Premium
Support vector machines for the classification and prediction of β‐turn types
Author(s) -
Cai YuDong,
Liu XiaoJun,
Xu Xuebiao,
Chou KuoChen
Publication year - 2002
Publication title -
journal of peptide science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.475
H-Index - 66
eISSN - 1099-1387
pISSN - 1075-2617
DOI - 10.1002/psc.401
Subject(s) - tetrapeptide , turn (biochemistry) , overfitting , support vector machine , artificial intelligence , beta (programming language) , sequence (biology) , artificial neural network , type (biology) , mathematics , chemistry , pattern recognition (psychology) , machine learning , computer science , peptide , biochemistry , biology , ecology , programming language
The support vector machines (SVMs) method is proposed because it can reflect the sequence‐coupling effect for a tetrapeptide in not only a β‐turn or non‐β‐turn, but also in different types of β‐turn. The results of the model for 6022 tetrapeptides indicate that the rates of self‐consistency for β‐turn types I, I′, II, II′, VI and VIII and non‐β‐turns are 99.92%, 96.8%, 98.02%, 97.75%, 100%, 97.19% and 100%, respectively. Using these training data, the rate of correct prediction by the SVMs for a given protein: rubredoxin (54 residues, 51 tetrapeptides) which includes 12 β‐turn type I tetrapeptides, 1 β‐turn type II tetrapeptide and 38 non‐β‐turns reached 82.4%. The high quality of prediction of the SVMs implies that the formation of different β‐turn types or non‐β‐turns is considerably correlated with the sequence of a tetrapeptide. The SVMs can save CPU time and avoid the overfitting problem compared with the neural network method. Copyright © 2002 European Peptide Society and John Wiley & Sons, Ltd.