Abstract
BackgroundHidden Markov Models (HMMs) have been extensively used in computational molecular biology, for modelling protein and nucleic acid sequences. In many applications, such as transmembrane protein topology prediction, the incorporation of limited amount of information regarding the topology, arising from biochemical experiments, has been proved a very useful strategy that increased remarkably the performance of even the top-scoring methods. However, no clear and formal explanation of the algorithms that retains the probabilistic interpretation of the models has been presented so far in the literature.ResultsWe present here, a simple method that allows incorporation of prior topological information concerning the sequences at hand, while at the same time the HMMs retain their full probabilistic interpretation in terms of conditional probabilities. We present modifications to the standard Forward and Backward algorithms of HMMs and we also show explicitly, how reliable predictions may arise by these modifications, using all the algorithms currently available for decoding HMMs. A similar procedure may be used in the training procedure, aiming at optimizing the labels of the HMM's classes, especially in cases such as transmembrane proteins where the labels of the membrane-spanning segments are inherently misplaced. We present an application of this approach developing a method to predict the transmembrane regions of alpha-helical membrane proteins, trained on crystallographically solved data. We show that this method compares well against already established algorithms presented in the literature, and it is extremely useful in practical applications.ConclusionThe algorithms presented here, are easily implemented in any kind of a Hidden Markov Model, whereas the prediction method (HMM-TM) is freely available for academic users at , offering the most advanced decoding options currently available.
Highlights
Hidden Markov Models (HMMs) have been extensively used in computational molecular biology, for modelling protein and nucleic acid sequences
With the development of fast and reliable methods to utilise gene fusion technology in order to determine the location of a protein's C-terminus, it has been shown that incorporating topological information into the prediction methods improves largely the performance of even the top-scoring methods, making easy to screen newly sequenced genomes [24,25]
The particular method is currently not available to the public. This method is based in an Ensemble network combining the results of two independent HMM modules with these of a Neural Network predictor, using a dynamic programming algorithm that filters the prediction in a last step
Summary
Hidden Markov Models (HMMs) have been extensively used in computational molecular biology, for modelling protein and nucleic acid sequences. BMC Bioinformatics 2006, 7:189 http://www.biomedcentral.com/1471-2105/7/189 been found to perform significantly better compared to other sophisticated Machine-Learning techniques such as Neural Networks (NNs) or Support Vector Machines (SVMs) This is the case irrespective to which class of transmembrane proteins we refer to, since it has been shown that the best currently available predictors for alpha-helical membrane proteins [12,13] as well as for beta-barrel outer membrane proteins [14], are methods based on HMMs. Experimental techniques are routinely used to partially uncover the transmembrane topology of membrane proteins (in contrast to the more expensive and difficult method of crystallography with which the detailed threedimensional structure is elucidated). A global topological analysis of the E. coli inner membrane proteins was performed, providing reliable models for more than 600 membrane proteins [28]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
