Closed liamxg closed 4 years ago
Well you can re-compute log-likelihood scores using one of the program, and pick a tree with the best score. This is what we did in raxml-ng paper, please read supp. material:
https://academic.oup.com/bioinformatics/article/doi/10.1093/bioinformatics/btz305/5487384/
@stamatak @amkozlov , I use the three software to test the topologies, but all the three are different from each other, how can I know which is the best one? I do not access to google, so hope you can image that.