A deep dilated convolutional residual network for predicting interchain contacts of protein homodimers

Deep learning has revolutionized protein tertiary structure prediction recently. The cutting-edge deep learning methods such as AlphaFold can predict high-accuracy tertiary structures for most individual protein chains. However, the accuracy of predicting quaternary structures of protein complexes consisting of multiple chains is still relatively low due to lack of advanced deep learning methods in the field. Because interchain residue-residue contacts can be used as distance restraints to guide quaternary structure modeling, here we develop a deep dilated convolutional residual network method (DRCon) to predict interchain residue-residue contacts in homodimers from residue-residue co-evolutionary signals derived from multiple sequence alignments of monomers, intrachain residue-residue contacts of monomers extracted from true/predicted tertiary structures or predicted by deep learning, and other sequence and structural features. Tested on three homodimer test datasets (Homo_std dataset, DeepHomo dataset, and CASP14-CAPRI dataset), the precision of DRCon for top L/5 interchain contact predictions (L: length of monomer in a homodimer) is 43.46%, 47.15%, and 24.81% respectively, which is substantially better than two existing deep learning interchain contact prediction methods. Moreover, our experiments demonstrate that using predicted tertiary structure or intrachain contacts of monomers in the unbound state as input, DRCon still performs reasonably well, even though its accuracy is lower than when true tertiary structures in the bound state are used as input. Finally, our case study shows that good interchain contact predictions can be used to build high-accuracy quaternary structure models of homodimers.

COMTOP: Protein Residue–Residue Contact Prediction through Mixed Integer Linear Optimization

Protein contact prediction helps reconstruct the tertiary structure that greatly determines a protein’s function; therefore, contact prediction from the sequence is an important problem. Recently there has been exciting progress on this problem, but many of the existing methods are still low quality of prediction accuracy. In this paper, we present a new mixed integer linear programming (MILP)-based consensus method: a Consensus scheme based On a Mixed integer linear opTimization method for prOtein contact Prediction (COMTOP). The MILP-based consensus method combines the strengths of seven selected protein contact prediction methods, including CCMpred, EVfold, DeepCov, NNcon, PconsC4, plmDCA, and PSICOV, by optimizing the number of correctly predicted contacts and achieving a better prediction accuracy. The proposed hybrid protein residue–residue contact prediction scheme was tested in four independent test sets. For 239 highly non-redundant proteins, the method showed a prediction accuracy of 59.68%, 70.79%, 78.86%, 89.04%, 94.51%, and 97.35% for top-5L, top-3L, top-2L, top-L, top-L/2, and top-L/5 contacts, respectively. When tested on the CASP13 and CASP14 test sets, the proposed method obtained accuracies of 75.91% and 77.49% for top-L/5 predictions, respectively. COMTOP was further tested on 57 non-redundant ɑ-helical transmembrane proteins and achieved prediction accuracies of 64.34% and 73.91% for top-L/2 and top-L/5 predictions, respectively. For all test datasets, the improvement of COMTOP in accuracy over the seven individual methods increased with the increasing number of predicted contacts. For example, COMTOP performed much better for large number of contact predictions (such as top-5L and top-3L) than for small number of contact predictions such as top-L/2 and top-L/5. The results and analysis demonstrate that COMTOP can significantly improve the performance of the individual methods; therefore, COMTOP is more robust against different types of test sets. COMTOP also showed better/comparable predictions when compared with the state-of-the-art predictors.

On the effect of phylogenetic correlations in coevolution-based contact prediction in proteins

Coevolution-based contact prediction, either directly by coevolutionary couplings resulting from global statistical sequence models or using structural supervision and deep learning, has found widespread application in protein-structure prediction from sequence. However, one of the basic assumptions in global statistical modeling is that sequences form an at least approximately independent sample of an unknown probability distribution, which is to be learned from data. In the case of protein families, this assumption is obviously violated by phylogenetic relations between protein sequences. It has turned out to be notoriously difficult to take phylogenetic correlations into account in coevolutionary model learning. Here, we propose a complementary approach: we develop strategies to randomize or resample sequence data, such that conservation patterns and phylogenetic relations are preserved, while intrinsic (i.e. structure- or function-based) coevolutionary couplings are removed. A comparison between the results of Direct Coupling Analysis applied to real and to resampled data shows that the largest coevolutionary couplings, i.e. those used for contact prediction, are only weakly influenced by phylogeny. However, the phylogeny-induced spurious couplings in the resampled data are compatible in size with the first false-positive contact predictions from real data. Dissecting functional from phylogeny-induced couplings might therefore extend accurate contact predictions to the range of intermediate-size couplings.

The whole is greater than its parts: ensembling improves protein contact prediction

The prediction of amino acid contacts from protein sequence is an important problem, as protein contacts are a vital step towards the prediction of folded protein structures. We propose that a powerful concept from deep learning, called ensembling, can increase the accuracy of protein contact predictions by combining the outputs of different neural network models. We show that ensembling the predictions made by different groups at the recent Critical Assessment of Protein Structure Prediction (CASP13) outperforms all individual groups. Further, we show that contacts derived from the distance predictions of three additional deep neural networks—AlphaFold, trRosetta, and ProSPr—can be substantially improved by ensembling all three networks. We also show that ensembling these recent deep neural networks with the best CASP13 group creates a superior contact prediction tool. Finally, we demonstrate that two ensembled networks can successfully differentiate between the folds of two highly homologous sequences. In order to build further on these findings, we propose the creation of a better protein contact benchmark set and additional open-source contact prediction methods.

Protein Model Accuracy Estimation Empowered by Deep Learning and Inter-residue Distance Prediction in CASP14

The inter-residue contact prediction and deep learning showed the promise to improve the estimation of protein model accuracy (CASP13) in the 13th Critical Assessment of Protein Structure Prediction (CASP13). During the 2020 CASP14 experiment, we developed and tested several EMA predictors that used deep learning with the new features based on inter-residue distance/contact predictions as well as the existing model quality features. The average global distance test (GDT-TS) score loss of ranking CASP14 structural models by three multi-model MULTICOM EMA predictors (MULTICOM-CONSTRUCT, MULTICOM-AI, and MULTICOM-CLUSTER) is 0.073, 0.079, and 0.081, respectively, which are ranked first, second, and third places out of 68 CASP14 EMA predictors. The single-model EMA predictor (MULTICOM-DEEP) is ranked 10th place among all the single-model EMA methods in terms of GDT-TS score loss. The results show that deep learning and contact/distance predictions are useful in ranking and selecting protein structural models.

pyconsFold: A fast and easy tool for modelling and docking using distance predictions

MotivationContact predictions within a protein has recently become a viable method for accurate prediction of protein structure. Using predicted distance distributions has been shown in many cases to be superior to only using a binary contact annotation. Using predicted inter-protein distances has also been shown to be able to dock some protein dimers.ResultsHere we present pyconsFold. Using CNS as its underlying folding mechanism and predicted contact distance it outperforms regular contact prediction based modelling on our dataset of 210 proteins. It performs marginally worse than the state of the art pyRosetta folding pipeline but is on average about 20 times faster per model. More importantly pyconsFold can also be used as a fold-and-dock protocol by using predicted inter-protein contacts to simultaneously fold and dock two protein chains.Availability and implementationpyconsFold is implemented in Python 3 with a strong focus on using as few dependencies as possible for longevity. It is available both as a pip package in Python 3 and as source code on GitHub and is published under the GPLv3 [email protected] materialInstall instructions, examples and parameters can be found in the supplemental notes.Availability of dataThe data underlying this article together with source code are available on github, at https://github.com/johnlamb/pyconsfold.

Protein model accuracy estimation empowered by deep learning and inter-residue distance prediction in CASP14

The inter-residue contact prediction and deep learning showed the promise to improve the estimation of protein model accuracy (EMA) in the 13th Critical Assessment of Protein Structure Prediction (CASP13). During the 2020 CASP14 experiment, we developed and tested several EMA predictors that used deep learning with the new features based on inter-residue distance/contact predictions as well as the existing model quality features. The average global distance test (GDT-TS) score loss of ranking CASP14 structural models by three multi-model MULTICOM EMA predictors (MULTICOM-CONSTRUCT, MULTICOM-AI, and MULTICOM-CLUSTER) is 0.073, 0.079, and 0.081, respectively, which are ranked first, second, and third places out of 68 CASP14 EMA predictors. The single-model EMA predictor (MULTICOM-DEEP) is ranked 10th place among all the single-model EMA methods in terms of GDT_TS score loss. The results show that deep learning and contact/distance predictions are useful in ranking and selecting protein structural models.

DeepDist: real-value inter-residue distance prediction with deep residual convolutional network

Background Driven by deep learning, inter-residue contact/distance prediction has been significantly improved and substantially enhanced ab initio protein structure prediction. Currently, most of the distance prediction methods classify inter-residue distances into multiple distance intervals instead of directly predicting real-value distances. The output of the former has to be converted into real-value distances to be used in tertiary structure prediction. Results To explore the potentials of predicting real-value inter-residue distances, we develop a multi-task deep learning distance predictor (DeepDist) based on new residual convolutional network architectures to simultaneously predict real-value inter-residue distances and classify them into multiple distance intervals. Tested on 43 CASP13 hard domains, DeepDist achieves comparable performance in real-value distance prediction and multi-class distance prediction. The average mean square error (MSE) of DeepDist’s real-value distance prediction is 0.896 Å2 when filtering out the predicted distance ≥ 16 Å, which is lower than 1.003 Å2 of DeepDist’s multi-class distance prediction. When distance predictions are converted into contact predictions at 8 Å threshold (the standard threshold in the field), the precision of top L/5 and L/2 contact predictions of DeepDist’s multi-class distance prediction is 79.3% and 66.1%, respectively, higher than 78.6% and 64.5% of its real-value distance prediction and the best results in the CASP13 experiment. Conclusions DeepDist can predict inter-residue distances well and improve binary contact prediction over the existing state-of-the-art methods. Moreover, the predicted real-value distances can be directly used to reconstruct protein tertiary structures better than multi-class distance predictions due to the lower MSE. Finally, we demonstrate that predicting the real-value distance map and multi-class distance map at the same time performs better than predicting real-value distances alone.

Combination of deep neural network with attention mechanism enhances the explainability of protein contact prediction

Deep learning has emerged as a revolutionary technology for protein residue-residue contact prediction since the 2012 CASP10 competition. Considerable advancements in the predictive power of the deep learning-based contact predictions have been achieved since then. However, little effort has been put into interpreting the black-box deep learning methods. Algorithms that can interpret the relationship between predicted contact maps and the internal mechanism of the deep learning architectures are needed to explore the essential components of contact inference and improve their explainability. In this study, we present an attention-based convolutional neural network for protein contact prediction, which consists of two attention mechanism-based modules: sequence attention and regional attention. Our benchmark results on the CASP13 free-modeling (FM) targets demonstrate that the two attention modules added on top of existing typical deep learning models exhibit a complementary effect that contributes to predictive improvements. More importantly, the inclusion of the attention mechanism provides interpretable patterns that contain useful insights into the key fold-determining residues in proteins. We expect the attention-based model can provide a reliable and practically interpretable technique that helps break the current bottlenecks in explaining deep neural networks for contact prediction.

DeepDist: real-value inter-residue distance prediction with deep residual convolutional network

MotivationDriven by deep learning techniques, inter-residue contact/distance prediction has been significantly improved and substantially enhanced ab initio protein structure prediction. Currently all the distance prediction methods classify inter-residue distances into multiple distance intervals (i.e. a multi-classification problem) instead of directly predicting real-value distances (i.e. a regression problem). The output of the former has to be converted into real-value distances in order to be used in tertiary structure prediction.ResultsTo explore the potentials of predicting real-value inter-residue distances, we develop a multi-task deep learning distance predictor (DeepDist) based on new residual convolutional network architectures to simultaneously predict real-value inter-residue distances and classify them into multiple distance intervals. We demonstrate that predicting the real-value distance map and multi-class distance map at the same time performs better than predicting real-value distances alone, indicating their complementarity. On 43 CASP13 hard domains, the average mean square error (MSE) of DeepDist’s real-value distance predictions is 0.896 Å when filtering out the predicted distance >=16 Å, which is lower than 1.003 Å of DeepDist’s multi-class distance predictions. When the predicted real-value distances are converted to binary contact predictions at 8Å threshold, the precisions of top L/5 and L/2 contact predictions are 78.6% and 64.5%, respectively, higher than the best results reported in the CASP13 experiment. These results demonstrate that the real-value distance prediction can predict inter-residue distances well and improve binary contact prediction over the existing state-of-the-art methods. Moreover, the predicted real-value distances can be directly used to reconstruct protein tertiary structures better than multi-class distance predictions due to the lower MSE.