탠덤 질량 스펙트럼에서 추출된 갭 태그를 단백질데이터베이스에서 검색하는 효율적인 알고리즘

초록

Proteomics is the study of proteins, particularly expression, structures, functions and interactions of proteins. Protein identification searches protein sequence database for given amino acid sequences. Recently, MS-GappedDictionary proposed to search the databases using gapped tags. Using this method, more candidate amino-acid sequences are found than previous method. For searching gapped tags, it takes O(n) time without database indexing and O(p) time with indexing where n is the size of database and p is the average number of gapped tags. For UniProt protein sequence database which is widely-used in general, n = 35,856,033 and p = 719, so it is really necessary to index the database to reduce time. In previous works, a gap size is limited to 500Da because it uses a lot of memory to index the database. However, the size of a gap is actually more than 500Da. We propose an efficient algorithm which searches gapped tags whose size more than 500Da in O(p) time. Our method uses the same memory, but search time is 652 and 98 times faster than original method from 501Da to 600Da and to 650Da, respectively.

키워드