BWTsearch 0.1.15

Burrows Wheeler Transform (BWT) search

Import the BWT class into your namespace for convenience:

` from BWTsearch import BWT `

Initialize a BWT of text with instantiating the BWT class:

` text = "AGATA" `

` bwt = BWT(text) `

The instance bwt now will contain all necessary indices for efficiently searching:

• bwt.bwt … contains the BWT string

• bwt.position_list … contains the start positions of the rotations for each character of bwt

• bwt.ranks … contains rank indices for each character of the alphabet in bwt

• bwt.cnums … contains number of character occurances of the alphabet in bwt

Use the search() function to obtain a list of positions of exact matches for pattern in text:

` pattern = "AT" `

` bwt.search(pattern) `

This returns a list with indexes into text where an exact match occurs. Empty list if none. The pattern will be searched case-sensitive and should not contain $.

Use the get_text() function to retrieve the original text from the BWT instance:

` bwt.get_text() `

If a sequence file is available, a genome can be loaded from a FASTA file:

` phix = BWT.from_fast_file("phix.fa") `

` phix.search("AGATA") `