11-11-2014, 09:53 AM
Abstracts: Pattern matching is a fundamental and upcoming area in computational molecular biology. The problem in pattern discovery is to determine how often a candidate pattern occurs, as well as possibly some information on its frequency distribution across the sequence/text. With the increasing need for instant information, pattern matching will continue to grow and change as needed from time to time. So new techniques are required that consumes less time for sequence alignment by avoiding unnecessary comparison in sequences. There are many sequence alignment algorithms. We have taken Smith -Waterman algorithm for sequence alignment. Like many applications in computational science, the Smith Waterman(s-w) algorithm is constrained by the memory access speed. If we run this algorithm of computational biology in sequential manner , then they are slow , less accurate. Thus, such a cheap, desirable device is required. GPU has provided powerful platform to accelerate algorithms. So we have implemented S-W algorithm by using OpenCL library to exploit GPU support of AMD Radeon graphics card.