PPMuSA: PROSITE-Pattern Matcher Using Suffix Array

PPMuSA is a pattern matching program for PROSITE [2] signature patterns against amino acid sequence database. To characterize a group of protein among others, it is important work to find occurrences of a signature in an amino acid sequence of the protein. Such signatures are compiled in databases. The PROSITE is one of the databases of signatures that distinguish members of a protein families or domains from other unrelated proteins. In the PROSITE, the signature is described in several forms. One of the forms is a “pattern” that is a motif description based on a regular expressionlike syntax. For instance, the signature pattern is described in a form such as [AC]-x-V-x(4)-{ED}; or <A-x-[ST](2)-x(0,1)-v. These patterns would be interpreted as [Ala or Cys]-any-Val-any-any-any-any{any but Glu or Asp}; and Nterm-Ala-any-[Ser or Thr]-[Ser or Thr]-(any or none)-Val, respectively. ScanProsite [1] is a reference implementation of a PROSITE scanning tool that can take a pattern of the form as the input and report occurrences of the pattern in amino acid sequences. ScanProsite uses a sequential algorithm as opposed to the indexed one. PPMuSA uses prebuilt suffix array [3] indices on the amino acid sequences in order to achieve a fast pattern matching. The suffix array is a data structure for the information retrieval, and is made of an array of the indices of the suffixes obtained by sorting all suffixes of the text. In the case of string retrieval using suffix array, the time complexity is O(m · log n), (O(m+log n) when auxiliary data structure is used), where m is the length of a pattern string and n is the length of the text. PPMuSA takes a pattern of the form as the input, and reports all the occurrences of the pattern in amino acid sequence database.