Predicting the Real-Valued Inter-Residue Distances for Proteins.
Ontology highlight
ABSTRACT:
SUBMITTER: Ding W
PROVIDER: S-EPMC7539185 | biostudies-literature | 2020 Oct
REPOSITORIES: biostudies-literature
Publications
Predicting the Real-Valued Inter-Residue Distances for Proteins.
Advanced science (Weinheim, Baden-Wurttemberg, Germany) 20200810 19
Predicting protein structure from the amino acid sequence has been a challenge with theoretical and practical significance in biophysics. Despite the recent progresses elicited by improved inter-residue contact prediction, contact-based structure prediction has gradually reached the performance ceiling. New methods have been proposed to predict the inter-residue distance, but unanimously by simplifying the real-valued distance prediction into a multiclass classification problem. Here, a lightwei ...[more]
Similar Datasets
Project description:BackgroundProtein inter-residue contact maps provide a translation and rotation invariant topological representation of a protein. They can be used as an intermediary step in protein structure predictions. However, the prediction of contact maps represents an unbalanced problem as far fewer examples of contacts than non-contacts exist in a protein structure.In this study we explore the possibility of completely eliminating the unbalanced nature of the contact map prediction problem by predicting real-value distances between residues. Predicting full inter-residue distance maps and applying them in protein structure predictions has been relatively unexplored in the past.ResultsWe initially demonstrate that the use of native-like distance maps is able to reproduce 3D structures almost identical to the targets, giving an average RMSD of 0.5Å. In addition, the corrupted physical maps with an introduced random error of ±6Å are able to reconstruct the targets within an average RMSD of 2Å.After demonstrating the reconstruction potential of distance maps, we develop two classes of predictors using two-dimensional recursive neural networks: an ab initio predictor that relies only on the protein sequence and evolutionary information, and a template-based predictor in which additional structural homology information is provided. We find that the ab initio predictor is able to reproduce distances with an RMSD of 6Å, regardless of the evolutionary content provided. Furthermore, we show that the template-based predictor exploits both sequence and structure information even in cases of dubious homology and outperforms the best template hit with a clear margin of up to 3.7Å.Lastly, we demonstrate the ability of the two predictors to reconstruct the CASP9 targets shorter than 200 residues producing the results similar to the state of the machine learning art approach implemented in the Distill server.ConclusionsThe methodology presented here, if complemented by more complex reconstruction protocols, can represent a possible path to improve machine learning algorithms for 3D protein structure prediction. Moreover, it can be used as an intermediary step in protein structure predictions either on its own or complemented by NMR restraints.
| S-EPMC3893389 | biostudies-literature
Project description:As deep learning algorithms drive the progress in protein structure prediction, a lot remains to be studied at this merging superhighway of deep learning and protein structure prediction. Recent findings show that inter-residue distance prediction, a more granular version of the well-known contact prediction problem, is a key to predicting accurate models. However, deep learning methods that predict these distances are still in the early stages of their development. To advance these methods and develop other novel methods, a need exists for a small and representative dataset packaged for faster development and testing. In this work, we introduce protein distance net (PDNET), a framework that consists of one such representative dataset along with the scripts for training and testing deep learning methods. The framework also includes all the scripts that were used to curate the dataset, and generate the input features and distance maps. Deep learning models can also be trained and tested in a web browser using free platforms such as Google Colab. We discuss how PDNET can be used to predict contacts, distance intervals, and real-valued distances.
| S-EPMC7414848 | biostudies-literature
Project description:BackgroundThe wealth of information on protein structure has led to a variety of statistical analyses of the role played by individual amino acid types in the protein fold. In particular, the contact propensities between the various amino acids can be converted into folding energies that have proved useful in structure prediction. The present study addresses the relationship of protein folding propensities to the evolutionary relationship between residues.ResultsThe contact preferences of residue types observed in a representative sample of protein structures are converted into a residue similarity matrix or inter-residue distance matrix. Remarkably, these distances correlate excellently with evolutionary substitution costs. Residue vectors are derived from the distance matrix. The residue vectors give a concrete picture of the grouping of residues into families sharing properties crucial for protein folding.ConclusionsInter-residue distances have proved useful in showing the explicit relationship between contact preferences and evolutionary substitution rates. It is proposed that the distance matrix derived from structural analysis may be useful in aligning proteins where remote homologs share structural features. Residue vectors derived from the distance matrix illustrate the spatial arrangement of residues and point to ways in which they can be grouped.
| S-EPMC526251 | biostudies-literature
Project description:In this paper, based on a recent work by McAllister and Floudas who developed a mathematical optimization model to predict the contacts in transmembrane alpha-helical proteins from a limited protein data set [1], we have enhanced this method by 1) building a more comprehensive data set for transmembrane alpha-helical proteins and this enhanced data set is then used to construct the probability sets, MIN-1N and MIN-2N, for residue contact prediction, 2) enhancing the mathematical model via modifications of several important physical constraints and 3) applying a new blind contact prediction scheme on different protein sets proposed from analyzing the contact prediction on 65 proteins from Fuchs et al. [2]. The blind contact prediction scheme has been tested on two different membrane protein sets. Firstly it is applied to five carefully selected proteins from the training set. The contact prediction of these five proteins uses probability sets built by excluding the target protein from the training set, and an average accuracy of 56% was obtained. Secondly, it is applied to six independent membrane proteins with complicated topologies, and the prediction accuracies are 73% for 2ZY9A, 21% for 3KCUA, 46% for 2W1PA, 64% for 3CN5A, 77% for 3IXZA and 83% for 3K3FA. The average prediction accuracy for the six proteins is 60.7%. The proposed approach is also compared with a support vector machine method (TMhit [3]) and it is shown that it exhibits better prediction accuracy.
| S-EPMC3164537 | biostudies-literature
Project description:BackgroundInversion Symmetry is a generalization of the second Chargaff rule, stating that the count of a string of k nucleotides on a single chromosomal strand equals the count of its inverse (reverse-complement) k-mer. It holds for many species, both eukaryotes and prokaryotes, for ranges of k which may vary from 7 to 10 as chromosomal lengths vary from 2Mbp to 200 Mbp. Building on this formalism we introduce the concept of k-mer distances between chromosomes. We formulate two k-mer distance measures, D1 and D2, which depend on k. D1 takes into account all k-mers (for a single k) appearing on single strands of the two compared chromosomes, whereas D2 takes into account both strands of each chromosome. Both measures reflect dissimilarities in global chromosomal structures.ResultsAfter defining the various distance measures and summarizing their properties, we also define proximities that rely on the existence of synteny blocks between chromosomes of different bacterial strains. Comparing pairs of strains of bacteria, we find negative correlations between synteny proximities and k-mer distances, thus establishing the meaning of the latter as measures of evolutionary distances among bacterial strains. The synteny measures we use are appropriate for closely related bacterial strains, where considerable sections of chromosomes demonstrate high direct or reversed equality. These measures are not appropriate for comparing different bacteria or eukaryotes. K-mer structural distances can be defined for all species. Because of the arbitrariness of strand choices, we employ only the D2 measure when comparing chromosomes of different species. The results for comparisons of various eukaryotes display interesting behavior which is partially consistent with conventional understanding of evolutionary genomics. In particular, we define ratios of minimal k-mer distances (KDR) between unmasked and masked chromosomes of two species, which correlate with both short and long evolutionary scales.Conclusionsk-mer distances reflect dissimilarities among global chromosomal structures. They carry information which aggregates all mutations. As such they can complement traditional evolution studies , which mainly concentrate on coding regions.
| S-EPMC8422766 | biostudies-literature
Project description:BACKGROUND:Accurate prediction of inter-residue contacts of a protein is important to calculating its tertiary structure. Analysis of co-evolutionary events among residues has been proved effective in inferring inter-residue contacts. The Markov random field (MRF) technique, although being widely used for contact prediction, suffers from the following dilemma: the actual likelihood function of MRF is accurate but time-consuming to calculate; in contrast, approximations to the actual likelihood, say pseudo-likelihood, are efficient to calculate but inaccurate. Thus, how to achieve both accuracy and efficiency simultaneously remains a challenge. RESULTS:In this study, we present such an approach (called clmDCA) for contact prediction. Unlike plmDCA using pseudo-likelihood, i.e., the product of conditional probability of individual residues, our approach uses composite-likelihood, i.e., the product of conditional probability of all residue pairs. Composite likelihood has been theoretically proved as a better approximation to the actual likelihood function than pseudo-likelihood. Meanwhile, composite likelihood is still efficient to maximize, thus ensuring the efficiency of clmDCA. We present comprehensive experiments on popular benchmark datasets, including PSICOV dataset and CASP-11 dataset, to show that: i) clmDCA alone outperforms the existing MRF-based approaches in prediction accuracy. ii) When equipped with deep learning technique for refinement, the prediction accuracy of clmDCA was further significantly improved, suggesting the suitability of clmDCA for subsequent refinement procedure. We further present a successful application of the predicted contacts to accurately build tertiary structures for proteins in the PSICOV dataset. CONCLUSIONS:Composite likelihood maximization algorithm can efficiently estimate the parameters of Markov Random Fields and can improve the prediction accuracy of protein inter-residue contacts.
| S-EPMC6821021 | biostudies-literature
Project description:This article contains inter-residue solvent accessible surface distances between lysines in a comprehensive dataset homo-oligomeric protein complex structures, downloaded from 3D Complex database. Solvent Accessible Surface Distances were calculated with Jwalk algorithm. To avoid unnecessary redundancy due to symmetry, we calculated only distances originating from the first subunit of each protein complex. Redundancy was further reduced by including only the shortest of the two possible inter-subunit alternatives in the final non-redundant dataset. For each protein complexes we also calculated weight ob subunits, number of lysines radius of gyration and average distances. This data can be used for structural analyses of homo-oligomeric protein complexes and for further optimization of distance-based restraints, such as those based on data obtained from chemical cross-linking coupled with mass spectrometry.
| S-EPMC6909155 | biostudies-literature
Project description:BackgroundDriven 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.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. 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.ConclusionsDeepDist 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.
| S-EPMC7831258 | biostudies-literature
Project description:MotivationDNA sequences can be represented by sequences of four symbols, but it is often useful to convert the symbols into real or complex numbers for further analysis. Several mapping schemes have been used in the past, but they seem unrelated to any intrinsic characteristic of DNA. The objective of this work was to find a mapping scheme directly related to DNA characteristics and that would be useful in discriminating between different species. Mathematical models to explore DNA correlation structures may contribute to a better knowledge of the DNA and to find a concise DNA description.ResultsWe developed a methodology to process DNA sequences based on inter-nucleotide distances. Our main contribution is a method to obtain genomic signatures for complete genomes, based on the inter-nucleotide distances, that are able to discriminate between different species. Using these signatures and hierarchical clustering, it is possible to build phylogenetic trees. Phylogenetic trees lead to genome differentiation and allow the inference of phylogenetic relations. The phylogenetic trees generated in this work display related species close to each other, suggesting that the inter-nucleotide distances are able to capture essential information about the genomes. To create the genomic signature, we construct a vector which describes the inter-nucleotide distance distribution of a complete genome and compare it with the reference distance distribution, which is the distribution of a sequence where the nucleotides are placed randomly and independently. It is the residual or relative error between the data and the reference distribution that is used to compare the DNA sequences of different organisms.
| S-EPMC2778338 | biostudies-literature
Project description:MOTIVATION: Helix-helix interactions play a critical role in the structure assembly, stability and function of membrane proteins. On the molecular level, the interactions are mediated by one or more residue contacts. Although previous studies focused on helix-packing patterns and sequence motifs, few of them developed methods specifically for contact prediction. RESULTS: We present a new hierarchical framework for contact prediction, with an application in membrane proteins. The hierarchical scheme consists of two levels: in the first level, contact residues are predicted from the sequence and their pairing relationships are further predicted in the second level. Statistical analyses on contact propensities are combined with other sequence and structural information for training the support vector machine classifiers. Evaluated on 52 protein chains using leave-one-out cross validation (LOOCV) and an independent test set of 14 protein chains, the two-level approach consistently improves the conventional direct approach in prediction accuracy, with 80% reduction of input for prediction. Furthermore, the predicted contacts are then used to infer interactions between pairs of helices. When at least three predicted contacts are required for an inferred interaction, the accuracy, sensitivity and specificity are 56%, 40% and 89%, respectively. Our results demonstrate that a hierarchical framework can be applied to eliminate false positives (FP) while reducing computational complexity in predicting contacts. Together with the estimated contact propensities, this method can be used to gain insights into helix-packing in membrane proteins.
| S-EPMC2666818 | biostudies-other