scholarly journals Predicting direct physical interactions in multimeric proteins with deep learning

2021 ◽  
Author(s):  
Mu Gao ◽  
Davi Nakajima An ◽  
Jerry M Parks ◽  
Jeffrey Skolnick
Keyword(s):  
Protein Interactions ◽  
Protein Complexes ◽  
Network Models ◽  
Protein Docking ◽  
Protein Protein Interactions ◽  
Neural Network Models ◽  
Sequence Alignments ◽  
Multiple Sequence ◽  
Atomic Structures ◽  
Cytochrome C Biogenesis

Accurate descriptions of protein-protein interactions are essential for understanding biological systems. Very recently, AlphaFold2 has been shown to be remarkably accurate for predicting the atomic structures of individual proteins. Here, we demonstrate that the same neural network models developed for AlphaFold2 can be adapted to predict the structures of multimeric protein complexes without retraining. In contrast to common approaches that require paired multiple sequence alignments, our method, AF2Complex, works without using such paired alignments. It achieves higher accuracy than complex strategies that combine AlphaFold2 and protein-protein docking. New metrics are then introduced for predicting direct protein-protein interactions between arbitrary protein pairs. The approach is successfully validated on some challenging CASP14 multimeric targets, a small but appropriate benchmark set, and the E. coli proteome. Lastly, using the cytochrome c biogenesis system as an example, we present high-confidence models of three sought-after assemblies formed by eight members of this system.

scholarly journals Conservation of coevolving protein interfaces bridges prokaryote–eukaryote homologies in the twilight zone

2016 ◽  
Vol 113 (52) ◽  
pp. 15018-15023 ◽  
Author(s):  
Juan Rodriguez-Rivas ◽  
Simone Marsili ◽  
David Juan ◽  
Alfonso Valencia
Keyword(s):  
Protein Interactions ◽  
Protein Complexes ◽  
Accurate Information ◽  
Twilight Zone ◽  
Sequence Information ◽  
Protein Protein Interactions ◽  
Sequence Alignments ◽  
Multiple Sequence ◽  
Protein Interfaces ◽  
Recent Developments

Protein–protein interactions are fundamental for the proper functioning of the cell. As a result, protein interaction surfaces are subject to strong evolutionary constraints. Recent developments have shown that residue coevolution provides accurate predictions of heterodimeric protein interfaces from sequence information. So far these approaches have been limited to the analysis of families of prokaryotic complexes for which large multiple sequence alignments of homologous sequences can be compiled. We explore the hypothesis that coevolution points to structurally conserved contacts at protein–protein interfaces, which can be reliably projected to homologous complexes with distantly related sequences. We introduce a domain-centered protocol to study the interplay between residue coevolution and structural conservation of protein–protein interfaces. We show that sequence-based coevolutionary analysis systematically identifies residue contacts at prokaryotic interfaces that are structurally conserved at the interface of their eukaryotic counterparts. In turn, this allows the prediction of conserved contacts at eukaryotic protein–protein interfaces with high confidence using solely mutational patterns extracted from prokaryotic genomes. Even in the context of high divergence in sequence (the twilight zone), where standard homology modeling of protein complexes is unreliable, our approach provides sequence-based accurate information about specific details of protein interactions at the residue level. Selected examples of the application of prokaryotic coevolutionary analysis to the prediction of eukaryotic interfaces further illustrate the potential of this approach.

scholarly journals Improving Protein Docking with Constraint Programming and Coevolution Data

2014 ◽  
Author(s):  
Ludwig Krippahl ◽  
Fábio Madeira
Keyword(s):  
Protein Interactions ◽  
Constraint Programming ◽  
Protein Complexes ◽  
Protein Sequences ◽  
Protein Docking ◽  
Statistical Parameters ◽  
Sequence Alignments ◽  
Multiple Sequence ◽  
Simplified Approach ◽  
Multiple Sequence Alignments

Background: Constraint programming (CP) is usually seen as a rigid approach, focusing on crisp, precise, distinctions between what is allowed as a solution and what is not. At first sight, this makes it seem inadequate for bioinformatics applications that rely mostly on statistical parameters and optimisation. The prediction of protein interactions, or protein docking, is one such application. And this apparent problem with CP is particularly evident when constraints are provided by noisy data, as it is the case when using the statistical analysis of Multiple Sequence Alignments (MSA) to extract coevolution information. The goal of this paper is to show that this first impression is misleading and that CP is a useful technique for improving protein docking even with data as vague and noisy as the coevolution indicators that can be inferred from MSA. Results: Here we focus on the study of two protein complexes. In one case we used a simplified estimator of interaction propensity to infer a set of five candidate residues for the interface and used that set to constrain the docking models. Even with this simplified approach and considering only the interface of one of the partners, there is a visible focusing of the models around the correct configuration. Considering a set of 400 models with the best geometric contacts, this constraint increases the number of models close to the target (RMSD ¡5Å) from 2 to 5 and decreases the RMSD of all retained models from 26Å to 17.5Å. For the other example we used a more standard estimate of coevolving residues, from the Co-Evolution Analysis using Protein Sequences (CAPS) software. Using a group of three residues identified from the sequence alignment as potentially co-evolving to constrain the search, the number of complexes similar to the target among the 50 highest scoring docking models increased from 3 in the unconstrained docking to 30 in the constrained docking. Conclusions: Although only a proof-of-concept application, our results show that, with suitably designed constraints, CP allows us to integrate coevolution data, which can be inferred from databases of protein sequences, even though the data is noisy and often “fuzzy”, with no well-defined discontinuities. This also shows, more generally, that CP in bioinformatics needs not be limited to the more crisp cases of finite domains and explicit rules but can also be applied to a broader range of problems that depend on statistical measurements and continuous data.

scholarly journals Improved prediction of protein-protein interactions using AlphaFold2 and extended multiple-sequence alignments

2021 ◽  
Author(s):  
Patrick Bryant ◽  
Gabriele Pozzati ◽  
Arne Elofsson
Keyword(s):  
Protein Interactions ◽  
Error Rate ◽  
Protein Complexes ◽  
Sequence Information ◽  
Protein Protein Interactions ◽  
Single Chain ◽  
Sequence Alignments ◽  
Multiple Sequence ◽  
Test Set ◽  
Interacting Protein

Predicting the structure of single-chain proteins is now close to being a solved problem due to the recent achievement of AlphaFold2 (AF2). However, predicting the structure of interacting protein chains is still a challenge. Here, we utilise AF2 to optimise a protocol for predicting the structure of heterodimeric protein complexes using only sequence information. We find that using the default AF2 protocol, 32% of the models in the Dockground test set can be modelled accurately. By tuning the input alignment and identifying the best model, we adjusted the performance to 43%. Our protocol uses MSAs generated by AF2 and MSAs paired on the organism level generated with HHblits. In a more extensive, more realistic, independent test set, the accuracy is 59%. In comparison, the alternative fold-and-dock method RoseTTAFold is only successful in 10% of the cases on this set and traditional docking methods 22%. However, for the traditional method, the performance would be lower if the bound form of both monomers was not known. The success is higher for bacterial protein pairs, pairs with large interaction areas consisting of helices or sheets, and many homologous sequences. We can distinguish acceptable (DockQ>0.23) from incorrect models with an AUC of 0.84 on the test set by analysing the predicted interfaces. At an error rate of 1%, 13% are acceptable (at a 10% error rate, 40% of the models are acceptable). All scripts and tools to run our protocol are freely available at: https://gitlab.com/ElofssonLab/FoldDock.

scholarly journals Mass spectrometry-based cross-linking study shows that the Psb28 protein binds to cytochrome b559 in Photosystem II

2017 ◽  
Vol 114 (9) ◽  
pp. 2224-2229 ◽  
Author(s):  
Daniel A. Weisz ◽  
Haijun Liu ◽  
Hao Zhang ◽  
Sundarapandian Thangapandian ◽  
Emad Tajkhorshid ◽  
...  
Keyword(s):  
Mass Spectrometry ◽  
Photosystem Ii ◽  
Protein Interactions ◽  
Protein Complexes ◽  
Protein Docking ◽  
Cross Linking ◽  
Protein Protein Interactions ◽  
Cytochrome B559 ◽  
Reaction Center Complex ◽  
Assembly Intermediate

Photosystem II (PSII), a large pigment protein complex, undergoes rapid turnover under natural conditions. During assembly of PSII, oxidative damage to vulnerable assembly intermediate complexes must be prevented. Psb28, the only cytoplasmic extrinsic protein in PSII, protects the RC47 assembly intermediate of PSII and assists its efficient conversion into functional PSII. Its role is particularly important under stress conditions when PSII damage occurs frequently. Psb28 is not found, however, in any PSII crystal structure, and its structural location has remained unknown. In this study, we used chemical cross-linking combined with mass spectrometry to capture the transient interaction of Psb28 with PSII. We detected three cross-links between Psb28 and the α- and β-subunits of cytochrome b559, an essential component of the PSII reaction-center complex. These distance restraints enable us to position Psb28 on the cytosolic surface of PSII directly above cytochrome b559, in close proximity to the QB site. Protein–protein docking results also support Psb28 binding in this region. Determination of the Psb28 binding site and other biochemical evidence allow us to propose a mechanism by which Psb28 exerts its protective effect on the RC47 intermediate. This study also shows that isotope-encoded cross-linking with the “mass tags” selection criteria allows confident identification of more cross-linked peptides in PSII than has been previously reported. This approach thus holds promise to identify other transient protein–protein interactions in membrane protein complexes.

scholarly journals Protein-protein docking using learned three-dimensional representations

2019 ◽  
Author(s):  
Georgy Derevyanko ◽  
Guillaume Lamoureux
Keyword(s):  
Protein Interactions ◽  
Network Architecture ◽  
Protein Complexes ◽  
Three Dimensional ◽  
Spatial Arrangement ◽  
Protein Docking ◽  
Protein Protein Interactions ◽  
Translational Invariance ◽  
Shape Complementarity ◽  
Spatial Features

AbstractProtein-protein interactions are determined by a number of hard-to-capture features related to shape complementarity, electrostatics, and hydrophobicity. These features may be intrinsic to the protein or induced by the presence of a partner. A conventional approach to protein-protein docking consists in engineering a small number of spatial features for each protein, and in minimizing the sum of their correlations with respect to the spatial arrangement of the two proteins. To generalize this approach, we introduce a deep neural network architecture that transforms the raw atomic densities of each protein into complex three-dimensional representations. Each point in the volume containing the protein is described by 48 learned features, which are correlated and combined with the features of a second protein to produce a score dependent on the relative position and orientation of the two proteins. The architecture is based on multiple layers of SE(3)-equivariant convolutional neural networks, which provide built-in rotational and translational invariance of the score with respect to the structure of the complex. The model is trained end-to-end on a set of decoy conformations generated from 851 nonredundant protein-protein complexes and is tested on data from the Protein-Protein Docking Benchmark Version 4.0.

scholarly journals InterPep2: global peptide–protein docking using interaction surface templates

2020 ◽  
Vol 36 (8) ◽  
pp. 2458-2465 ◽  
Author(s):  
Isak Johansson-Åkhe ◽  
Claudio Mirabello ◽  
Björn Wallner
Keyword(s):  
Protein Interactions ◽  
Protein Complexes ◽  
Structural Features ◽  
Protein Docking ◽  
Supplementary Information ◽  
Peptide Ligand ◽  
Protein Protein Interactions ◽  
Intrinsically Disordered ◽  
Intrinsically Disordered Regions ◽  
Improved Performance

Abstract Motivation Interactions between proteins and peptides or peptide-like intrinsically disordered regions are involved in many important biological processes, such as gene expression and cell life-cycle regulation. Experimentally determining the structure of such interactions is time-consuming and difficult because of the inherent flexibility of the peptide ligand. Although several prediction-methods exist, most are limited in performance or availability. Results InterPep2 is a freely available method for predicting the structure of peptide–protein interactions. Improved performance is obtained by using templates from both peptide–protein and regular protein–protein interactions, and by a random forest trained to predict the DockQ-score for a given template using sequence and structural features. When tested on 252 bound peptide–protein complexes from structures deposited after the complexes used in the construction of the training and templates sets of InterPep2, InterPep2-Refined correctly positioned 67 peptides within 4.0 Å LRMSD among top10, similar to another state-of-the-art template-based method which positioned 54 peptides correctly. However, InterPep2 displays a superior ability to evaluate the quality of its own predictions. On a previously established set of 27 non-redundant unbound-to-bound peptide–protein complexes, InterPep2 performs on-par with leading methods. The extended InterPep2-Refined protocol managed to correctly model 15 of these complexes within 4.0 Å LRMSD among top10, without using templates from homologs. In addition, combining the template-based predictions from InterPep2 with ab initio predictions from PIPER-FlexPepDock resulted in 22% more near-native predictions compared to the best single method (22 versus 18). Availability and implementation The program is available from: http://wallnerlab.org/InterPep2. Supplementary information Supplementary data are available at Bioinformatics online.

scholarly journals Improved prediction of protein-protein interactions using AlphaFold2

2021 ◽  
Author(s):  
Patrick Bryant ◽  
Gabriele Pozzati ◽  
Arne Elofsson
Keyword(s):  
Protein Interactions ◽  
Protein Complexes ◽  
Interacting Proteins ◽  
Bacterial Protein ◽  
Protein Protein Interactions ◽  
Multiple Sequence ◽  
Test Set ◽  
Interacting Protein ◽  
Identification Rate ◽  
Homologous Sequences

Abstract Predicting the structure of interacting protein chains is fundamental for understanding the function of proteins. Here, we examine the use of AlphaFold2 (AF2) for predicting the structure of heterodimeric protein complexes. We find that using the default AF2 protocol, 44% of the models in a test set can be predicted accurately. However, by optimising the multiple sequence alignment, we can increase the accuracy to 59%. In comparison, the alternative fold-and-dock method RoseTTAFold is only successful in 10% of the cases on this set, template-based docking 35% and traditional docking methods 22%. We can distinguish acceptable (DockQ>0.23) from incorrect models with an AUC of 0.85 on the test set by analysing the predicted interfaces. The success is higher for bacterial protein pairs, pairs with large interaction areas consisting of helices or sheets, and many homologous sequences. Further, we test the possibility to distinguish interacting from non-interacting proteins and find that by analysing the predicted interfaces, we can separate truly interacting from non-interacting proteins with an AUC of 0.82 in the ROC curve, compared to 0.76 with a recently published method. In addition, when using a more realistic negative set, including mammalian proteins, the identification rate remains (AUC=0.83), resulting in that 27% of interactions can be identified at a 1% FPR. All scripts and tools to run our protocol are freely available at: https://gitlab.com/ElofssonLab/FoldDock.

scholarly journals Conservation of co-evolving protein interfaces bridges prokaryote-eukaryote homologies in the twilight zone

2016 ◽  
Author(s):  
Juan Rodriguez-Rivas ◽  
Simone Marsili ◽  
David Juan ◽  
Alfonso Valencia
Keyword(s):  
Protein Interactions ◽  
Homology Modelling ◽  
Accurate Information ◽  
Sequence Information ◽  
Evolutionary Analysis ◽  
Interacting Proteins ◽  
Protein Protein Interactions ◽  
Sequence Alignments ◽  
Multiple Sequence ◽  
Protein Interfaces

AbstractProtein-protein interactions are fundamental for the proper functioning of the cell. As a result, protein interaction surfaces are subject to strong evolutionary constraints. Recent developments have shown that residue co-evolution provides accurate predictions of heterodimeric protein interfaces from sequence information. So far these approaches have been limited to the analysis of families of prokaryotic complexes for which large multiple sequence alignments of homologous sequences can be compiled. We explore the hypothesis that co-evolution points to structurally conserved contacts at protein-protein interfaces, which can be reliably projected to homologous complexes with distantly related sequences. We introduce a novel domain-centred protocol to study the interplay between residue co-evolution and structural conservation of protein-protein interfaces. We show that sequence-based co-evolutionary analysis systematically identifies residue contacts at prokaryotic interfaces that are structurally conserved at the interface of their eukaryotic counterparts. In turn, this allows the prediction of conserved contacts at eukaryotic protein-protein interfaces with high confidence using solely mutational patterns extracted from prokaryotic genomes. Even in the context of high divergence in sequence, where standard homology modelling of protein complexes is unreliable, our approach provides sequence-based accurate information about specific details of protein interactions at the residue level. Selected examples of the application of prokaryotic co-evolutionary analysis to the prediction of eukaryotic interfaces further illustrates the potential of this novel approach.Significance statementInteracting proteins tend to co-evolve through interdependent changes at the interaction interface. This phenomenon leads to patterns of coordinated mutations that can be exploited to systematically predict contacts between interacting proteins in prokaryotes. We explore the hypothesis that co-evolving contacts at protein interfaces are preferentially conserved through long evolutionary periods. We demonstrate that co-evolving residues in prokaryotes identify inter-protein contacts that are particularly well conserved in the corresponding structure of their eukaryotic homologues. Therefore, these contacts have likely been important to maintain protein-protein interactions during evolution. We show that this property can be used to reliably predict interacting residues between eukaryotic proteins with homologues in prokaryotes even if they are very distantly related in sequence.

scholarly journals Text mining for modeling of protein complexes enhanced by machine learning

2020 ◽  
Author(s):  
Varsha D Badal ◽  
Petras J Kundrotas ◽  
Ilya A Vakser
Keyword(s):  
Machine Learning ◽  
Text Mining ◽  
Protein Interactions ◽  
Full Text ◽  
Protein Complexes ◽  
Protein Docking ◽  
Supplementary Information ◽  
Support Vector ◽  
Learning Approaches ◽  
Protein Protein Interactions

Abstract Motivation Procedures for structural modeling of protein-protein complexes (protein docking) produce a number of models which need to be further analyzed and scored. Scoring can be based on independently determined constraints on the structure of the complex, such as knowledge of amino acids essential for the protein interaction. Previously, we showed that text mining of residues in freely available PubMed abstracts of papers on studies of protein-protein interactions may generate such constraints. However, absence of post-processing of the spotted residues reduced usability of the constraints, as a significant number of the residues were not relevant for the binding of the specific proteins. Results We explored filtering of the irrelevant residues by two machine learning approaches, Deep Recursive Neural Network (DRNN) and Support Vector Machine (SVM) models with different training/testing schemes. The results showed that the DRNN model is superior to the SVM model when training is performed on the PMC-OA full-text articles and applied to classification (interface or non-interface) of the residues spotted in the PubMed abstracts. When both training and testing is performed on full-text articles or on abstracts, the performance of these models is similar. Thus, in such cases, there is no need to utilize computationally demanding DRNN approach, which is computationally expensive especially at the training stage. The reason is that SVM success is often determined by the similarity in data/text patterns in the training and the testing sets, whereas the sentence structures in the abstracts are, in general, different from those in the full text articles. Availability The code and the datasets generated in this study are available at https://gitlab.ku.edu/vakser-lab-public/text-mining/-/tree/2020-09-04. Supplementary information Supplementary data are available at Bioinformatics online.

scholarly journals Harnessing protein folding neural networks for peptide-protein docking

2021 ◽  
Author(s):  
Tomer Tsaban ◽  
Julia K Varga ◽  
Orly Avraham ◽  
Ziv Ben Aharon ◽  
Alisa Khramushin ◽  
...  
Keyword(s):  
Neural Networks ◽  
Protein Interactions ◽  
Protein Complexes ◽  
Protein Docking ◽  
Peptide Sequence ◽  
Multiple Sequence ◽  
C Terminus ◽  
Docking Protocol ◽  
Wide Range ◽  
Simple Implementation

Highly accurate protein structure predictions by the recently published deep neural networks such as AlphaFold2 and RoseTTAFold are truly impressive achievements, and will have a tremendous impact far beyond structural biology. If peptide-protein binding can be seen as a final complementing step in the folding of a protein monomer, we reasoned that these approaches might be applicable to the modeling of such interactions. We present a simple implementation of AlphaFold2 to model the structure of peptide-protein interactions, enabled by linking the peptide sequence to the protein c-terminus via a poly glycine linker. We show on a large non-redundant set of 162 peptide-protein complexes that peptide-protein interactions can indeed be modeled accurately. Importantly, prediction is fast and works without multiple sequence alignment information for the peptide partner. We compare performance on a smaller, representative set to the state-of-the-art peptide docking protocol PIPER-FlexPepDock, and describe in detail specific examples that highlight advantages of the two approaches, pointing to possible further improvements and insights in the modeling of peptide-protein interactions. Peptide-mediated interactions play important regulatory roles in functional cells. Thus the present advance holds much promise for significant impact, by bringing into reach a wide range of peptide-protein complexes, and providing important starting points for detailed study and manipulation of many specific interactions.

Sign in / Sign up

Export Citation Format

Share Document