Accurate prediction of protein structures and interactions using a three-track neural network

被引:3205
作者
Baek, Minkyung [1 ,2 ]
DiMaio, Frank [1 ,2 ]
Anishchenko, Ivan [1 ,2 ]
Dauparas, Justas [1 ,2 ]
Ovchinnikov, Sergey [3 ,4 ]
Lee, Gyu Rie [1 ,2 ]
Wang, Jue [1 ,2 ]
Cong, Qian [5 ,6 ]
Kinch, Lisa N. [7 ]
Schaeffer, R. Dustin [6 ]
Millan, Claudia [8 ]
Park, Hahnbeom [1 ,2 ]
Adams, Carson [1 ,2 ]
Glassman, Caleb R. [9 ,10 ,11 ]
DeGiovanni, Andy [12 ]
Pereira, Jose H. [12 ]
Rodrigues, Andria V. [12 ]
van Dijk, Alberdina A. [13 ]
Ebrecht, Ana C. [13 ]
Opperman, Diederik J. [14 ]
Sagmeister, Theo [15 ]
Buhlheller, Christoph [15 ,16 ]
Pavkov-Keller, Tea [15 ,17 ]
Rathinaswamy, Manoj K. [18 ]
Dalwadi, Udit [19 ]
Yip, Calvin K. [19 ]
Burke, John E. [18 ]
Garcia, K. Christopher [9 ,10 ,11 ,20 ]
Grishin, Nick V. [6 ,7 ,21 ]
Adams, Paul D. [12 ,22 ]
Read, Randy J. [8 ]
Baker, David [1 ,2 ,23 ]
机构
[1] Univ Washington, Dept Biochem, Seattle, WA 98195 USA
[2] Univ Washington, Inst Prot Design, Seattle, WA 98195 USA
[3] Harvard Univ, Div Sci, Fac Arts & Sci, Cambridge, MA 02138 USA
[4] Harvard Univ, John Harvard Distinguished Sci Fellowship Program, Cambridge, MA 02138 USA
[5] Univ Texas Southwestern Med Ctr Dallas, Eugene McDermott Ctr Human Growth & Dev, Dallas, TX 75390 USA
[6] Univ Texas Southwestern Med Ctr Dallas, Dept Phys, Dallas, TX 75390 USA
[7] Univ Texas Southwestern Med Ctr Dallas, Howard Hughes Med Inst, Dallas, TX USA
[8] Univ Cambridge, Cambridge Inst Med Res, Dept Haematol, Cambridge, England
[9] Stanford Univ, Sch Med, Program Immunol, Stanford, CA 94305 USA
[10] Stanford Univ, Sch Med, Dept Mol & Cellular Physiol, Stanford, CA 94305 USA
[11] Stanford Univ, Sch Med, Dept Biol Struct, Stanford, CA 94305 USA
[12] Lawrence Berkeley Natl Lab, Mol Biophys & Integrated Bioimaging Div, Berkeley, CA USA
[13] North West Univ, Dept Biochem, Focus Area Human Metabol, ZA-2531 Potchefstroom, South Africa
[14] Univ Free State, Dept Biotechnol, 205 Nelson Mandela Dr, ZA-9300 Bloemfontein, South Africa
[15] Karl Franzens Univ Graz, Inst Mol Biosci, Humboldtstr 50, A-8010 Graz, Austria
[16] Med Univ Graz, Graz, Austria
[17] BioTechMed Graz, Graz, Austria
[18] Univ Victoria, Dept Biochem & Microbiol, Victoria, BC, Canada
[19] Univ British Columbia, Life Sci Inst, Dept Biochem & Mol Biol, Vancouver, BC, Canada
[20] Stanford Univ, Howard Hughes Med Inst, Sch Med, Stanford, CA 94305 USA
[21] Univ Texas Southwestern Med Ctr Dallas, Dept Biochem, Dallas, TX 75390 USA
[22] Univ Calif Berkeley, Dept Bioengn, Berkeley, CA 94720 USA
[23] Univ Washington, Howard Hughes Med Inst, Seattle, WA 98195 USA
基金
英国惠康基金; 奥地利科学基金会; 美国国家卫生研究院; 美国国家科学基金会;
关键词
MULTIPLE SEQUENCE ALIGNMENT; MOLECULAR-REPLACEMENT; STRUCTURE REFINEMENT; CRYSTAL-STRUCTURES; METABOLIC CRISES; FAMILY; MODELS; DOMAIN; DISINTEGRIN; FEATURES;
D O I
10.1126/science.abj8754
中图分类号
O [数理科学和化学]; P [天文学、地球科学]; Q [生物科学]; N [自然科学总论];
学科分类号
07 ; 0710 ; 09 ;
摘要
DeepMind presented notably accurate predictions at the recent 14th Critical Assessment of Structure Prediction (CASP14) conference. We explored network architectures that incorporate related ideas and obtained the best performance with a three-track network in which information at the one-dimensional (1D) sequence level, the 2D distance map level, and the 3D coordinate level is successively transformed and integrated. The three-track network produces structure predictions with accuracies approaching those of DeepMind in CASP14, enables the rapid solution of challenging x-ray crystallography and cryo-electron microscopy structure modeling problems, and provides insights into the functions of proteins of currently unknown structure. The network also enables rapid generation of accurate protein-protein complex models from sequence information alone, short-circuiting traditional approaches that require modeling of individual subunits followed by docking. We make the method available to the scientific community to speed biological research.
引用
收藏
页码:871 / +
页数:52
相关论文
共 83 条
[1]   Towards automated crystallographic structure refinement with phenix.refine [J].
Afonine, Pavel V. ;
Grosse-Kunstleve, Ralf W. ;
Echols, Nathaniel ;
Headd, Jeffrey J. ;
Moriarty, Nigel W. ;
Mustyakimov, Marat ;
Terwilliger, Thomas C. ;
Urzhumtsev, Alexandre ;
Zwart, Peter H. ;
Adams, Paul D. .
ACTA CRYSTALLOGRAPHICA SECTION D-STRUCTURAL BIOLOGY, 2012, 68 :352-367
[2]   OMA orthology in 2021: website overhaul, conserved isoforms, ancestral gene order and more [J].
Altenhoff, Adrian M. ;
Train, Clement-Marie ;
Gilbert, Kimberly J. ;
Mediratta, Ishita ;
de Farias, Tarcisio Mendes ;
Moi, David ;
Nevers, Yannis ;
Radoykova, Hale-Seda ;
Rossier, Victor ;
Vesztrocy, Alex Warwick ;
Glover, Natasha M. ;
Dessimoz, Christophe .
NUCLEIC ACIDS RESEARCH, 2021, 49 (D1) :D373-D379
[3]  
Anishchenko I., 2020, BIORXIV, DOI [10.1101/2020.07.22.211482, DOI 10.1101/2020.07.22.211482]
[4]  
Anishchenko I., 2021, ZENODO
[5]  
Apweiler R, 2004, NUCLEIC ACIDS RES, V32, pD115, DOI [10.1093/nar/gkh131, 10.1093/nar/gkw1099]
[6]   RUPEE: A fast and accurate purely geometric protein structure search [J].
Ayoub, Ronald ;
Lee, Yugyung .
PLOS ONE, 2019, 14 (03)
[7]   Improving homology modeling from low-sequence identity templates in Rosetta: A case study in GPCRs [J].
Bender, Brian Joseph ;
Marlow, Brennica ;
Meiler, Jens .
PLOS COMPUTATIONAL BIOLOGY, 2020, 16 (10)
[8]   Toward the estimation of the absolute quality of individual protein structure models [J].
Benkert, Pascal ;
Biasini, Marco ;
Schwede, Torsten .
BIOINFORMATICS, 2011, 27 (03) :343-350
[9]   The SWISS-MODEL Repository-new features and functionality [J].
Bienert, Stefan ;
Waterhouse, Andrew ;
de Beer, Tjaart A. P. ;
Tauriello, Gerardo ;
Studer, Gabriel ;
Bordoli, Lorenza ;
Schwede, Torsten .
NUCLEIC ACIDS RESEARCH, 2017, 45 (D1) :D313-D319
[10]  
Bunkoczi G., 2011, Comput Crystallogr Newsl, V2, P8