using a different version of codeml

[Bessem CHOUAIA]

Hi,

I have been using the ete3 toolkit to carry out some dn/ds analyses with codeml. For my datasets, this is taking too much time (more than a month for each set). Recently, I discovered that some researchers had developed a GPU version of codeml (https://github.com/rmingming/CuCodeML), and I wanted to use it with ete3 instead of the normal codeml. Is there a way to tell ete3 to use a different version of codeml?

I installed ete3 through conda under ubuntu 22.04

Thank you for your help
Bessem

[ziqi deng]

Hi,

thank you for sharing this very useful repo, it seems very interesting. Users are allowed to define the codeml binary path in EvolTree module, for example 

from ete3 import EvolTree

t = EvolTree('(A:1,(B:1,(E:1,D:1):0.5):0.5);', binpath=''/home/deng/miniconda3/envs/ete4_official/bin/ete3_apps/bin/codeml") 

if you are using command line version, then you can set it up with the flag --codeml_binary [path]

I recommend you to try the above methos, if it didn't work, maybe try to use change the name CuCodeML to codeml to trick the program. Good luck!

Best,

Ziqi 

Bessem CHOUAIA <bessem....@unive.it> 于2022年10月16日周日 11:56写道：

--
You received this message because you are subscribed to the Google Groups "The ETE toolkit" group.
To unsubscribe from this group and stop receiving emails from it, send an email to etetoolkit+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/etetoolkit/c6113975-22c7-4e65-b30a-95ec2314b7b5n%40googlegroups.com.

[Bessem CHOUAIA]

Dear Ziqi,

Thanks for your help. It is finally working. I had to solve a few other problems of compatibility between CUDA and my GPU, but now it's working.

Best regards

Bessem