2C-O1-1050 PDF
データベース検索エンジンを用いたタンパク質同定における特異性向上
The developing integrated proteome database “jPOST” plans to employ a methodology to remove unreliable results that can be false positives from the database search results. False positive assignments can be estimated by decoy sequence assignments, and on the sorted result, higher-ranked assignments than the highest-ranked decoy assignment can be assumed to contain no false positive assignments. In order to move down the highest decoy assignment, we have developed a new index “jPOST score,” which employs annotated mass peak information, such as the concept of peptide sequence tags, and is equivalent to manual inspection of spectra. By re-sorting the database search result with this jPOST score, the rank of highest-ranked decoy assignment was lower by more than 10%, and similarly the number of obtained sequences was increased more than 10% under the condition FDR=0.1%.