Proteins are made up of basic units called amino acids which are held together by bonds namely hydrogen and ionic bond. The way in which the amino acids are sequenced has been categorized into two dimensional and three dimensional structures. The main advantage of predicting secondary structure is to produce tertiary structure predictions which are in great demand to the continuous discovery of proteins. This paper reviews the different methods adopted for predicting the protein secondary structure and provides a comparative analysis of accuracies obtained from various input datasets.
Keywords : Protein Secondary Structure, Auto Encoder, Bayes Classifier, Margin Infused Relaxed Algorithm(MIRA), Deep Neural Residual Network (Deepnrn), PSI-BLAST, Cullpdb, Support Vector Machines, Position Specific Scoring Matrix(PSSM).