HGM2002 Poster Abstracts: 1. Genome Informatics and Annotation
POSTER NO: 13
A Method to Build Standardized Phylogenetic Trees Based on Relative Evolutionary Rates
Toshinori Endo, Soichi Ogishima, Hiroshi Tanaka
Adaptive evolution is an evolutionary change which enables an organism to adapt to a given environment where it lives, which usually is derived from a diversity of genetic background that in turn to be interpreted later as genetic lineages. Individuals that have advantageous characters in terms of fitness should have some genetic difference among others in its genome, such as mutation, duplication and deletion of one or more genes or gene clusters which should have been born in the generation or earlier. Because the primary change would be a single and small event, it is rarely optimal in regard of adaptation and hence it should be followed by a series of adaptive modification, which in turn an accelerated molecular evolution after all. In other words, if an accelerated rate of molecular evolution is observed, it is likely to imply an existence of adaptation pressure in that evolutionary period, or the evolutionary lineage. In this sense, attempts have been made to detect evolutionary rate acceleration using methods such as comparison between synonymous (silent) and nonsynonymous (amino-acid altering) substitution rates and relative-rate test. Both methods are useful and successful to detect adaptive evolutions, i.e. Malaria merozoite surface antigen 2 gene and the primate lysozyme genes. Those methods, however, bear a serious limitation by nature in application. The first one is valid only in a short range of evolutionary time dates back from the present because synonymous substitution is in nature almost neutral to natural selection and thus multiple substitution on the same site would easily be occur in over the time, which prevents us from estimating the number of synonymous substitutions accurately. That means, the comparison is non-sense or impossible if the evolutionary time period is long enough to allow too many synonymous substitution. Unfortunately, many of interesting morphological evolutionary changes seem to occur out of the valid. The method has an advantage in this aspect, because it is applicable to both the nucleotide change and amino-acid change which is more conservative than the former. It uses the relative evolutionary rates of the genes in question for rate-acceleration test of evolutionary changes because the evolutionary rate varies depending on the functional constraint of the genes. This means that this method requires a reference gene to compare so that the evolutionary rate variation can be detected. The problem is that there is not always a good reference with confidence because fluctuation of evolutionary rate, and more importantly, there would not be an gene which evolves like an ideal molecular clock. Therefore, even though an accelerated evolutionary rate of a gene could be detected, it is likely that it was affected by such parameters. In this study, we developed a method to build a standardized phylogenetic tree which is derived from multiple gene trees. Use of multiple gene trees have the fluctuation of evolutionary rate averaged and the confidence limit given as the standard deviation of branch lengths. In cases that branching patterns are different among gene trees, the dominant topology is selected as the template and the different patterned branch is given a negative branch length, which makes a broader standard deviation of particular branch and thus the branching pattern there can be interpreted as less significant. As the consequence, the acceleration of evolutionary rate can be tested by comparing the branch lengths of a gene tree with that of standardized tree in regard of the standard deviation of branch length. The standardized trees constructed by this method is presented. By combination of this tree with a sliding-window analysis to test accelerated evolutionary rate, a functionally evolving sites of genes may be suggested from a large-collection of genomic sequence data.
Other abstracts in same session