Genoogle: an indexed and parallelized search engine for similar DNA sequences (1507.02987v1)

Abstract: The search for similar genetic sequences is one of the main bioinformatics tasks. The genetic sequences data banks are growing exponentially and the searching techniques that use linear time are not capable to do the search in the required time anymore. Another problem is that the clock speed of the modern processors are not growing as it did before, instead, the processing capacity is growing with the addiction of more processing cores and the techniques which does not use parallel computing does not have benefits from these extra cores. This work aims to use data indexing techniques to reduce the searching process computation cost united with the parallelization of the searching techniques to use the computational capacity of the multi core processors. To verify the viability of using these two techniques simultaneously, a software which uses parallelization techniques with inverted indexes was developed. Experiments were executed to analyze the performance gain when parallelism is utilized, the search time gain, and also the quality of the results when it compared with others searching tools. The results of these experiments were promising, the parallelism gain overcame the expected speedup, the searching time was 20 times faster than the parallelized NCBI BLAST, and the searching results showed a good quality when compared with this tool. The software source code is available at https://github.com/felipealbrecht/Genoogle .