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

DOI: 10.1126/science.abj8754

Authors: Minkyung Baek (University of Washington) , Frank Dimaio (University of Washington) , Ivan Anishchenko (University of Washington) , Justas Dauparas (University of Washington) , Sergey Ovchinnikov (Harvard University) , Gyu Rie Lee (University of Washington) , Jue Wang (University of Washington) , Qian Cong (University of Texas Southwestern Medical Center) , Lisa N. Kinch (University of Texas Southwestern Medical Center,) , R. Dustin Schaeffer (University of Texas Southwestern Medical Center) , Claudia Millán (Cambridge Institute for Medical Research) , Hahnbeom Park (University of Washington) , Carson Adams (University of Washington) , Caleb R. Glassman (Stanford University School of Medicine) , Andy Degiovanni (Lawrence Berkeley National Laboratory) , Jose H. Pereira (Lawrence Berkeley National Laboratory) , Andria V. Rodrigues (Lawrence Berkeley National Laboratory) , Alberdina A. Van Dijk (North-West University) , Ana C. Ebrecht (North-West University) , Diederik J. Opperman (University of the Free State) , Theo Sagmeister (University of Graz) , Christoph Buhlheller (University of Graz; Medical University of Graz) , Tea Pavkov-Keller (University of Graz; BioTechMed-Graz) , Manoj K. Rathinaswamy (University of Victoria) , Udit Dalwadi (The University of British Columbia) , Calvin K. Yip (The University of British Columbia) , John E. Burke (University of Victoria) , K. Christopher Garcia (Stanford University School of Medicine) , Nick V. Grishin (University of Texas Southwestern Medical Center) , Paul D. Adams (Lawrence Berkeley National Laboratory; University of California, Berkeley) , Randy J. Read (University of Cambridge) , David Baker (University of Washington)
Co-authored by industrial partner: No

Type: Journal Paper
Journal: Science , VOL 2

State: Published (Approved)
Published: July 2021
Diamond Proposal Number(s): 20303

Abstract: DeepMind presented remarkably accurate predictions at the recent CASP14 protein structure prediction assessment conference. We explored network architectures incorporating related ideas and obtained the best performance with a three-track network in which information at the 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-EM 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 which require modeling of individual subunits followed by docking. We make the method available to the scientific community to speed biological research.

Subject Areas: Biology and Bio-materials, Information and Communication Technology


Instruments: I04-Macromolecular Crystallography

Other Facilities: ID30-3 at ESRF; P11 at PETRA III

Added On: 20/07/2021 08:19

Discipline Tags:

Artificial Intelligence Information & Communication Technologies Structural biology Life Sciences & Biotech

Technical Tags:

Diffraction Macromolecular Crystallography (MX)