No retraction, correction or concern found when last checked on Oct 16, 2026.

Accurate De Novo Prediction of Protein Contact Map by Ultra-Deep Learning Model

Sheng Wang, Siqi Sun, Zhen Li and 2 more

PLOS Computational Biology · 2017 · CC BY

Abstract

Motivation. Protein contacts contain key information for the understanding of protein structure and function and thus, contact prediction from sequence is an important problem. Recently exciting progress has been made on this problem, but the predicted contacts for proteins without many sequence homologs is still of low quality and not very useful for de novo structure prediction.

Method. This paper presents a new deep learning method that predicts contacts by integrating both evolutionary coupling (EC) and sequence conservation information through an ultra-deep neural network formed by two deep residual neural networks. The first residual network conducts a series of 1-dimensional convolutional transformation of sequential features; the second residual network conducts a series of 2-dimensional convolutional transformation of pairwise information including output of the first residual network, EC information and pairwise potential. By using very deep residual networks, we can accurately model contact occurrence patterns and complex sequence-structure relationship and thus, obtain higher-quality contact prediction regardless of how many sequence homologs are available for proteins in question.

Results. Our method greatly outperforms existing methods and leads to much more accurate contact-assisted folding. Tested on 105 CASP11 targets, 76 past CAMEO hard targets, and 398 membrane proteins, the average top L long-range prediction accuracy obtained by our method, one representative EC method CCMpred and the CASP11 winner MetaPSICOV is 0.47, 0.21 and 0.30, respectively; the average top L/10 long-range accuracy of our method, CCMpred and MetaPSICOV is 0.77, 0.47 and 0.59, respectively. Ab initio folding using our predicted contacts as restraints but without any force fields can yield correct folds (i.e., TMscore>0.6) for 203 of the 579 test proteins, while that using MetaPSICOV- and CCMpred-predicted contacts can do so for only 79 and 62 of them, respectively. Our contact-assisted models also have much better quality than template-based models especially for membrane proteins. The 3D models built from our contact prediction have TMscore>0.5 for 208 of the 398 membrane proteins, while those from homology modeling have TMscore>0.5 for only 10 of them. Further, even if trained mostly by soluble proteins, our deep learning method works very well on membrane proteins. In the recent blind CAMEO benchmark, our fully-automated web server implementing this method successfully folded 6 targets with a new fold and only 0.3L-2.3L effective sequence homologs, including one β protein of 182 residues, one α+β protein of 125 residues, one α protein of 140 residues, one α protein of 217 residues, one α/β of 260 residues and one α protein of 462 residues. Our method also achieved the highest F1 score on free-modeling targets in the latest CASP (Critical Assessment of Structure Prediction), although it was not fully implemented back then.

Availability. http://raptorx.uchicago.edu/ContactMap/

Full text

Its licence does not allow us to show the full text. Read it at PubMed Central:

References 49

1.
Kim DE, DiMaio F, Yu‐Ruei Wang R, Song Y, Baker D. One contact for every twelve residues allows robust and accurate topology‐level protein structure modeling. Proteins: Structure, Function, and Bioinformatics. 2014;82(S2):208–18.
Not in our library10.1002/prot.24374
2.
de Juan D, Pazos F, Valencia A. Emerging methods in protein co-evolution. Nature Reviews Genetics. 2013;14(4):249–61.
Not in our library10.1038/nrg3414
3.
Weigt M, White RA, Szurmant H, Hoch JA, Hwa T. Identification of direct residue contacts in protein-protein interaction by message passing. P Natl Acad Sci USA. 2009;106(1):67–72.
Not in our library10.1073/pnas.0805923106
4.
Seemayer S, Gruber M, Söding J. CCMpred—fast and precise prediction of protein residue–residue contacts from correlated mutations. Bioinformatics. 2014;30(21):3128–30.
Not in our library10.1093/bioinformatics/btu500
5.
Jones DT, Buchan DW, Cozzetto D, Pontil M. PSICOV: precise structural contact prediction using sparse inverse covariance estimation on large multiple sequence alignments. Bioinformatics. 2012;28(2):184–90.
Not in our library10.1093/bioinformatics/btr638
6.
Protein 3D Structure Computed from Evolutionary Sequence Variation · 2011
No notice
7.
Ekeberg M, Hartonen T, Aurell E. Fast pseudolikelihood maximization for direct-coupling analysis of protein structure from many homologous amino-acid sequences. J Comput Phys. 2014;276:341–56.
Not in our library
8.
Kamisetty H, Ovchinnikov S, Baker D. Assessing the utility of coevolution-based residue–residue contact predictions in a sequence-and structure-rich era. Proceedings of the National Academy of Sciences. 2013;110(39):15674–9.
Not in our library10.1073/pnas.1314045110
9.
MetaPSICOV: combining coevolution methods for accurate prediction of contacts and long range hydrogen bonding in proteins · 2014
No notice
10.
Ma J, Wang S, Wang Z, Xu J. Protein contact prediction by integrating joint evolutionary coupling analysis and supervised learning. Bioinformatics. 2015:btv472.
Not in our library10.1093/bioinformatics/btv472
11.
Adhikari B, Bhattacharya D, Cao R, Cheng J. CONFOLD: residue‐residue contact‐guided ab initio protein folding. Proteins: Structure, Function, and Bioinformatics. 2015;83(8):1436–49.
Not in our library10.1002/prot.24829
12.
Wang S, Li W, Zhang R, Liu S, Xu J. CoinFold: a web server for protein contact prediction and contact-assisted protein folding. Nucleic acids research. 2016:gkw307.
Not in our library10.1093/nar/gkw307
13.
Di Lena P, Nagata K, Baldi P. Deep architectures for protein contact map prediction. Bioinformatics. 2012;28(19):2449–57.
Not in our library10.1093/bioinformatics/bts475
14.
Ekeberg M, Lövkvist C, Lan Y, Weigt M, Aurell E. Improved contact prediction in proteins: using pseudolikelihoods to infer Potts models. Phys Rev E. 2013;87(1):012707.
Not in our library10.1103/physreve.87.012707
15.
Göbel U, Sander C, Schneider R, Valencia A. Correlated mutations and residue contacts in proteins. Proteins: Structure, Function, and Bioinformatics. 1994;18(4):309–17.
Not in our library10.1002/prot.340180402
16.
Morcos F, Pagnani A, Lunt B, Bertolino A, Marks DS, Sander C, et al Direct-coupling analysis of residue coevolution captures native contacts across many protein families. Proceedings of the National Academy of Sciences. 2011;108(49):E1293–E301.
Not in our library10.1073/pnas.1111471108
17.
Improved Contact Predictions Using the Recognition of Protein Like Contact Patterns · 2014
No notice
18.
Wu S, Zhang Y. A comprehensive assessment of sequence-based and template-based methods for protein contact prediction. Bioinformatics. 2008;24(7):924–31.
Not in our library10.1093/bioinformatics/btn069
19.
Wang Z, Xu J. Predicting protein contact map using evolutionary and physical constraints by integer programming. Bioinformatics. 2013;29(13):i266–i73.
Not in our library10.1093/bioinformatics/btt211
20.
He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. arXiv preprint arXiv:151203385. 2015.
Not in our library
21.
Krizhevsky A, Hinton G. Convolutional deep belief networks on cifar-10. Unpublished manuscript. 2010;40.
Not in our library
22.
Srivastava RK, Greff K, Schmidhuber J, editors. Training very deep networks. Advances in neural information processing systems; 2015.
Not in our library
23.
Hinton G, Deng L, Yu D, Dahl GE, Mohamed A-r, Jaitly N, et al Deep neural networks for acoustic modeling in speech recognition: The shared views of four research groups. IEEE Signal Processing Magazine. 2012;29(6):82–97.
Not in our library
24.
Deep learning · 2015
No notice
25.
Szegedy C, Liu W, Jia Y, Sermanet P, Reed S, Anguelov D, et al., editors. Going deeper with convolutions. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition; 2015.
Not in our library
26.
Moult J, Fidelis K, Kryshtafovych A, Schwede T, Tramontano A. Critical assessment of methods of protein structure prediction (CASP)—round x. Proteins: Structure, Function, and Bioinformatics. 2014;82(S2):1–6.
Not in our library10.1002/prot.24452
27.
Moult J, Fidelis K, Kryshtafovych A, Schwede T, Tramontano A. Critical assessment of methods of protein structure prediction: Progress and new directions in round XI. Proteins: Structure, Function, and Bioinformatics. 2016.
Not in our library10.1002/prot.25064
28.
Haas J, Roth S, Arnold K, Kiefer F, Schmidt T, Bordoli L, et al The Protein Model Portal—a comprehensive resource for protein structure and model information. Database. 2013;2013:bat031.
Not in our library10.1093/database/bat031
29.
Pinheiro PH, Collobert R, editors. Recurrent Convolutional Neural Networks for Scene Labeling. ICML; 2014.
Not in our library
30.
Nair V, Hinton GE, editors. Rectified linear units improve restricted boltzmann machines. Proceedings of the 27th International Conference on Machine Learning (ICML-10); 2010.
Not in our library
31.
Monastyrskyy B, D'Andrea D, Fidelis K, Tramontano A, Kryshtafovych A. New encouraging developments in contact prediction: Assessment of the CASP11 results. Proteins: Structure, Function, and Bioinformatics. 2015.
Not in our library10.1002/prot.24943
32.
Briinger AT, Adams PD, Clore GM, DeLano WL, Gros P, Grosse-Kunstleve RW, et al Crystallography & NMR system: A new software suite for macromolecular structure determination. Acta Crystallogr D Biol Crystallogr. 1998;54(5):905–21.
Not in our library10.1107/s0907444998003254
33.
Zhang Y, Skolnick J. Scoring function for automated assessment of protein structure template quality. Proteins: Structure, Function, and Bioinformatics. 2004;57(4):702–10.
Not in our library10.1002/prot.20264
34.
Xu JR, Zhang Y. How significant is a protein structure similarity with TM-score = 0.5? Bioinformatics. 2010;26(7):889–95.
Not in our library10.1093/bioinformatics/btq066
35.
Kim DE, Chivian D, Baker D. Protein structure prediction and analysis using the Robetta server. Nucleic Acids Research. 2004;32:W526–W31.
Not in our library10.1093/nar/gkh468
36.
Kelley LA, Mezulis S, Yates CM, Wass MN, Sternberg MJE. The Phyre2 web portal for protein modeling, prediction and analysis. Nature protocols. 2015;10(6):845–58.
Not in our library10.1038/nprot.2015.053
37.
Källberg M, Wang H, Wang S, Peng J, Wang Z, Lu H, et al Template-based protein structure modeling using the RaptorX web server. Nature protocols. 2012;7(8):1511–22.
Not in our library10.1038/nprot.2012.085
38.
SWISS-MODEL: modelling protein tertiary and quaternary structure using evolutionary information · 2014
No notice
39.
Remmert M, Biegert A, Hauser A, Söding J. HHblits: lightning-fast iterative protein sequence searching by HMM-HMM alignment. Nature methods. 2012;9(2):173–5.
Not in our library10.1038/nmeth.1818
40.
Wang S, Ma J, Peng J, Xu J. Protein structure alignment beyond spatial proximity. Sci Rep. 2013;3:1448 Epub 2013/03/15. PubMed Central PMCID: PMC3596798.
Not in our library10.1038/srep01448
41.
Zalewski JK, Mo JH, Heber S, Heroux A, Gardner RG, Hildebrand JD, et al Structure of the Shroom-Rho kinase complex reveals a binding interface with monomeric Shroom that regulates cell morphology and stimulates kinase activity. J Biol Chem. 2016. Epub 2016/10/21.
Not in our library10.1074/jbc.m116.738559
42.
Mohan S, Rizaldy R, Das D, Bauer RJ, Heroux A, Trakselis MA, et al Structure of Shroom domain 2 reveals a three-segmented coiled-coil required for dimerization, Rock binding, and apical constriction. Mol Biol Cell. 2012;23(11):2131–42. Epub 2012/04/12. PubMed Central PMCID: PMC3364177.
Not in our library10.1091/mbc.e11-11-0937
43.
Ioffe S, Szegedy C, editors. Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift. Proceedings of The 32nd International Conference on Machine Learning; 2015.
Not in our library
44.
Protein Secondary Structure Prediction Using Deep Convolutional Neural Fields · 2016
No notice
45.
Miyazawa S, Jernigan RL. Estimation of effective interresidue contact energies from protein crystal structures: quasi-chemical approximation. Macromolecules. 1985;18(3):534–52.
Not in our library
46.
Betancourt MR, Thirumalai D. Pair potentials for protein folding: choice of reference states and sensitivity of predicted native states to variations in the interaction schemes. Protein Science. 1999;8(02):361–9.
Not in our library10.1110/ps.8.2.361
47.
Bergstra J, Breuleux O, Bastien F, Lamblin P, Pascanu R, Desjardins G, et al., editors. Theano: A CPU and GPU math compiler in Python. Proc 9th Python in Science Conf; 2010.
Not in our library
48.
JPred4: a protein secondary structure prediction server · 2015
No notice
49.
Wang S, Li W, Liu S, Xu J. RaptorX-Property: a web server for protein structure property prediction. Nucleic acids research. 2016:gkw306.
Not in our library10.1093/nar/gkw306

Cited by papers we hold 26

Modeling protein conformational ensembles by guiding AlphaFold2 with Double Electron Electron Resonance (DEER) distance distributions · 2025
Not checked
When will RNA get its AlphaFold moment? · 2023
No notice
Artificial Intelligence and Machine Learning Technology Driven Modern Drug Discovery and Development · 2023
No notice
Reaching alignment-profile-based accuracy in predicting protein secondary and tertiary structural properties without alignment · 2022
Not checked
A structural biology community assessment of AlphaFold2 applications · 2022
No notice
Artificial intelligence in cancer target identification and drug discovery · 2022
No notice
Current progress and open challenges for applying deep learning across the biosciences · 2022
No notice
Highly accurate protein structure prediction with AlphaFold · 2021
No notice
Improved Protein Structure Prediction Using a New Multi‐Scale Network and Homologous Templates · 2021
Not checked
Structure-based protein function prediction using graph convolutional networks · 2021
No notice
Generating functional protein variants with variational autoencoders · 2021
No notice
Artificial intelligence and machine learning‐aided drug discovery in central nervous system diseases: State‐of‐the‐arts and future directions · 2020
No notice
Deep Learning in Proteomics · 2020
No notice
An ER translocon for multi-pass membrane protein biogenesis · 2020
No notice
Biological network analysis with deep learning · 2020
No notice
Deep learning methods in protein structure prediction · 2020
No notice
DeepCDpred: Inter-residue distance and contact prediction for improved prediction of protein structure. · 2019
Not checked
Common activation mechanism of class A GPCRs · 2019
No notice
Phenotype Prediction and Genome-Wide Association Study Using Deep Convolutional Neural Network of Soybean · 2019
No notice
RNA secondary structure prediction using an ensemble of two-dimensional deep neural networks and transfer learning · 2019
No notice

and 6 more

Reads

Read by people on this page

None yet.

Checked or read by AI through naulon

None yet.

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.

Are you an author?

Is this your paper?

Claim it to add your note and be paid when agents cite it. Free.