PAT-PseAAC: Prediction of acetyl threonine in protein sites with statistical moments and chou’s 5 step rule

Abstract

N-acetyl threonine is a post translational modification that involved in the adaption of cells and functional properties of a large number of proteins. N-acetyl threonine has a key importance in sub cellular concentration, transcriptional synchronization, chromatin remodeling, response to stress and regulation of mitosis. Sumo is associated to development defects and a number of diseases in humans such as cancer, Huntington's, Alzheimer's, Parkinson's diseases, Spinocerebellar ataxia 1 and amyotrophic lateral sclerosis. The involvement of M-acetyl threonine in different disease, it is important to find a way to predict the presence of N-acetyl threonine in protein sequences. We proposed a model to predict the N-acetyl threonine sites by using the Chou's Pseudo Amino Acid Composition (PseAAc), which based on relative positive features. For the evaluation of our predictor namely PATPseAAC, self-consistency testing as well as 10-fold cross validation are implemented with the help of accuracy metrics. The result of self-consistency testing was 100% Acc, 100% Sn, 100% Sp and 100% MCC, while the result of 10-fold cross validation was 94.23% Acc, 94.83% Sn, 95.06% Sp and 0.93% MCC noticed. Whereas Jackknife test, 96.01% Acc, 98.75% Sn, 98.83% Sp and 0.93% MCC.