An efficient framework for accelerating Needleman-Wunsch algorithm using GPU

The Needleman-Wunsch algorithm is considered the benchmark for global alignment, this work proposes a new implementation for the parallel NW algorithm over the graphical processing unit (GPU). Focusing on enhancing the second phase of the algorithm (The Fill) the most time demanding phase. The idea of filling a percentage of the matrix is presented which guarantees a decrease in execution time, the key was to find the minimum needed percentage to be filled while ensuring the same result as filling the whole matrix of the algorithm. Experiments show the effectiveness of the proposed model in execution time when compared with the sequential algorithm.