Abstract

Oral Sessions

Day 3, May 17（Fri.）　10:10-10:30　Room D (202)

3D-O1-1010　　PDF

High Accuracy Protein Identification Using Posterior Peptide Sequence Tags

(¹Kyoto Univ., ²Kyoto Univ., ³KTH (Royal Inst. of Tech.), Sweden)
^oTsuyoshi Tabata^1,2, Akiyasu Yoshizawa¹, Mio Iwasaki², Naoyuki Sugiyama¹, Lukas Käll³, Yasushi Ishihama¹

We developed an approach to identify peptides based on product ion spectra with peak annotation obtained by multiple search engines, which are implemented in jPOST, the integrated database for proteomes. In order to merge the results from multiple search engines and to remove possible false positives, we have developed a new index “jPOST score,” in which peptide sequence tags by multiple search engines are posteriorly used to increase the accuracy of peptide identification. By re-sorting the peptides according to jPOST score for a large-scale dataset of colon cancers registered in jPOST, the number of identified peptides increased from 63,127 to 111,270 at the FDR=1%. In this presentation, we will describe how to calculate jPOST scores and how to improve the scoring algorithm.