Qi Wu1, Zhenling Peng2, Ivan Anishchenko3,4, Qian Cong3,4, David Baker3,4, Jianyi Yang1. 1. School of Mathematical Sciences, Nankai University, Tianjin 300071, China. 2. Center for Applied Mathematics, Tianjin University, Tianjin, China. 3. Department of Biochemistry, Seattle, WA 98105, USA. 4. Institute for Protein Design, University of Washington, Seattle, WA 98105, USA.
Abstract
MOTIVATION: Almost all protein residue contact prediction methods rely on the availability of deep multiple sequence alignments (MSAs). However, many proteins from the poorly populated families do not have sufficient number of homologs in the conventional UniProt database. Here we aim to solve this issue by exploring the rich sequence data from the metagenome sequencing projects. RESULTS: Based on the improved MSA constructed from the metagenome sequence data, we developed MapPred, a new deep learning-based contact prediction method. MapPred consists of two component methods, DeepMSA and DeepMeta, both trained with the residual neural networks. DeepMSA was inspired by the recent method DeepCov, which was trained on 441 matrices of covariance features. By considering the symmetry of contact map, we reduced the number of matrices to 231, which makes the training more efficient in DeepMSA. Experiments show that DeepMSA outperforms DeepCov by 10-13% in precision. DeepMeta works by combining predicted contacts and other sequence profile features. Experiments on three benchmark datasets suggest that the contribution from the metagenome sequence data is significant with P-values less than 4.04E-17. MapPred is shown to be complementary and comparable the state-of-the-art methods. The success of MapPred is attributed to three factors: the deeper MSA from the metagenome sequence data, improved feature design in DeepMSA and optimized training by the residual neural networks. AVAILABILITY AND IMPLEMENTATION: http://yanglab.nankai.edu.cn/mappred/. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.
MOTIVATION: Almost all protein residue contact prediction methods rely on the availability of deep multiple sequence alignments (MSAs). However, many proteins from the poorly populated families do not have sufficient number of homologs in the conventional UniProt database. Here we aim to solve this issue by exploring the rich sequence data from the metagenome sequencing projects. RESULTS: Based on the improved MSA constructed from the metagenome sequence data, we developed MapPred, a new deep learning-based contact prediction method. MapPred consists of two component methods, DeepMSA and DeepMeta, both trained with the residual neural networks. DeepMSA was inspired by the recent method DeepCov, which was trained on 441 matrices of covariance features. By considering the symmetry of contact map, we reduced the number of matrices to 231, which makes the training more efficient in DeepMSA. Experiments show that DeepMSA outperforms DeepCov by 10-13% in precision. DeepMeta works by combining predicted contacts and other sequence profile features. Experiments on three benchmark datasets suggest that the contribution from the metagenome sequence data is significant with P-values less than 4.04E-17. MapPred is shown to be complementary and comparable the state-of-the-art methods. The success of MapPred is attributed to three factors: the deeper MSA from the metagenome sequence data, improved feature design in DeepMSA and optimized training by the residual neural networks. AVAILABILITY AND IMPLEMENTATION: http://yanglab.nankai.edu.cn/mappred/. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.
Authors: Faruck Morcos; Andrea Pagnani; Bryan Lunt; Arianna Bertolino; Debora S Marks; Chris Sander; Riccardo Zecchina; José N Onuchic; Terence Hwa; Martin Weigt Journal: Proc Natl Acad Sci U S A Date: 2011-11-21 Impact factor: 11.205
Authors: Milot Mirdita; Lars von den Driesch; Clovis Galiez; Maria J Martin; Johannes Söding; Martin Steinegger Journal: Nucleic Acids Res Date: 2016-11-28 Impact factor: 16.971
Authors: Thomas A Hopf; Lucy J Colwell; Robert Sheridan; Burkhard Rost; Chris Sander; Debora S Marks Journal: Cell Date: 2012-05-10 Impact factor: 41.582
Authors: Owen N Vickery; Satchal K Erramilli; Carmen M Herrera; Thomas H McConville; Khuram U Ashraf; Rie Nygaard; Vasileios I Petrou; Sabrina I Giacometti; Meagan Belcher Dufrisne; Kamil Nosol; Allen P Zinkle; Chris L B Graham; Michael Loukeris; Brian Kloss; Karolina Skorupinska-Tudek; Ewa Swiezewska; David I Roper; Oliver B Clarke; Anne-Catrin Uhlemann; Anthony A Kossiakoff; M Stephen Trent; Phillip J Stansfeld; Filippo Mancia Journal: Nature Date: 2022-04-06 Impact factor: 69.504
Authors: Jose Alberto de la Paz; Charisse M Nartey; Monisha Yuvaraj; Faruck Morcos Journal: Proc Natl Acad Sci U S A Date: 2020-03-02 Impact factor: 11.205
Authors: Daniel P Farrell; Ivan Anishchenko; Shabih Shakeel; Anna Lauko; Lori A Passmore; David Baker; Frank DiMaio Journal: IUCrJ Date: 2020-08-20 Impact factor: 4.769