Prediction of helix-helix contacts and interacting helices in polytopic membrane proteins using neural networks

Despite rapidly increasing numbers of available 3D structures, membrane proteins still account for less than 1% of all structures in the Protein Data Bank. Recent high-resolution structures indicate a clearly broader structural diversity of membrane proteins than initially anticipated, motivating the development of reliable structure prediction methods specifically tailored for this class of molecules. One important prediction target capturing all major aspects of a protein's 3D structure is its contact map. Our analysis shows that computational methods trained to predict residue contacts in globular proteins perform poorly when applied to membrane proteins. We have recently published a method to identify interacting alpha-helices in membrane proteins based on the analysis of coevolving residues in predicted transmembrane regions. Here, we present a substantially improved algorithm for the same problem, which uses a newly developed neural network approach to predict helix-helix contacts. In addition to the input features commonly used for contact prediction of soluble proteins, such as windowed residue profiles and residue distance in the sequence, our network also incorporates features that apply to membrane proteins only, such as residue position within the transmembrane segment and its orientation toward the lipophilic environment. The obtained neural network can predict contacts between residues in transmembrane segments with nearly 26% accuracy. It is therefore the first published contact predictor developed specifically for membrane proteins performing with equal accuracy to state-of-the-art contact predictors available for soluble proteins. The predicted helix-helix contacts were employed in a second step to identify interacting helices. For our dataset consisting of 62 membrane proteins of solved structure, we gained an accuracy of 78.1%. Because the reliable prediction of helix interaction patterns is an important step in the classification and prediction of membrane protein folds, our method will be a helpful tool in compiling a structural census of membrane proteins.

Prediction of membrane protein structures with complex topologies using limited constraints

Reliable structure-prediction methods for membrane proteins are important because the experimental determination of high-resolution membrane protein structures remains very difficult, especially for eukaryotic proteins. However, membrane proteins are typically longer than 200 aa and represent a formidable challenge for structure prediction. We have developed a method for predicting the structures of large membrane proteins by constraining helix-helix packing arrangements at particular positions predicted from sequence or identified by experiments. We tested the method on 12 membrane proteins of diverse topologies and functions with lengths ranging between 190 and 300 residues. Enforcing a single constraint during the folding simulations enriched the population of near-native models for 9 proteins. In 4 of the cases in which the constraint was predicted from the sequence, 1 of the 5 lowest energy models was superimposable within 4 A on the native structure. Near-native structures could also be selected for heme-binding and pore-forming domains from simulations in which pairs of conserved histidine-chelating hemes and one experimentally determined salt bridge were constrained, respectively. These results suggest that models within 4 A of the native structure can be achieved for complex membrane proteins if even limited information on residue-residue interactions can be obtained from protein structure databases or experiments.

Predicting helix-helix interactions from residue contacts in membrane proteins

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.

Availability:http://bio-cluster.iis.sinica.edu.tw/TMhit/

Contact:tsung@iis.sinica.edu.tw

Supplementary information:Supplementary data are available at Bioinformatics online.

Detecting coevolution without phylogenetic trees? Tree-ignorant metrics of coevolution perform as well as tree-aware metrics

BACKGROUND

Identifying coevolving positions in protein sequences has myriad applications, ranging from understanding and predicting the structure of single molecules to generating proteome-wide predictions of interactions. Algorithms for detecting coevolving positions can be classified into two categories: tree-aware, which incorporate knowledge of phylogeny, and tree-ignorant, which do not. Tree-ignorant methods are frequently orders of magnitude faster, but are widely held to be insufficiently accurate because of a confounding of shared ancestry with coevolution. We conjectured that by using a null distribution that appropriately controls for the shared-ancestry signal, tree-ignorant methods would exhibit equivalent statistical power to tree-aware methods. Using a novel t-test transformation of coevolution metrics, we systematically compared four tree-aware and five tree-ignorant coevolution algorithms, applying them to myoglobin and myosin. We further considered the influence of sequence recoding using reduced-state amino acid alphabets, a common tactic employed in coevolutionary analyses to improve both statistical and computational performance.

RESULTS

Consistent with our conjecture, the transformed tree-ignorant metrics (particularly Mutual Information) often outperformed the tree-aware metrics. Our examination of the effect of recoding suggested that charge-based alphabets were generally superior for identifying the stabilizing interactions in alpha helices. Performance was not always improved by recoding however, indicating that the choice of alphabet is critical.

CONCLUSION

The results suggest that t-test transformation of tree-ignorant metrics can be sufficient to control for patterns arising from shared ancestry.

Reassessing a sparse energetic network within a single protein domain

Understanding the molecular principles that govern allosteric communication is an important goal in protein science. One way allostery could be transmitted is via sparse energetic networks of residues, and one such evolutionary conserved network was identified in the PDZ domain family of proteins by multiple sequence alignment [Lockless SW, Ranganathan R (1999) Science 286:295-299]. We have reassessed the energetic coupling of these residues by double mutant cycles together with ligand binding and stability experiments and found that coupling is not a special property of the coevolved network of residues in PDZ domains. The observed coupling for ligand binding is better explained by a distance relationship, where residues close in space are more likely to couple than distal residues. Our study demonstrates that statistical coupling from sequence analysis is not necessarily a reporter of energetic coupling and allostery.