Closed singing-scientist closed 1 year ago
Hi Chase!
Thanks for checking out Pythia :-) Unfortunately multithreading is not supported. The largest portion of runtime is spent to infer the maximum parsimony trees with RAxML-NG, and RAxML-NG does currently not implement parallelization for the maximum parsimony tree inference. I will raise this issue with my team members (who develop RAxML-NG) and see if we can implement parallelization, but I can't promise anything. I will keep this issue open as a reminder and I will report back if there are any updates :-)
Greetings, Julia
Thank you so much, Julia! Totally understood — I wasn't sure if it supported multithreading or not, but wanted to ask to be sure. It's still lightning fast compared to the tree search, and ~2 hours is not prohibitive for a dataset that ends up being difficulty 0.83! Thanks a ton for the clarification, and the fabulous tool!
Sincerely, Chase
Hi Chase! With our new release 1.1.0 you can specify multiple threads using the --threads
option for parallel parsimony inference. For this to work you need to update RAxML-NG to version >= 1.2.0.
Greetings, and thanks for the awesome tool! Is it possible to specify multiple threads? The documentation suggests that Pythia might be able to take advantage of multiple threads, but I cannot see how to specify it at the command line. (A particularly hard alignment of mine takes a few hours to finish.)
Thanks so much! Chase