin, Eric R.;Marqusee, Susan
通讯作者：
Marqusee, Susan
DOI：
10.1126/science.abf7945
发表时间：
2021-05-07
期刊：
Science (New York, N.Y.)
影响因子：
--
作者：
Günther S;Reinke PYA;Fernández-García Y;Lieske J;Lane TJ;Ginn HM;Koua FHM;Ehrt C;Ewert W;Oberthuer D;Yefanov O;Meier S;Lorenzen K;Krichel B;Kopicki JD;Gelisio L;Brehm W;Dunkel I;Seychell B;Gieseler H;Norton-Baker B;Escudero-Pérez B;Domaracky M;Saouane S;Tolstikova A;White TA;Hänle A;Groessler M;Fleckenstein H;Trost F;Galchenkova M;Gevorkov Y;Li C;Awel S;Peck A;Barthelmess M;Schlünzen F;Lourdu Xavier P;Werner N;Andaleeb H;Ullah N;Falke S;Srinivasan V;França BA;Schwinzer M;Brognaro H;Rogers C;Melo D;Zaitseva-Kinneberg JI;Knoska J;Peña-Murillo GE;Mashhour AR;Hennicke V;Fischer P;Hakanpää J;Meyer J;Gribbon P;Ellinger B;Kuzikov M;Wolf M;Beccari AR;Bourenkov G;von Stetten D;Pompidor G;Bento I;Panneerselvam S;Karpics I;Schneider TR;Garcia-Alai MM;Niebling S;Günther C;Schmidt C;Schubert R;Han H;Boger J;Monteiro DCF;Zhang L;Sun X;Pletzer-Zelgert J;Wollenhaupt J;Feiler CG;Weiss MS;Schulz EC;Mehrabi P;Karničar K;Usenik A;Loboda J;Tidow H;Chari A;Hilgenfeld R;Uetrecht C;Cox R;Zaliani A;Beck T;Rarey M;Günther S;Turk D;Hinrichs W;Chapman HN;Pearson AR;Betzel C;Meents A
通讯作者：
Meents A
影响因子：
3
作者：
Camacho C;Coulouris G;Avagyan V;Ma N;Papadopoulos J;Bealer K;Madden TL
通讯作者：
Madden TL