Authors : Mohammed Hasan Aldulaimi, Suhaila Zainudin and Azuraliza Abu Bakar
Abstract: The prediction of the structural classes of proteins is an important classification issue in bioinformatics research. Knowledge of these classes will give a clear understanding of the protein folding process. For this reason, research into the prediction of protein classes has become a major topic of concern. This research intends to discuss new development of features based on secondary structures information of proteins and hydropathy profile that categorized proteins into all-α, all-β, α/β and α+β with each category being vital in pinpointing the proteins’ structural classes. The folding patterns, functions and interactions between proteins is reliant upon the accurate prediction of its structural classes. This is especially true if one intend to synthesize new proteins possessing unique functionalities. This is however a complex undertaking, especially for structural classes of low-similarity sequences. There are a few computational methods being developed for this purpose (25-40%). The accuracy of the proposed method is on par with current methods being reported in literature.
Mohammed Hasan Aldulaimi, Suhaila Zainudin and Azuraliza Abu Bakar, 2016. An Improved Method for Predicting Protein Structure Classes by Incorporating Hydropathical and Secondary Information Based on Feature Selection Technique. Journal of Engineering and Applied Sciences, 11: 430-435.