Sequence-Based Prediction of Protein Folding Rates Using Contacts, Secondary Structures and Support Vector Machines

Guan Ning Lin, Zheng Wang, Dong Xu, Jianlin Cheng
2009 2009 IEEE International Conference on Bioinformatics and Biomedicine  
Predicting protein folding rate is useful for understanding protein folding process and guiding protein design. Here we developed a method, SeqRate, to predict both protein folding kinetic type (two-state versus multi-state) and real-value folding rate using features extracted from only protein sequence with support vector machines. On a standard benchmark dataset, the accuracy of folding kinetic type classification is 80%. The Pearson correlation coefficient and the mean absolute difference
more » ... ween predicted and experimental folding rates (sec -1 ) in the base-10 logarithmic scale are 0.81 and 0.79 for twostate protein folders, and 0.80 and 0.68 for three-state protein folders. SeqRate is the first sequence-based method for protein folding type classification and its accuracy of fold rate prediction is improved over previous sequence-based methods. Both the web server and software of predicting folding rate are publicly available at
doi:10.1109/bibm.2009.21 dblp:conf/bibm/LinWXC09 fatcat:cmw2trfpbzattgbee45hfbrlly