Distributed realization of parallel algorithm for global alignment of long sequences

Abstract

Distributed realization of parallel global alignment algorithm is proposed. The goal of this work is to collect the best practices and to build the most general model of a pairwise global alignment with the possibility of aligning as long sequences as possible. As a result, the versatile algorithm for global alignment is constructed on the basis of parallel implementation FastLSA of the Needleman–Wunsch algorithm with an arbitrary matrix of substitutions and Gotoh algorithm for the affine system of penalties for gaps. The main feature of the algorithm is low memory consumption. The algorithm is implemented in the Javascript programming language as web application. Distributed computing is implemented in the PHP programming language and is produced on clients that use this web service.