Comparisons among the Novel Measurements Based on Chi Square Criterion for Sequence Dissimilarity and Their Applications to Phylogeny

In this paper, some new measurements based on Chi square test are presented. Protein sequences are characterized by the frequency of occurrence of the 20 amino acids. Each frequency of a protein sequence is deemed as a sample for various populations. The value of Chi square in the Chi square test is used to measure the dissimilarity of each pair of the protein sequence. Furthermore, some transformations based on the Chi square value are done to measure the dissimilarity. For example, taking the unbalance of the length of protein sequences into consideration, we standardize the length of protein sequence as 1000 with the same frequency of amino acid. Other transformations are listed as follow, such as, the P value according to the Chi square distribution and the normal distribution quantile according to the P value. Based on the data for the Eutherian orders using concatenated H-stranded amino acid sequences, we compare the phylogeny trees with these measurements for sequence dissimilarity. In line with the results, some phylogeny trees are agreed with the commonly accepted one for the Eutherians.