Parallel Biological Sequence Comparison on Heterogeneous High Performance Computing Platforms with BSP++

Abstract

Biological Sequence Comparison is an important operation inBioinformatics that is often used to relate organisms. Smith and Waterman proposed an exact algorithm (SW) that compares two sequences in quadratic time and space. Due to high computing and memory requirements, SW is usually executed on HPC platforms such as multicore clusters and CellBEs. Since HPC architectures exhibit very different hardware characteristics, porting an application between them is an error-prone time-consuming task. BSP++ is an implementation of BSP that aims to reduce the effort to write parallel code. In this paper, we propose and evaluate a parallel BSP++ strategy to execute SW in multiple platforms like MPI, OpenMP, MPI/OpenMP, CellBE and MPI/CellBE. The results obtained with real DNA sequences show that the performance of our versions is comparable to the ones in the literature, evidencing the appropriateness and flexibility of our approach.

Implementation and performance analysis of efficient index structures for DNA search algorithms in parallel platforms

“…Finally, it is worth noting that further improvements could still be obtained by parallelizing the processing of these algorithms on a large cluster by using a processing model entirely similar to the one presented in . This approach would be particularly advantageous in data‐level parallelization schemes, when considering the alignment of a large number of queries to a large number of sequences.…”

“…This approach significantly reduces the computational requirements when compared with applying the S-W directly to the original sequences. Even when highly optimized and parallel implementations of the S-W algorithm are considered [6,7], the usage of heuristics algorithms is still preferred when the runtime is a critical factor. On the other hand, whenever high sensitivity levels are required, the S-W algorithm must be used on the full sequences, because of the fact that the heuristic algorithms may miss some important alignments.Many of the currently adopted heuristic algorithms make use of a pre-prepared index of the reference sequence to accelerate the search of the initial match (the seeds).…”