Prediction of membrane proteins based on classification of transmembrane segments

The number of transmembrane segments often corresponds to a structural or f unctional class of membrane proteins such as to seven-transmembrane recepto rs and six-transmembrane ion channels. We have developed a new prediction m ethod to detect the membrane protein class that is defined by the number of transmembrane segments, as well as to locate the transmembrane segments in the amino acid sequence. Each membrane protein class is represented by a m odel of ordering different types of transmembrane segments. Specifically, w e have classified the transmembrane segments in known membrane proteins int o five groups (types) using the Mahalanobis distance with the average hydro phobicity and the periodicity of hydrophobicity as a measure of similarity. The discriminant functions derived for these groups were then used to dete ct transmembrane segments and to match with the models for one- to fourteen -spanning membrane proteins and for globular proteins. Using the test data set of 89 membrane proteins whose transmembrane positions are known by expe rimental evidence, 61.8% of the proteins and 85.1% of the transmembrane seg ments were correctly predicted. Because of the new feature to predict membr ane protein classes, the method should be useful in the functional assignme nt of genomic sequences.