Feature Extraction From Wavelet Coefficients for Pattern Recognition Tasks
IEEE Transactions on Pattern Analysis and Machine Intelligence
A comparison of methods for multiclass support vector machines
IEEE Transactions on Neural Networks
| Hi-index | 0.00 |
In the protein universe, many proteins are composed of two or more polypeptide chains, generally referred to as subunits, which associate through noncovalent interactions and, occasionally, disulfide bonds to form protein quaternary structures. It has been known for long that the functions of proteins are closely related to their quaternary structure. With the number of protein sequences entering into data banks rapidly increasing, it is highly desirable to predict protein quaternary structures automatically from their primary sequences. Here, multi-scale energy of factor solution scores and feature combination were employed to form various input feature vectors, and the multi-class support vector machine (SVM) classifier modules were adopted for predicting protein quaternary structures. The rates of correct identification suggest that the individual primary sequence of an oligomeric protein do contain the information of its quaternary structure. The results of multi-scale energy of Factor 1 solution scores indirectly prove that biologically relevant complex formation is driven predominantly by the hydrophobic effect. The current approach is quite promising and may at least play a complimentary role to the existing methods.