|
With the avalanche of gene products in the postgenomic age, the gap between newly found protein sequences and the knowledge of their 3D (three dimensional) structures is becoming increasingly wide. It is highly desired to develop a method by which one can predict the folding rates of proteins based on their amino acid sequence information alone. To address this problem, an ensemble predictor, called FoldRate, was developed by fusing the folding-correlated features that can be either directly obtained or easily derived from the sequences of proteins. It was demonstrated by the jackknife cross-validation on a benchmark dataset constructed recently that FoldRate is at least comparable with or even better than the existing methods that, however, need both the sequence and 3D structure information for predicting the folding rate. As a user-friendly web-server, FoldRate is freely accessible to the public at www.csbio.sjtu.edu.cn/bioinf/FoldRate/, by which one can get the desired result for a query protein sequence in around 30 seconds.
|