Distributed Phylogenetic Tree Processing on Biology Sequences Using Mapreduce

Abstract

Multiple Sequence Alignment (MSA) is an important process in the analysis of biological sequences by making comparisons of several biological sequences. In some MSA algorithms (such as CLUSTALW, for example), the formation of phylogenetic trees as a guideline in the alignment process has an important role in determining the accuracy of the final alignment results. From the whole MSA process, the formation of phylogenetic tree computational time increased as the number of sequences increases. Computing the similarity score for all sequence pairs takes a lot of time, causing a long computation time problem. In addition, this research uses the neighbor join method as a technique to build a phyloghenetic tree This Research examines the potential efficiency of computational phylogenetic trees in parallel and is distributed to the Hadoop environment using MapReduce. The results showed that a phylogenetic tree can be generated using MapReduce computation and results in 19%-time efficiency from 2 and 3 computation nodes.