{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,6]],"date-time":"2026-03-06T01:42:58Z","timestamp":1772761378714,"version":"3.50.1"},"update-to":[{"DOI":"10.1371\/journal.pcbi.1008291","type":"new_version","label":"New version","source":"publisher","updated":{"date-parts":[[2020,12,10]],"date-time":"2020-12-10T00:00:00Z","timestamp":1607558400000}}],"reference-count":73,"publisher":"Public Library of Science (PLoS)","issue":"11","license":[{"start":{"date-parts":[[2020,11,30]],"date-time":"2020-11-30T00:00:00Z","timestamp":1606694400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.ploscompbiol.org"],"crossmark-restriction":false},"short-container-title":["PLoS Comput Biol"],"abstract":"<jats:p>\n                    Predicting mutation-induced changes in protein thermodynamic stability (\u0394\u0394G) is of great interest in protein engineering, variant interpretation, and protein biophysics. We introduce ThermoNet, a deep, 3D-convolutional neural network (3D-CNN) designed for structure-based prediction of \u0394\u0394Gs upon point mutation. To leverage the image-processing power inherent in CNNs, we treat protein structures as if they were multi-channel 3D images. In particular, the inputs to ThermoNet are uniformly constructed as multi-channel voxel grids based on biophysical properties derived from raw atom coordinates. We train and evaluate ThermoNet with a curated data set that accounts for protein homology and is balanced with direct and reverse mutations; this provides a framework for addressing biases that have likely influenced many previous \u0394\u0394G prediction methods. ThermoNet demonstrates performance comparable to the best available methods on the widely used S\n                    <jats:sup>sym<\/jats:sup>\n                    test set. In addition, ThermoNet accurately predicts the effects of both stabilizing and destabilizing mutations, while most other methods exhibit a strong bias towards predicting destabilization. We further show that homology between S\n                    <jats:sup>sym<\/jats:sup>\n                    and widely used training sets like S2648 and VariBench has likely led to overestimated performance in previous studies. Finally, we demonstrate the practical utility of ThermoNet in predicting the \u0394\u0394Gs for two clinically relevant proteins, p53 and myoglobin, and for pathogenic and benign missense variants from ClinVar. Overall, our results suggest that 3D-CNNs can model the complex, non-linear interactions perturbed by mutations, directly from biophysical properties of atoms.\n                  <\/jats:p>","DOI":"10.1371\/journal.pcbi.1008291","type":"journal-article","created":{"date-parts":[[2020,11,30]],"date-time":"2020-11-30T14:28:41Z","timestamp":1606746521000},"page":"e1008291","update-policy":"https:\/\/doi.org\/10.1371\/journal.pcbi.corrections_policy","source":"Crossref","is-referenced-by-count":120,"title":["Predicting changes in protein thermodynamic stability upon point mutation with deep 3D convolutional neural networks"],"prefix":"10.1371","volume":"16","author":[{"given":"Bian","family":"Li","sequence":"first","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6873-5279","authenticated-orcid":true,"given":"Yucheng T.","family":"Yang","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9743-1795","authenticated-orcid":true,"given":"John A.","family":"Capra","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9746-3719","authenticated-orcid":true,"given":"Mark B.","family":"Gerstein","sequence":"additional","affiliation":[]}],"member":"340","published-online":{"date-parts":[[2020,11,30]]},"reference":[{"issue":"1","key":"pcbi.1008291.ref001","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1080\/10409238.2017.1380596","article-title":"Finding the needle in the haystack: towards solving the protein-folding problem computationally","volume":"53","author":"B Li","year":"2018","journal-title":"Crit Rev Biochem Mol Biol"},{"issue":"7571","key":"pcbi.1008291.ref002","doi-asserted-by":"crossref","first-page":"68","DOI":"10.1038\/nature15393","article-title":"A global reference for human genetic variation","volume":"526","author":"A Auton","year":"2015","journal-title":"Nature"},{"issue":"4","key":"pcbi.1008291.ref003","doi-asserted-by":"crossref","first-page":"263","DOI":"10.1002\/humu.22","article-title":"SNPs, protein structure, and disease","volume":"17","author":"Z Wang","year":"2001","journal-title":"Hum Mutat"},{"issue":"2","key":"pcbi.1008291.ref004","doi-asserted-by":"crossref","first-page":"459","DOI":"10.1016\/j.jmb.2005.08.020","article-title":"Loss of protein structure stability as a major causative factor in monogenic disease","volume":"353","author":"P Yue","year":"2005","journal-title":"J Mol Biol"},{"issue":"7","key":"pcbi.1008291.ref005","doi-asserted-by":"crossref","first-page":"575","DOI":"10.1016\/j.tibs.2019.01.003","article-title":"Biophysical and Mechanistic Models for Disease-Causing Protein Variants","volume":"44","author":"A Stein","year":"2019","journal-title":"Trends Biochem Sci"},{"issue":"7620","key":"pcbi.1008291.ref006","doi-asserted-by":"crossref","first-page":"320","DOI":"10.1038\/nature19946","article-title":"The coming of age of de novo protein design","volume":"537","author":"PS Huang","year":"2016","journal-title":"Nature"},{"issue":"26","key":"pcbi.1008291.ref007","doi-asserted-by":"crossref","first-page":"7364","DOI":"10.1002\/anie.201510054","article-title":"Accurate and Rigorous Prediction of the Changes in Protein Free Energies in a Large-Scale Mutation Scan","volume":"55","author":"V Gapsys","year":"2016","journal-title":"Angew Chem Int Edit"},{"issue":"3","key":"pcbi.1008291.ref008","doi-asserted-by":"crossref","first-page":"830","DOI":"10.1002\/prot.22921","article-title":"Role of conformational sampling in computing mutation-induced changes in protein structure and stability","volume":"79","author":"EH Kellogg","year":"2011","journal-title":"Proteins-Structure Function and Bioinformatics"},{"issue":"34","key":"pcbi.1008291.ref009","doi-asserted-by":"crossref","first-page":"4748","DOI":"10.1021\/acs.biochem.6b00444","article-title":"Protocols for Molecular Modeling with Rosetta3 and RosettaScripts","volume":"55","author":"BJ Bender","year":"2016","journal-title":"Biochemistry"},{"issue":"6","key":"pcbi.1008291.ref010","doi-asserted-by":"crossref","first-page":"466","DOI":"10.1038\/nmeth0607-466","article-title":"Eris: an automated estimator of protein stability","volume":"4","author":"SY Yin","year":"2007","journal-title":"Nature Methods"},{"key":"pcbi.1008291.ref011","doi-asserted-by":"crossref","first-page":"W215","DOI":"10.1093\/nar\/gkr363","article-title":"SDM-a server for predicting effects of mutations on protein stability and malfunction","volume":"39","author":"CL Worth","year":"2011","journal-title":"Nucleic Acids Research"},{"key":"pcbi.1008291.ref012","doi-asserted-by":"crossref","DOI":"10.1186\/1471-2105-12-151","article-title":"PoPMuSiC 2.1: a web server for the estimation of protein stability changes upon mutation and sequence optimality","volume":"12","author":"Y Dehouck","year":"2011","journal-title":"BMC bioinformatics"},{"issue":"2","key":"pcbi.1008291.ref013","doi-asserted-by":"crossref","first-page":"369","DOI":"10.1016\/S0022-2836(02)00442-4","article-title":"Predicting changes in the stability of proteins and protein complexes: A study of more than 1000 mutations","volume":"320","author":"R Guerois","year":"2002","journal-title":"Journal of Molecular Biology"},{"key":"pcbi.1008291.ref014","doi-asserted-by":"crossref","first-page":"W239","DOI":"10.1093\/nar\/gkl190","article-title":"CUPSAT: prediction of protein stability upon point mutations","volume":"34","author":"V Parthiban","year":"2006","journal-title":"Nucleic Acids Research"},{"issue":"19","key":"pcbi.1008291.ref015","doi-asserted-by":"crossref","first-page":"2936","DOI":"10.1093\/bioinformatics\/btw361","article-title":"STRUM: structure-based prediction of protein stability changes upon single-point mutation","volume":"32","author":"LJ Quan","year":"2016","journal-title":"Bioinformatics"},{"issue":"3","key":"pcbi.1008291.ref016","doi-asserted-by":"crossref","first-page":"335","DOI":"10.1093\/bioinformatics\/btt691","article-title":"mCSM: predicting the effects of mutations in proteins using graph-based signatures","volume":"30","author":"DEV Pires","year":"2014","journal-title":"Bioinformatics"},{"key":"pcbi.1008291.ref017","doi-asserted-by":"crossref","first-page":"W306","DOI":"10.1093\/nar\/gki375","article-title":"I-Mutant2.0: predicting stability changes upon mutation from the protein sequence or structure","volume":"33","author":"E Capriotti","year":"2005","journal-title":"Nucleic Acids Research"},{"issue":"18","key":"pcbi.1008291.ref018","doi-asserted-by":"crossref","first-page":"2002","DOI":"10.1093\/bioinformatics\/btn353","article-title":"Accurate prediction of stability changes in protein mutants by combining machine learning with structure based computational mutagenesis","volume":"24","author":"M Masso","year":"2008","journal-title":"Bioinformatics"},{"issue":"17","key":"pcbi.1008291.ref019","doi-asserted-by":"crossref","first-page":"2816","DOI":"10.1093\/bioinformatics\/btv291","article-title":"INPS: predicting the impact of non-synonymous variations on protein stability from sequence","volume":"31","author":"P Fariselli","year":"2015","journal-title":"Bioinformatics"},{"key":"pcbi.1008291.ref020","article-title":"DeepDDG: Predicting the Stability Change of Protein Point Mutations Using Neural Networks","author":"H Cao","year":"2019","journal-title":"Journal of Chemical Information and Modeling"},{"issue":"1","key":"pcbi.1008291.ref021","doi-asserted-by":"crossref","first-page":"204","DOI":"10.1021\/jacs.8b08243","article-title":"Contribution of Cotranslational Folding Defects to Membrane Protein Homeostasis","volume":"141","author":"FJ Roushar","year":"2019","journal-title":"J Am Chem Soc"},{"key":"pcbi.1008291.ref022","doi-asserted-by":"crossref","first-page":"25","DOI":"10.1016\/j.csbj.2018.01.002","article-title":"FoldX as Protein Engineering Tool: Better Than Random Based Approaches?","volume":"16","author":"O Buss","year":"2018","journal-title":"Comput Struct Biotechnol J"},{"issue":"10","key":"pcbi.1008291.ref023","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0046084","article-title":"Assessing Predictors of Changes in Protein Stability upon Mutation Using Self-Consistency","volume":"7","author":"G Thiltgen","year":"2012","journal-title":"Plos One"},{"issue":"21","key":"pcbi.1008291.ref024","doi-asserted-by":"crossref","first-page":"3659","DOI":"10.1093\/bioinformatics\/bty348","article-title":"Quantification of biases in predictions of protein stability changes upon mutations","volume":"34","author":"F Pucci","year":"2018","journal-title":"Bioinformatics"},{"issue":"21","key":"pcbi.1008291.ref025","doi-asserted-by":"crossref","first-page":"3653","DOI":"10.1093\/bioinformatics\/bty340","article-title":"Self-consistency test reveals systematic bias in programs for prediction change of stability upon mutation","volume":"34","author":"DR Usmanova","year":"2018","journal-title":"Bioinformatics"},{"key":"pcbi.1008291.ref026","article-title":"A critical review of five machine learning-based algorithms for predicting protein stability changes upon mutation","author":"J Fang","year":"2019","journal-title":"Brief Bioinform"},{"issue":"19","key":"pcbi.1008291.ref027","doi-asserted-by":"crossref","first-page":"3036","DOI":"10.1093\/bioinformatics\/btx350","article-title":"DeepSite: protein-binding site predictor using 3D-convolutional neural networks","volume":"33","author":"J Jimenez","year":"2017","journal-title":"Bioinformatics"},{"issue":"2","key":"pcbi.1008291.ref028","doi-asserted-by":"crossref","first-page":"287","DOI":"10.1021\/acs.jcim.7b00650","article-title":"KDEEP: Protein-Ligand Absolute Binding Affinity Prediction via 3D-Convolutional Neural Networks","volume":"58","author":"J Jimenez","year":"2018","journal-title":"J Chem Inf Model"},{"issue":"7553","key":"pcbi.1008291.ref029","doi-asserted-by":"crossref","first-page":"436","DOI":"10.1038\/nature14539","article-title":"Deep learning","volume":"521","author":"Y LeCun","year":"2015","journal-title":"Nature"},{"issue":"1","key":"pcbi.1008291.ref030","doi-asserted-by":"crossref","first-page":"302","DOI":"10.1186\/s12859-017-1702-0","article-title":"3D deep convolutional neural networks for amino acid environment similarity analysis","volume":"18","author":"W Torng","year":"2017","journal-title":"BMC bioinformatics"},{"issue":"9","key":"pcbi.1008291.ref031","doi-asserted-by":"crossref","first-page":"1503","DOI":"10.1093\/bioinformatics\/bty813","article-title":"High precision protein functional site detection using 3D convolutional neural networks","volume":"35","author":"W Torng","year":"2019","journal-title":"Bioinformatics"},{"key":"pcbi.1008291.ref032","unstructured":"Wallach I, Dzamba M, Heifets A. AtomNet: A Deep Convolutional Neural Network for Bioactivity Prediction in Structure-based Drug Discovery. arXiv. 2015; arXiv:1510.02855"},{"key":"pcbi.1008291.ref033","doi-asserted-by":"crossref","first-page":"545","DOI":"10.1016\/B978-0-12-381270-4.00019-6","article-title":"ROSETTA3: an object-oriented software suite for the simulation and design of macromolecules","volume":"487","author":"A Leaver-Fay","year":"2011","journal-title":"Methods Enzymol"},{"issue":"19","key":"pcbi.1008291.ref034","doi-asserted-by":"crossref","first-page":"2537","DOI":"10.1093\/bioinformatics\/btp445","article-title":"Fast and accurate predictions of protein stability changes upon mutations using statistical potentials and neural networks: PoPMuSiC-2.0","volume":"25","author":"Y Dehouck","year":"2009","journal-title":"Bioinformatics"},{"issue":"12","key":"pcbi.1008291.ref035","doi-asserted-by":"crossref","first-page":"1589","DOI":"10.1093\/bioinformatics\/btg224","article-title":"PISCES: a protein sequence culling server","volume":"19","author":"G Wang","year":"2003","journal-title":"Bioinformatics"},{"issue":"4","key":"pcbi.1008291.ref036","doi-asserted-by":"crossref","DOI":"10.3390\/ijms19041009","article-title":"PON-tstab: Protein Variant Stability Predictor. Importance of Training Data Quality","volume":"19","author":"Y Yang","year":"2018","journal-title":"Int J Mol Sci"},{"issue":"Suppl 14","key":"pcbi.1008291.ref037","doi-asserted-by":"crossref","first-page":"335","DOI":"10.1186\/s12859-019-2923-1","article-title":"DDGun: an untrained method for the prediction of protein stability changes upon single and multiple point variations","volume":"20","author":"L Montanucci","year":"2019","journal-title":"BMC bioinformatics"},{"issue":"14","key":"pcbi.1008291.ref038","doi-asserted-by":"crossref","first-page":"2525","DOI":"10.1093\/bioinformatics\/bty979","article-title":"On the biases in predictions of protein stability changes upon variations: the INPS test case","volume":"35","author":"L Montanucci","year":"2019","journal-title":"Bioinformatics"},{"issue":"2","key":"pcbi.1008291.ref039","doi-asserted-by":"crossref","DOI":"10.1063\/1.4947493","article-title":"High-quality Thermodynamic Data on the Stability Changes of Proteins Upon Single-site Mutations","volume":"45","author":"F Pucci","year":"2016","journal-title":"Journal of Physical and Chemical Reference Data"},{"issue":"6","key":"pcbi.1008291.ref040","doi-asserted-by":"crossref","first-page":"607","DOI":"10.1002\/humu.10081","article-title":"The IARC TP53 database: new online mutation analysis and recommendations to users","volume":"19","author":"M Olivier","year":"2002","journal-title":"Hum Mutat"},{"issue":"Pt 20","key":"pcbi.1008291.ref041","doi-asserted-by":"crossref","first-page":"3441","DOI":"10.1242\/jeb.01172","article-title":"Myoglobin: an essential hemoprotein in striated muscle","volume":"207","author":"GA Ordway","year":"2004","journal-title":"J Exp Biol"},{"issue":"10 Pt A","key":"pcbi.1008291.ref042","doi-asserted-by":"crossref","first-page":"1239","DOI":"10.1016\/j.bbapap.2015.06.002","article-title":"Towards a \"Golden Standard\" for computing globin stability: Stability and structure sensitivity of myoglobin mutants","volume":"1854","author":"KP Kepp","year":"2015","journal-title":"Biochimica et biophysica acta"},{"issue":"2","key":"pcbi.1008291.ref043","doi-asserted-by":"crossref","first-page":"607","DOI":"10.1016\/j.jmb.2010.11.008","article-title":"Alternate states of proteins revealed by detailed energy landscape mapping","volume":"405","author":"MD Tyka","year":"2011","journal-title":"J Mol Biol"},{"issue":"Suppl 8","key":"pcbi.1008291.ref044","doi-asserted-by":"crossref","first-page":"S8","DOI":"10.1186\/1471-2105-10-S8-S8","article-title":"Correlating protein function and stability through the analysis of single amino acid substitutions","volume":"10","author":"Y Bromberg","year":"2009","journal-title":"BMC bioinformatics"},{"issue":"1","key":"pcbi.1008291.ref045","doi-asserted-by":"crossref","first-page":"4480","DOI":"10.1038\/s41598-018-22531-2","article-title":"Prediction and interpretation of deleterious coding variants in terms of protein structural stability","volume":"8","author":"F Ancien","year":"2018","journal-title":"Sci Rep"},{"issue":"D1","key":"pcbi.1008291.ref046","doi-asserted-by":"crossref","first-page":"D1062","DOI":"10.1093\/nar\/gkx1153","article-title":"ClinVar: improving access to variant interpretations and supporting evidence","volume":"46","author":"MJ Landrum","year":"2018","journal-title":"Nucleic Acids Res"},{"issue":"9","key":"pcbi.1008291.ref047","doi-asserted-by":"crossref","first-page":"678","DOI":"10.1038\/nrg1672","article-title":"Missense meanderings in sequence space: a biophysical view of protein evolution","volume":"6","author":"MA DePristo","year":"2005","journal-title":"Nat Rev Genet"},{"issue":"5","key":"pcbi.1008291.ref048","first-page":"1856","article-title":"On the critical review of five machine learning-based algorithms for predicting protein stability changes upon mutation","volume":"21","author":"C Savojardo","year":"2019","journal-title":"Brief Bioinform"},{"issue":"1","key":"pcbi.1008291.ref049","doi-asserted-by":"crossref","first-page":"458","DOI":"10.1016\/j.ifacol.2015.05.068","article-title":"Symmetry Principles in Optimization Problems: an application to Protein Stability Prediction","volume":"48","author":"F Pucci","year":"2015","journal-title":"IFAC-PapersOnLine"},{"key":"pcbi.1008291.ref050","unstructured":"Boomsma W, Frellsen J, editors. Spherical convolutions and their application in molecular modelling. Advances in Neural Information Processing Systems; 2017; p3433\u20133443"},{"issue":"12","key":"pcbi.1008291.ref051","doi-asserted-by":"crossref","first-page":"1792","DOI":"10.1016\/j.drudis.2017.08.004","article-title":"Structural coverage of the proteome for pharmaceutical applications","volume":"22","author":"JC Somody","year":"2017","journal-title":"Drug Discov Today"},{"issue":"8","key":"pcbi.1008291.ref052","doi-asserted-by":"crossref","first-page":"801","DOI":"10.1038\/nmeth.3027","article-title":"Deep mutational scanning: a new style of protein science","volume":"11","author":"DM Fowler","year":"2014","journal-title":"Nature Methods"},{"key":"pcbi.1008291.ref053","doi-asserted-by":"crossref","first-page":"D204","DOI":"10.1093\/nar\/gkj103","article-title":"ProTherm and ProNIT: thermodynamic databases for proteins and protein-nucleic acid interactions","volume":"34","author":"MDS Kumar","year":"2006","journal-title":"Nucleic Acids Research"},{"issue":"3","key":"pcbi.1008291.ref054","doi-asserted-by":"crossref","first-page":"403","DOI":"10.1016\/S0022-2836(05)80360-2","article-title":"Basic local alignment search tool","volume":"215","author":"SF Altschul","year":"1990","journal-title":"J Mol Biol"},{"issue":"D1","key":"pcbi.1008291.ref055","first-page":"D271","article-title":"The RCSB protein data bank: integrative view of protein, gene and 3D structural information","volume":"45","author":"PW Rose","year":"2017","journal-title":"Nucleic Acids Research"},{"issue":"6","key":"pcbi.1008291.ref056","doi-asserted-by":"crossref","first-page":"3031","DOI":"10.1021\/acs.jctc.7b00125","article-title":"The Rosetta All-Atom Energy Function for Macromolecular Modeling and Design","volume":"13","author":"RF Alford","year":"2017","journal-title":"J Chem Theory Comput"},{"issue":"16","key":"pcbi.1008291.ref057","doi-asserted-by":"crossref","first-page":"2785","DOI":"10.1002\/jcc.21256","article-title":"AutoDock4 and AutoDockTools4: Automated Docking with Selective Receptor Flexibility","volume":"30","author":"GM Morris","year":"2009","journal-title":"Journal of Computational Chemistry"},{"issue":"4","key":"pcbi.1008291.ref058","doi-asserted-by":"crossref","first-page":"1845","DOI":"10.1021\/acs.jctc.6b00049","article-title":"HTMD: High-Throughput Molecular Dynamics for Molecular Discovery","volume":"12","author":"S Doerr","year":"2016","journal-title":"Journal of Chemical Theory and Computation"},{"issue":"12","key":"pcbi.1008291.ref059","doi-asserted-by":"crossref","first-page":"1092","DOI":"10.1002\/prot.25779","article-title":"Prediction of interresidue contacts with DeepMetaPSICOV in CASP13","volume":"87","author":"SM Kandathil","year":"2019","journal-title":"Proteins"},{"issue":"Suppl 1","key":"pcbi.1008291.ref060","doi-asserted-by":"crossref","first-page":"51","DOI":"10.1002\/prot.25407","article-title":"Assessment of contact predictions in CASP12: Co-evolution and deep learning coming of age","volume":"86","author":"J Schaarschmidt","year":"2018","journal-title":"Proteins"},{"key":"pcbi.1008291.ref061","doi-asserted-by":"crossref","first-page":"67","DOI":"10.1002\/prot.25377","article-title":"Analysis of deep learning methods for blind protein contact prediction in CASP12","volume":"86","author":"S Wang","year":"2018","journal-title":"Proteins-Structure Function and Bioinformatics"},{"issue":"12","key":"pcbi.1008291.ref062","doi-asserted-by":"crossref","first-page":"1058","DOI":"10.1002\/prot.25819","article-title":"Assessing the accuracy of contact predictions in CASP13","volume":"87","author":"R Shrestha","year":"2019","journal-title":"Proteins"},{"issue":"34","key":"pcbi.1008291.ref063","doi-asserted-by":"crossref","first-page":"16856","DOI":"10.1073\/pnas.1821309116","article-title":"Distance-based protein folding powered by deep learning","volume":"116","author":"J Xu","year":"2019","journal-title":"Proc Natl Acad Sci U S A"},{"issue":"12","key":"pcbi.1008291.ref064","doi-asserted-by":"crossref","first-page":"1069","DOI":"10.1002\/prot.25810","article-title":"Analysis of distance-based protein structure prediction by deep learning in CASP13","volume":"87","author":"J Xu","year":"2019","journal-title":"Proteins"},{"issue":"12","key":"pcbi.1008291.ref065","doi-asserted-by":"crossref","first-page":"1179","DOI":"10.1002\/prot.25824","article-title":"Recent developments in deep learning applied to protein structure prediction","volume":"87","author":"SM Kandathil","year":"2019","journal-title":"Proteins"},{"issue":"1","key":"pcbi.1008291.ref066","doi-asserted-by":"crossref","first-page":"3977","DOI":"10.1038\/s41467-019-11994-0","article-title":"Deep learning extends de novo protein modelling coverage of genomes using iteratively predicted structural constraints","volume":"10","author":"JG Greener","year":"2019","journal-title":"Nat Commun"},{"key":"pcbi.1008291.ref067","unstructured":"Chollet F. keras. \\url{https:\/\/github.com\/fchollet\/keras}; 2015."},{"key":"pcbi.1008291.ref068","unstructured":"Abadi M, Barham P, Chen J, Chen Z, Davis A, Dean J, et al. TensorFlow: a system for large-scale machine learning. Proceedings of the 12th USENIX conference on Operating Systems Design and Implementation; Savannah, GA, USA. 3026899: USENIX Association; 2016. p. 265\u201383."},{"key":"pcbi.1008291.ref069","volume-title":"Deep Learning with Python","author":"F Chollet","year":"2018"},{"key":"pcbi.1008291.ref070","unstructured":"Kingma DP, Ba JL. Adam: a method for stochastic optimization. 2015;arXiv:1412.6980"},{"issue":"1","key":"pcbi.1008291.ref071","doi-asserted-by":"crossref","first-page":"122","DOI":"10.1186\/s13059-016-0974-4","article-title":"The Ensembl Variant Effect Predictor","volume":"17","author":"W McLaren","year":"2016","journal-title":"Genome Biol"},{"issue":"D1","key":"pcbi.1008291.ref072","doi-asserted-by":"crossref","first-page":"D482","DOI":"10.1093\/nar\/gky1114","article-title":"SIFTS: updated Structure Integration with Function, Taxonomy and Sequences resource allows 40-fold increase in coverage of structure-based annotations for proteins","volume":"47","author":"JM Dana","year":"2019","journal-title":"Nucleic Acids Res"},{"key":"pcbi.1008291.ref073","doi-asserted-by":"crossref","first-page":"539","DOI":"10.1038\/msb.2011.75","article-title":"Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega","volume":"7","author":"F Sievers","year":"2011","journal-title":"Mol Syst Biol"}],"updated-by":[{"DOI":"10.1371\/journal.pcbi.1008291","type":"new_version","label":"New version","source":"publisher","updated":{"date-parts":[[2020,12,10]],"date-time":"2020-12-10T00:00:00Z","timestamp":1607558400000}}],"container-title":["PLOS Computational Biology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pcbi.1008291","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2020,12,10]],"date-time":"2020-12-10T15:48:10Z","timestamp":1607615290000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pcbi.1008291"}},"subtitle":[],"editor":[{"given":"Piero","family":"Fariselli","sequence":"first","affiliation":[]}],"short-title":[],"issued":{"date-parts":[[2020,11,30]]},"references-count":73,"journal-issue":{"issue":"11","published-online":{"date-parts":[[2020,11,30]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pcbi.1008291","relation":{"has-preprint":[{"id-type":"doi","id":"10.1101\/2020.02.28.959874","asserted-by":"object"}]},"ISSN":["1553-7358"],"issn-type":[{"value":"1553-7358","type":"electronic"}],"subject":[],"published":{"date-parts":[[2020,11,30]]}}}