The GTR+Gamma+I model of DNA substitution is probably the most widely used model in practical phylogenetic inference. However, a valid proof of the identifiability of its parameters (tree topology, root distribution, rate matrix, proportion of invariable sites, and alpha shape parameter) has not been given.
After indicating the gaps in the previously-published proof of the GTR+Gamma+I model, and discussing various mathematical approaches for establishing identifiability results, a sketch of a recent proof of the identifiability of GTR+Gamma will be given.
This is joint work with Cecile Ane' and John Rhodes.
view more