Multi-layer Perceptron Architecture for Tertiary Structure Prediction of Helical Content of Proteins from Peptide Sequences

The purpose of the present study is to deduce the novel method for tertiary structure prediction of various important unpredicted proteins i.e. metabolic, regulatory, signalling etc. due unavailability of template structure. Multi-layer perception architecture has been developed to predict the tertiary structure (Phi/Psi) of helical content of proteins. A novel codification scheme has been devised for data processing (I/O). The proposed system has been tested with different number of neural networks, training set sizes and training epochs. The overall successful prediction of residues for tertiary structure prediction (Phi/Psi) of helical content of protein has been reported according to window size as 15(51.4% / 57.8%), 17(57% / 64%), 19(52.2% / 54.2%), 21(52% / 57.4%). This study demonstrated the possibility of implementing fast and efficient structure prediction using neural network.