Frin: An Efficient Method for Representing Genome Evolutionary History.

Abstract

Phylogenetic analysis is important in understanding the process of biological evolution, and phylogenetic trees are used to represent the evolutionary history. Each taxon in a phylogenetic tree has not more than one parent, so phylogenetic trees cannot express the complex evolutionary information implicit in phylogeny. Phylogenetic networks can be used to express genome evolutionary histories. Therefore, it is great significance to research the construction of phylogenetic networks. Cass algorithm is an efficient method for constructing phylogenetic networks because it can construct a much simpler network. However, Cass relies heavily on the order of input data, i.e. different networks can be constructed for the same dataset with different input orders. Based on the frequency and incompatibility degree of taxa, we propose an efficiently improved algorithm of Cass, called as Frin. The experimental results show that the networks constructed by Frin are not only simpler than those constructed by other methods, but Frin can also construct more consistent phylogenetic networks when the treated data have different input orders. Furthermore, the phylogenetic network constructed by Frin is closer to the original information described by phylogenetic trees. Frin has been built as a Java software package and is freely available at https://github.com/wangjuanimu/Frin.