Aims to improve the accuracy of transmembrane helices (TMH) prediction. MemBrain is a prediction method to derive transmembrane inter-helix contacts from amino acid sequences by combining correlated mutations and multiple machine learning classifiers. This method is able to predict residue contacts for the entire transmembrane (TM) protein sequence. It was trained with inter-helix residue pairs and long-range residue pairs of type II and III.
Predicts the exposure or burial of transmembrane residues incorporating the structural specificities of channels. RHYTHM is a web application in which the position-specific matrices were updated using an enlarged data set of input structures. It also integrates the secondary structure prediction tool HMMTOP. After the upload of a single sequence file and the specification of the position specific matrix type, the prediction for tertiary structure contacts is started.
Assists users in detecting and evaluating of potential helix-turn-helix DNA-binding motifs in protein sequences. Helix-Turn-Helix prediction is a web server that uses a reference set of 91 presumed helix-turn-helix sequences. The scoring matrix obtained by this reference set has been calibrated against a large protein sequence database. Therefore, the score obtained by a sequence can be used to give a practical estimation of the probability that the sequence is a helix-turn-helix motif.
Predicts helix-helix contacts specifically within the transmembrane parts of membrane proteins. TMHcon is a neural-network based approach. It integrates sequence profiles, correlated mutations, protein topology, sequence separation, and predicted scores for lipid-exposure. It also includes predictions by two neural networks, one developed for helix-helix contacts on all transmembrane helix pairs, and the second specifically tailored for the prediction of helix-helix contacts lying on non-neighboring transmembrane helices.
A contact prediction method for α-helical transmembrane proteins, in which evolutionary couplings are combined with a machine learning approach. MemConP achieves a substantially improved accuracy (precision: 56.0%, recall: 17.5%, MCC: 0.288) compared to the use of either machine learning or co-evolution methods alone. The method also achieves 91.4% precision, 42.1% recall and a MCC of 0.490 in predicting helix-helix interactions based on predicted contacts. The approach was trained and rigorously benchmarked by cross-validation and independent testing on up-to-date non-redundant datasets of 90 and 30 experimental three dimensional structures, respectively.
You can access more results by creating a free plan account or unlimited content via a premium account.