GAREY, J.R.; Univ. of South Florida: Simulating DNA Evolution: application to genomic analysis

The simulation of DNA evolution is a powerful tool that has been used in the past to evaluate the accuracy of various tree-making algorithms and to explore poorly understood phenomena such as long-branch attraction. One of the problems with previous DNA simulations is that they focused on substitution of nucleotides and did not address the problems associated with insertions and deletions. We are developing a program (DNAsim) that simulates evolution along a user input tree in which substitution rates and indel rates can be independently controlled and different substitution and indel rates are allowed along different branches of the tree. The program maintains the true alignment during the simulation. The output consists of two files, one containing the true alignment in MEGA format and the other with the unaligned final sequences in PIR format. We have used this program to explore the contribution of alignment to long-branch attraction. We are currently implementing the ability to incorporate site-to-site variation during the simulation that will allow the realistic simulation of ribosomal RNA genes as well as protein coding genes. One approach to using genome datasets in phylogenetic analysis is to concatenate multiple genes into a single alignment. DNAsim will allow us to explore the validity of this approach by examining the effect of parameters such as long-branch attraction, inadvertent inclusion of paralogs and the inclusion of genes with drastically different patterns of evolution.