Poster #RP221
EVALUATION OF COMPUTATIONAL TOOL FOR PEPTIDE IDENTIFICATION BASED ON MULTIPLE DATABASE CONSTRAINS
Itaraju Brum*, Eduardo Galembeck*
*Depto. Bioquímica - IB - UNICAMP, Campinas, Brasil
The main goal of this work is to evaluate a computational tool designed to simulate protein digestion by trypsin, and to calculate the statistical features from the theoretical digested peptide set. The feasibility of peptide identification according to the calculated properties such as protein MW, peptide pI, peptide MW and presence of cysteins in aminoacid sequences (ICAT labeling) are showed. The setup parameters can be adjusted leading to the best identification levels according to experimental conditions. We tested the computational tool employing genomes from Escherichia coli, Xylella fastidiosa and Rattus norvegicus leading to 80% average degree of peptide identification. In addition, proteomic data from MS/MS experiments with Trypanosoma cruzi were obtained in literature and submitted to analysis. The simulated condition identified 37 proteins from the database among the 41 proteins identified experimentally. The developed computational tool is customizable to obtain the best results according to the inputed parameters. One can also automatically submit the output to metabolic pathway maps, such as ECPath and GenPath.
Supported by CAPES