Locating Apoptosis Proteins By Predicted Signal Peptide Cleavage Sites Of Protein Sequences

Apoptosis proteins play an essential role in the development andhomeostasis of an organism the accurate prediction of subcellular location forapoptosis proteins is helpful for understanding the mechanism of programmed celldeath and their biological functions. Recently, researchers have been focusing on theproteins localization researching with a lot of outstanding achievement abtained.Things for apoptosis proteins localization are quite different.In this paper, a new apoptosis proteins localization algorithm is proposedbased on the predicted cleavage sites of primary protein sequences. Protein sortingsignals were often overlooked as researchers extracted Protein sequence featureswhen at a study previously，of which Signal peptide was a significant one. Signalpeptide are present on the nascent proteins, it is the signal peptide that determinethe whereabouts of the protein molecules on the cell. When predicting proteinslocalization, it is helpful and rational to make use of the signal peptide. Our mainwork are as follows：First, protein chains are divided into N-terminal signal parts and matureprotein parts according to their predicted cleavage sites by SignalP. Then, amino acidcomposition (ACC) of the individual subsequence together with pseudo-amino acidcomposition (Pse-AAC) and stereochemical properties (SP) of whole chain wereextracted to represent a given protein sequence. Jackknife test by Support VectorMachine on three broadly-used datasets (ZD98, ZW225and CL317datasets) ofapoptosis proteins demonstrated that the total accuracies by this approach are93.9%,87.6%and91.5%, respectively. In addition, an independent non-apoptosisbenchmark dataset (NNPSL) was also used to evaluate the performance of thismethod, and predictive accuracies for eukaryotic and prokaryotic proteins are alsocomparable to existing methods.