Question_process_Orthofinder

[cristoyerenahs]

Good Afternoon. Orthofinder team.

I have run the Orthofinder (./orthofinder -f mydata) process with 1000 proteomes of different bacteria species, After some days, I had some erros in order to generate the outputs files.

I want to attach the errors.

Could you help about the erros?

Thanks

[Jonathan-Holmes-Bioinformatics]

Hi cristoyerenahs,

OrthoFinder was built for handling eukaryotic genomes in smaller batches than you are working on (1000). It looks like the issue is due to the program opening to many files during its multiprocessing stage for writing the Orthogroup matrices.

I would recommend trying some of the solutions on this thread:

https://stackoverflow.com/questions/16526783/python-subprocess-too-many-open-files

Hope this helps.

[davidemms]

Hi

You can also try the code in the branch 'linear_species_addition', which will be merged into the main branch soon. It has a solution to the problem of too many open files. You will need to install some extras python dependencies: biopython six scikit-learn.

All the best David