FilterDCA: Interpretable supervised contact prediction using inter-domain coevolution

PLoS Comput Biol. 2020 Oct 9;16(10):e1007621. doi: 10.1371/journal.pcbi.1007621. eCollection 2020 Oct.

Abstract

Predicting three-dimensional protein structure and assembling protein complexes using sequence information belongs to the most prominent tasks in computational biology. Recently substantial progress has been obtained in the case of single proteins using a combination of unsupervised coevolutionary sequence analysis with structurally supervised deep learning. While reaching impressive accuracies in predicting residue-residue contacts, deep learning has a number of disadvantages. The need for large structural training sets limits the applicability to multi-protein complexes; and their deep architecture makes the interpretability of the convolutional neural networks intrinsically hard. Here we introduce FilterDCA, a simpler supervised predictor for inter-domain and inter-protein contacts. It is based on the fact that contact maps of proteins show typical contact patterns, which results from secondary structure and are reflected by patterns in coevolutionary analysis. We explicitly integrate averaged contacts patterns with coevolutionary scores derived by Direct Coupling Analysis, improving performance over standard coevolutionary analysis, while remaining fully transparent and interpretable. The FilterDCA code is available at http://gitlab.lcqb.upmc.fr/muscat/FilterDCA.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Computational Biology / methods*
  • Models, Molecular
  • Protein Conformation*
  • Proteins / chemistry*
  • Sequence Analysis, Protein / methods*
  • Software
  • Supervised Machine Learning

Substances

  • Proteins

Grants and funding

MW acknowledges funding by the EU H2020 research and innovation programme MSCA-RISE-2016 under grant agreement No. 734439 INFERNET. This work has been undertaken partially in the framework of CALSIMLAB and supported by the public grant ANR-11-LABX-0037-01 overseen by the French National Research Agency (ANR) as part of the "Investissements d’Avenir" programme (ANR-11-IDEX-0004-02). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.