Improved protein structure prediction using potentials from deep learning

Protein structure prediction can be used to determine the three-dimensional shape of a protein from its amino acid sequence 1 . This problem is of fundamental importance as the structure of a protein largely determines its function 2 ; however, protein structures can be difficult to determine experimentally. Considerable progress has recently been made by leveraging genetic information. It is possible to infer which amino acid residues are in contact by analysing covariation in homologous sequences, which aids in the prediction of protein structures 3 . Here we show that we can train a neural network to make accurate predictions of the distances between pairs of residues, which convey more information about the structure than contact predictions. Using this information, we construct a potential of mean force 4 that can accurately describe the shape of a protein. We find that the resulting potential can be optimized by a simple gradient descent algorithm to generate structures without complex sampling procedures. The resulting system, named AlphaFold, achieves high accuracy, even for sequences with fewer homologous sequences. In the recent Critical Assessment of Protein Structure Prediction 5 (CASP13)—a blind assessment of the state of the field—AlphaFold created high-accuracy structures (with template modelling (TM) scores 6 of 0.7 or higher) for 24 out of 43 free modelling domains, whereas the next best method, which used sampling and contact information, achieved such accuracy for only 14 out of 43 domains. AlphaFold represents a considerable advance in protein-structure prediction. We expect this increased accuracy to enable insights into the function and malfunction of proteins, especially in cases for which no structures for homologous proteins have been experimentally determined 7 . AlphaFold predicts the distances between pairs of residues, is used to construct potentials of mean force that accurately describe the shape of a protein and can be optimized with gradient descent to predict protein structures.

[1]  Marco Biasini,et al.  lDDT: a local superposition-free score for comparing protein structures and models using distance difference tests , 2013, Bioinform..

[2]  David E. Kim,et al.  Improved de novo structure prediction in CASP11 by incorporating coevolution information into Rosetta , 2016, Proteins.

[3]  David C. Jones Predicting novel protein folds by using FRAGFOLD , 2001, Proteins.

[4]  Petr Popov,et al.  Crystal structure of misoprostol bound to the labor inducer prostaglandin E2 receptor , 2018, Nature Chemical Biology.

[5]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[6]  David Baker,et al.  Macromolecular modeling with rosetta. , 2008, Annual review of biochemistry.

[7]  T. N. Bhat,et al.  The Protein Data Bank , 2000, Nucleic Acids Res..

[8]  Matteo Dal Peraro,et al.  A further leap of improvement in tertiary structure prediction in CASP13 prompts new routes for future assessments , 2019, Proteins.

[9]  Lloyd Allison,et al.  Minimum message length inference of secondary structure from protein coordinate data , 2012, Bioinform..

[10]  Maria Jesus Martin,et al.  Uniclust databases of clustered and deeply annotated protein sequences and alignments , 2017, Nucleic Acids Res..

[11]  T. Madden,et al.  Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. , 1997, Nucleic acids research.

[12]  C Kooperberg,et al.  Assembly of protein tertiary structures from fragments with similar local sequences using simulated annealing and Bayesian scoring functions. , 1997, Journal of molecular biology.

[13]  A. Lesk,et al.  Correlation of co-ordinated amino acid substitutions with function in viruses related to tobacco mosaic virus. , 1987, Journal of molecular biology.

[14]  Yang Zhang Protein structure prediction: when is it useful? , 2009, Current opinion in structural biology.

[15]  C. Sander,et al.  Direct-coupling analysis of residue coevolution captures native contacts across many protein families , 2011, Proceedings of the National Academy of Sciences.

[16]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[17]  C Venclovas,et al.  Processing and analysis of CASP3 protein structure predictions , 1999, Proteins.

[18]  Jinbo Xu,et al.  A position-specific distance-dependent statistical potential for protein structure and functional study. , 2012, Structure.

[19]  Yuan Yu,et al.  TensorFlow: A system for large-scale machine learning , 2016, OSDI.

[20]  Marcin J. Skwark,et al.  Improved Contact Predictions Using the Recognition of Protein Like Contact Patterns , 2014, PLoS Comput. Biol..

[21]  David A. Lee,et al.  CATH: an expanded resource to predict protein function through structure and sequence , 2017, Nucleic Acids Res..

[22]  K. Dill,et al.  The Protein-Folding Problem, 50 Years On , 2012, Science.

[23]  Jimin Pei,et al.  An automatic method for CASP9 free modeling structure prediction assessment , 2011, Bioinform..

[24]  C. D. Gelatt,et al.  Optimization by Simulated Annealing , 1983, Science.

[25]  Yang Zhang,et al.  Ensembling multiple raw coevolutionary features with deep residual neural networks for contact‐map prediction in CASP13 , 2019, Proteins.

[26]  Pushmeet Kohli,et al.  Protein structure prediction using multiple deep neural networks in the 13th Critical Assessment of Protein Structure Prediction (CASP13) , 2019, Proteins.

[27]  Vladlen Koltun,et al.  Multi-Scale Context Aggregation by Dilated Convolutions , 2015, ICLR.

[28]  David T Jones,et al.  Prediction of interresidue contacts with DeepMetaPSICOV in CASP13 , 2019, Proteins.

[29]  K. Dill,et al.  The protein folding problem. , 2008, Annual review of biophysics.

[30]  Heiga Zen,et al.  WaveNet: A Generative Model for Raw Audio , 2016, SSW.

[31]  Sepp Hochreiter,et al.  Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs) , 2015, ICLR.

[32]  Johannes Söding,et al.  The HHpred interactive server for protein homology detection and structure prediction , 2005, Nucleic Acids Res..

[33]  James Scott-Brown,et al.  Visualization and analysis of non-covalent contacts using the Protein Contacts Atlas , 2018, Nature structural & molecular biology.

[34]  David T. Jones,et al.  MetaPSICOV: combining coevolution methods for accurate prediction of contacts and long range hydrogen bonding in proteins , 2015, Bioinform..

[35]  Zhen Li,et al.  Accurate De Novo Prediction of Protein Contact Map by Ultra-Deep Learning Model , 2017, PLoS Comput. Biol..

[36]  David T. Jones,et al.  High precision in protein contact prediction using fully convolutional neural networks and minimal sequence features , 2018, Bioinform..

[37]  W. Taylor,et al.  Global fold determination from a small number of distance restraints. , 1995, Journal of molecular biology.

[38]  W. Taylor,et al.  Estimating polypeptideα-carbon distances from multiple sequence alignments , 1995 .

[39]  D. Baker,et al.  Robust and accurate prediction of residue–residue interactions across protein interfaces using evolutionary information , 2014, eLife.

[40]  Jinbo Xu,et al.  Analysis of distance-based protein structure prediction by deep learning in CASP13 , 2019 .

[41]  Ankur Taly,et al.  Axiomatic Attribution for Deep Networks , 2017, ICML.

[42]  Yang Zhang,et al.  Scoring function for automated assessment of protein structure template quality , 2004, Proteins.

[43]  A. Tramontano,et al.  Critical assessment of methods of protein structure prediction (CASP)—round IX , 2011, Proteins.

[44]  A. Biegert,et al.  HHblits: lightning-fast iterative protein sequence searching by HMM-HMM alignment , 2012, Nature Methods.

[45]  Torsten Schwede,et al.  Critical assessment of methods of protein structure prediction (CASP)—Round XIII , 2019, Proteins.

[46]  Kuldip K. Paliwal,et al.  Sixty-five years of the long march in protein secondary structure prediction: the final stretch? , 2018, Briefings Bioinform..

[47]  Yang Zhang,et al.  Template‐based and free modeling of I‐TASSER and QUARK pipelines using predicted contact maps in CASP12 , 2018, Proteins.

[48]  Randy J Read,et al.  Evaluation of template‐based modeling in CASP13 , 2019, Proteins.

[49]  Andriy Kryshtafovych,et al.  Assessment of contact predictions in CASP12: Co‐evolution and deep learning coming of age , 2018, Proteins.

[50]  J. Skolnick,et al.  TM-align: a protein structure alignment algorithm based on the TM-score , 2005, Nucleic acids research.

[51]  J. Kirkwood Statistical Mechanics of Fluid Mixtures , 1935 .

[52]  Jorge Nocedal,et al.  On the limited memory BFGS method for large scale optimization , 1989, Math. Program..

[53]  Ilya A Vakser,et al.  Docking of protein models , 2002, Protein science : a publication of the Protein Society.

[54]  Massimiliano Pontil,et al.  PSICOV: precise structural contact prediction using sparse inverse covariance estimation on large multiple sequence alignments , 2012, Bioinform..

[55]  Markus Gruber,et al.  CCMpred—fast and precise prediction of protein residue–residue contacts from correlated mutations , 2014, Bioinform..

[56]  W. Kabsch,et al.  Dictionary of protein secondary structure: Pattern recognition of hydrogen‐bonded and geometrical features , 1983, Biopolymers.