Accurate contact predictions for thousands of protein families using PconsC3

Abstract

Protein structure prediction was for decades one of the grand unsolved challenges in bioinformatics. A few years ago it was shown that by using a maximum entropy approach to describe couplings between columns in a multiple sequence alignment it was possible to significantly increase the accuracy of residue contact predictions. For very large protein families with more than 1000 effective sequences the accuracy is sufficient to produce accurate models of proteins as well as complexes. Today, for about half of all Pfam domain families no structure is known, but unfortunately most of these families have at most a few hundred members, i.e. are too small for existing contact prediction methods. To extend accurate contact predictions to the thousands of smaller protein families we present PconsC3, an improved method for protein contact predictions that can be used for families with as little as 100 effective sequence members. We estimate that PconsC3 provides accurate contact predictions for up to 4646 Pfam domain families. In addition, PconsC3 outperforms previous methods significantly independent on family size, secondary structure content, contact range, or the number of selected contacts. This improvement translates into improved de-novo prediction of three-dimensional structures. PconsC3 is available as a web server and downloadable version at http://c3.pcons.net. The downloadable version is free for all to use and licensed under the GNU General Public License, version 2.protein structure prediction | contact prediction | direct coupling analysis | protein family I n recent years great progress has been made in the area of residue contact prediction. The vast amount of available sequence data is utilized by direct coupling analysis (DCA) methods to predict contacts between residues with unprecedented quality [1,2]. This has enabled accurate blind predictions of the structure of soluble proteins [3][4][5], membrane proteins [6][7][8], and protein complexes [9,10]. However, the widespread use of such methods has been limited to protein families with more than 1000 members [11,12]. Unfortunately, the structure of at least one member of most large families is known (see Fig. S1). This limits the practical usefulness of DCA methods [13] and strongly suggests that methods that accurately predict residue contacts for smaller protein families would be of much greater utility.Before the advent of DCA methods there has been a longstanding effort in using machine learning techniques to predict residue contacts [14][15][16]. These methods utilize covariancebased evolutionary information (e.g. mutual information), as well as knowledge based constraints as inputs to a machine learning algoritm. The best non-DCA methods are less dependent on the size of the protein family and although their predictive quality is easily outperformed by DCA on large families, they perform significantly better on smaller families, Figure 1 (a). We have earlier used an iterative machine learning approach, built on the observation that contacts are ...

Prediction of Structures and Interactions from Genome Information

“…PconsC [94],C2 [96] MetaPSICOV PSICOV,mfDCA, A two stage neural network predictor; CONSIP2 pipeline [51,46,52] GREMLIN/CCMpred http://bioinf.cs.ucl.ac.uk/MetaPSICOV RaptorX [105] CCMpred Ultra-deep learning model consisting of 1-and 2-dimensional convolutional residual neural networks http://raptorx.uchicago.edu/ContactMap/ iFold [13] Deep neural network (DNN) EPSILON-CP PSICOV, GREMLIN, 4 hidden layer neural network mfDCA,CCMpred,GaussDCA with 400-200-200-50 neurons [91] All the DCA methods such as mfDCA, plmDCA, GREMLIN, and PSICOV predict significantly nonoverlapping sets of contacts [46,52,108]. Then, increasing prediction accuracy by combining their predictions together with other sequence/structure information have been attempted [94,96,95,51,46,52,105,91]; see Table 3.…”

“…3 Machine learning methods to augment the contact prediction accuracy based on amino acid coevolution All the DCA methods such as mfDCA, plmDCA, GREMLIN, and PSICOV predict significantly nonoverlapping sets of contacts [46,52,108]. Then, increasing prediction accuracy by combining their predictions together with other sequence/structure information have been attempted [94,96,95,51,46,52,105,91]; see Table 3.…”

“…PconsC [94] combines the predictions of PSICOV and plmDCA into a machine learning method, random forests, and employs alignments with HHblits [84] and jackHMMer [42] at four different e-value cut-offs. Five-layer neural network is employed instead of random forests in PconsC2 [96], and plmDCA and GaussDCA are employed in PconsC3 [95]. A receptive field consisting of 11 × 11 predicted contacts around each residue pair is taken into account in each layer except the first one.…”

“…In the direct-coupling-based methods, the accuracy of predicted contacts depends on the depth [66,48,108] and quality of multiple sequence alignment (MSA) for a target. 5× L (protein length) aligned sequences may be desirable for accurate contact predictions [48], although attempts to improve prediction methods for fewer aligned sequences have been made [94,96,95,105]. PconsC3 can be used for families with as little as 100 effective sequence members [95].…”

“…5× L (protein length) aligned sequences may be desirable for accurate contact predictions [48], although attempts to improve prediction methods for fewer aligned sequences have been made [94,96,95,105]. PconsC3 can be used for families with as little as 100 effective sequence members [95]. Also, RaptorX [105] attained top-L/2-accuracy> 0.3 for long-rang contacts even by using MSAs with 20 effective sequence members (M eff ∼ 20); see Eq.…”