DeepCys is a deep neural network-based multiple Cys function prediction based on the strucutre . The algorithm only predicts four Cys functions namely Disulphide , Metal-Binding , Sulfenylation and Thioether based on the strucutre . This neural network model was trained and tested on two independent datasets curated from the protein crystal strucutres . The prediction method requires three inputs namely PDB identifier (ID) , Chain ID and Residue ID for a given Cys and outputs the probabilities of four Cys functions .

In this study we present a deep learning-based approach to predict any one of the four most abundant Cys modifications. Novelty of this work was prediction of maximum number of Cys modifications . Moreover , Thioether prediction was not attempted earlier . The DeepCys model developed in this work requires the protein strucutre in PDB format , the residue number and the chain identifier. Six features were either extracted or computed from PDB file those were used by the deep learning approach. The final output was the probability values for four cysteine modifications, namely, Disulphide , Metal-Binding , Thioether and Sulphenylation. The modification with highest probability was reported as the predicted modification.