TY - JOUR
T1 - Protein contact prediction using metagenome sequence data and residual neural networks
AU - Wu, Qi
AU - Peng, Zhenling
AU - Anishchenko, Ivan
AU - Cong, Qian
AU - Baker, David
AU - Yang, Jianyi
N1 - Funding Information:
The work was supported in part by National Natural Science Foundation of China (NSFC 11871290 and 61873185), the Fundamental Research Funds for the Central Universities, Fok Ying-Tong Education Foundation (161003), China Scholarship Council, KLMDASR and the Thousand Youth Talents Plan of China.
Publisher Copyright:
© 2019 The Author(s) 2019. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
PY - 2020/1/1
Y1 - 2020/1/1
N2 - 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.
AB - 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.
UR - http://www.scopus.com/inward/record.url?scp=85076839273&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85076839273&partnerID=8YFLogxK
U2 - 10.1093/bioinformatics/btz477
DO - 10.1093/bioinformatics/btz477
M3 - Article
C2 - 31173061
AN - SCOPUS:85076839273
SN - 1367-4803
VL - 36
SP - 41
EP - 48
JO - Bioinformatics
JF - Bioinformatics
IS - 1
ER -