FilterDCA: interpretable supervised contact prediction using inter-domain coevolution

AbstractPredicting 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, reaching results comparable to more complex deep-learning approaches, while remaining fully transparent and interpretable. The FilterDCA code is available at http://gitlab.lcqb.upmc.fr/muscat/FilterDCA.Author summaryThe de novo prediction of tertiary and quaternary protein structures has recently seen important advances, by combining unsupervised, purely sequence-based coevolutionary analyses with structure-based supervision using deep learning for contact-map prediction. While showing impressive performance, deep-learning methods require large training sets and pose severe obstacles for their interpretability. Here we construct a simple, transparent and therefore fully interpretable inter-domain contact predictor, which uses the results of coevolutionary Direct Coupling Analysis in combination with explicitly constructed filters reflecting typical contact patterns in a training set of known protein structures, and which improves the accuracy of predicted contacts significantly. Our approach thereby sheds light on the question how contact information is encoded in coevolutionary signals.

Download Full-text

Accurate contact-based modelling of repeat proteins predicts the structure of new repeats protein families

PLoS Computational Biology ◽

10.1371/journal.pcbi.1008798 ◽

2021 ◽

Vol 17 (4) ◽

pp. e1008798

Author(s):

Claudio Bassot ◽

Arne Elofsson

Keyword(s):

Deep Learning ◽

Protein Structure ◽

High Accuracy ◽

Unique Sequence ◽

Direct Coupling ◽

Protein Families ◽

Coupling Analysis ◽

Repeat Proteins ◽

Eukaryotic Proteomes ◽

Direct Coupling Analysis

Repeat proteins are abundant in eukaryotic proteomes. They are involved in many eukaryotic specific functions, including signalling. For many of these proteins, the structure is not known, as they are difficult to crystallise. Today, using direct coupling analysis and deep learning it is often possible to predict a protein’s structure. However, the unique sequence features present in repeat proteins have been a challenge to use direct coupling analysis for predicting contacts. Here, we show that deep learning-based methods (trRosetta, DeepMetaPsicov (DMP) and PconsC4) overcomes this problem and can predict intra- and inter-unit contacts in repeat proteins. In a benchmark dataset of 815 repeat proteins, about 90% can be correctly modelled. Further, among 48 PFAM families lacking a protein structure, we produce models of forty-one families with estimated high accuracy.

Download Full-text

Large-scale identification of coevolution signals across homo-oligomeric protein interfaces by direct coupling analysis

Proceedings of the National Academy of Sciences ◽

10.1073/pnas.1615068114 ◽

2017 ◽

Vol 114 (13) ◽

pp. E2662-E2671 ◽

Cited By ~ 58

Author(s):

Guido Uguzzoni ◽

Shalini John Lovis ◽

Francesco Oteri ◽

Alexander Schug ◽

Hendrik Szurmant ◽

...

Keyword(s):

Structure Prediction ◽

Large Scale ◽

Structural Models ◽

Direct Coupling ◽

Pfam Family ◽

Sequence Information ◽

Structural Description ◽

Protein Families ◽

Coupling Analysis ◽

Direct Coupling Analysis

Proteins have evolved to perform diverse cellular functions, from serving as reaction catalysts to coordinating cellular propagation and development. Frequently, proteins do not exert their full potential as monomers but rather undergo concerted interactions as either homo-oligomers or with other proteins as hetero-oligomers. The experimental study of such protein complexes and interactions has been arduous. Theoretical structure prediction methods are an attractive alternative. Here, we investigate homo-oligomeric interfaces by tracing residue coevolution via the global statistical direct coupling analysis (DCA). DCA can accurately infer spatial adjacencies between residues. These adjacencies can be included as constraints in structure prediction techniques to predict high-resolution models. By taking advantage of the ongoing exponential growth of sequence databases, we go significantly beyond anecdotal cases of a few protein families and apply DCA to a systematic large-scale study of nearly 2,000 Pfam protein families with sufficient sequence information and structurally resolved homo-oligomeric interfaces. We find that large interfaces are commonly identified by DCA. We further demonstrate that DCA can differentiate between subfamilies with different binding modes within one large Pfam family. Sequence-derived contact information for the subfamilies proves sufficient to assemble accurate structural models of the diverse protein-oligomers. Thus, we provide an approach to investigate oligomerization for arbitrary protein families leading to structural models complementary to often-difficult experimental methods. Combined with ever more abundant sequential data, we anticipate that this study will be instrumental to allow the structural description of many heteroprotein complexes in the future.

Download Full-text

On the use of direct-coupling analysis with a reduced alphabet of amino acids combined with super-secondary structure motifs for protein fold prediction

NAR Genomics and Bioinformatics ◽

10.1093/nargab/lqab027 ◽

2021 ◽

Vol 3 (2) ◽

Author(s):

Bernat Anton ◽

Mireia Besalú ◽

Oriol Fornes ◽

Jaume Bonet ◽

Alexis Molina ◽

...

Keyword(s):

Amino Acids ◽

Protein Structure ◽

Secondary Structure ◽

Protein Structures ◽

Three Dimensional ◽

Direct Coupling ◽

Dimensional Structure ◽

Coupling Analysis ◽

Multiple Sequence ◽

Direct Coupling Analysis

Abstract Direct-coupling analysis (DCA) for studying the coevolution of residues in proteins has been widely used to predict the three-dimensional structure of a protein from its sequence. We present RADI/raDIMod, a variation of the original DCA algorithm that groups chemically equivalent residues combined with super-secondary structure motifs to model protein structures. Interestingly, the simplification produced by grouping amino acids into only two groups (polar and non-polar) is still representative of the physicochemical nature that characterizes the protein structure and it is in line with the role of hydrophobic forces in protein-folding funneling. As a result of a compressed alphabet, the number of sequences required for the multiple sequence alignment is reduced. The number of long-range contacts predicted is limited; therefore, our approach requires the use of neighboring sequence-positions. We use the prediction of secondary structure and motifs of super-secondary structures to predict local contacts. We use RADI and raDIMod, a fragment-based protein structure modelling, achieving near native conformations when the number of super-secondary motifs covers >30–50% of the sequence. Interestingly, although different contacts are predicted with different alphabets, they produce similar structures.

Download Full-text

Protein structure without structure determination: direct coupling analysis based on in vitro evolution

10.1101/582056 ◽

2019 ◽

Author(s):

Marco Fantini ◽

Simonetta Lisi ◽

Paolo De Los Rios ◽

Antonino Cattaneo ◽

Annalisa Pastore

Keyword(s):

Structural Information ◽

Sequence Data ◽

Protein Structures ◽

Direct Coupling ◽

In Vitro Mutagenesis ◽

Evolutionary Analysis ◽

Protein Families ◽

Coupling Analysis ◽

Direct Coupling Analysis

AbstractDirect Coupling Analysis (DCA) is a powerful technique that enables to extract structural information of proteins belonging to large protein families exclusively by in silico analysis. This method is however limited by sequence availability and various biases. Here, we propose a method that exploits molecular evolution to circumvent these limitations: instead of relying on existing protein families, we used in vitro mutagenesis of TEM-1 beta lactamase combined with in vivo functional selection to generate the sequence data necessary for evolutionary analysis. We could reconstruct by this strategy, which we called CAMELS (CouplingAnalysis byMolecularEvolutionLibrarySequencing), the lactamase fold exclusively from sequence data. Through generating and sequencing large libraries of variants, we can deal with any protein, ancient or recent, from any species, having the only constraint of setting up a functional phenotypic selection of the protein. This method allows us to obtain protein structures without solving the structure experimentally.

Download Full-text

Deep learning reveals many more inter-protein residue-residue contacts than direct coupling analysis

10.1101/240754 ◽

2017 ◽

Cited By ~ 4

Author(s):

Tian-ming Zhou ◽

Sheng Wang ◽

Jinbo Xu

Keyword(s):

Deep Learning ◽

Protein Interaction ◽

Interaction Network ◽

Protein Docking ◽

Residue Level ◽

Direct Coupling ◽

Coupling Analysis ◽

Multiple Sequence ◽

Contact Prediction ◽

Direct Coupling Analysis

AbstractIntra-protein residue-level contact prediction has drawn a lot of attentions in recent years and made very good progress, but much fewer methods are dedicated to inter-protein contact prediction, which are important for understanding how proteins interact at structure and residue level. Direct coupling analysis (DCA) is popular for intra-protein contact prediction, but extending it to inter-protein contact prediction is challenging since it requires too many interlogs (i.e., interacting homologs) to be effective, which cannot be easily fulfilled especially for a putative interacting protein pair in eukaryotes. We show that deep learning, even trained by only intra-protein contact maps, works much better than DCA for inter-protein contact prediction. We also show that a phylogeny-based method can generate a better multiple sequence alignment for eukaryotes than existing genome-based methods and thus, lead to better inter-protein contact prediction. Our method shall be useful for protein docking, protein interaction prediction and protein interaction network construction.

Download Full-text

Direct Coupling Analysis of Epistasis in Allosteric Materials

10.1101/519116 ◽

2019 ◽

Cited By ~ 1

Author(s):

Barbara Bravi ◽

Riccardo Ravasio ◽

Carolina Brito ◽

Matthieu Wyart

Keyword(s):

Drug Design ◽

Long Range ◽

Active Site ◽

De Novo ◽

Sequence Data ◽

Direct Coupling ◽

Coupling Analysis ◽

Promising Tool ◽

Direct Coupling Analysis ◽

The Impact

AbstractIn allosteric proteins, the binding of a ligand modifies function at a distant active site. Such al-losteric pathways can be used as target for drug design, generating considerable interest in inferring them from sequence alignment data. Currently, different methods lead to conflicting results, in particular on the existence of long-range evolutionary couplings between distant amino-acids mediating allostery. Here we propose a resolution of this conundrum, by studying epistasis and its inference in models where an allosteric material is evolved in silico to perform a mechanical task. We find four types of epistasis (Synergistic, Sign, Antagonistic, Saturation), which can be both short or long-range and have a simple mechanical interpretation. We perform a Direct Coupling Analysis (DCA) and find that DCA predicts well mutation costs but is a rather poor generative model. Strikingly, it can predict short-range epistasis but fails to capture long-range epistasis, in agreement with empirical findings. We propose that such failure is generic when function requires subparts to work in concert. We illustrate this idea with a simple model, which suggests that other methods may be better suited to capture long-range effects.Author summaryAllostery in proteins is the property of highly specific responses to ligand binding at a distant site. To inform protocols of de novo drug design, it is fundamental to understand the impact of mutations on allosteric regulation and whether it can be predicted from evolutionary correlations. In this work we consider allosteric architectures artificially evolved to optimize the cooperativity of binding at allosteric and active site. We first characterize the emergent pattern of epistasis as well as the underlying mechanical phenomena, finding four types of epistasis (Synergistic, Sign, Antagonistic, Saturation), which can be both short or long-range. The numerical evolution of these allosteric architectures allows us to benchmark Direct Coupling Analysis, a method which relies on co-evolution in sequence data to infer direct evolutionary couplings, in connection to allostery. We show that Direct Coupling Analysis predicts quantitatively mutation costs but underestimates strong long-range epistasis. We provide an argument, based on a simplified model, illustrating the reasons for this discrepancy and we propose neural networks as more promising tool to measure epistasis.

Download Full-text