LSCGAT: LONG SEQUENCES CUSTOMIZABLE GLOBAL ALIGNMENT TOOL

The most general model of a pairwise global alignment with the possibility of aligning long sequences is considered. The model features are alignment of sequences in different alphabets including nucleotides and amino acids, alternative alignments with the same score, predefined or fully customazible scoring matrix and gap penalty systems for each sequence including end gap penalties. Developed versatile parallel algorithm for global alignment is based on the Needleman-Wunsch algorithm with an arbitrary scoring matrix and Gotoh algorithm for the affine system of penalties for gaps. The main features of the algorithm include optimal memory consumption and parallel computation. It is proved that algorithm can align two sequences of length in memory space . The algorithm is implemented in the form of a high-performance web application in Javascript programming language with the usage of webworkers and is freely available on http://sbars.impb.ru/aligner.html.