Maximum Parsimony (MP) Method

Maximum parsimonyRH_Maximum_Parsimony_Principle (MP) methods originally were developed for morphological characters, and there are many different versions (see Nei and Kumar [2000]Nei_and_Kumar_2000 for a review).  In MEGA, we consider both of these methods for nucleotide and amino acid sequence data (Eck and Dayhoff 1966Eck_and_Dayhoff_1966; Fitch 1971Fitch_1971).

For constructing an MP tree, only sites at which there are at least two different kinds of nucleotides or amino acids, each represented at least twice, are used (parsimony-informative sites). Other variable sites are not used for constructing an MP tree, although they are informative for distance and maximum-likelihood methods.

MEGA estimates MP tree branch lengths by using the average pathway method for unrooted trees (see Nei and Kumar [2000]Nei_and_Kumar_2000, page 132).

To search for MP Trees, MEGA provides four different types of searches: the max-mini branch-and-bound searchRH_Max_mini_branch_and_bound_search, min-mini heuristic searchRH_Min_mini_algorithm, Subtree-Pruning-Regrafting (SPR)Subtree_Pruning_Regrafting, and Tree-Bisection-Reconnection (TBR)Tree_Bisection_Reconnection_TBR_ RH_Close_Neighbor_Interchange_CNI heuristic search.  Only the branch-and-bound search is guaranteed to find all the MP trees, but it takes prohibitive amount of time if the number of sequences is large (>15).  For details, please see chapter 7 in Nei and Kumar (2000)Nei_and_Kumar_2000