Dependencies required to run epinano toolset

[a-slide]

Hi @Huanle and @enovoa,

I could not find a detailed list of the dependencies required to run Epinano except the 2 packages mentioned in the INSTALL.sh script. I followed the instructions but I keep having dependencies issue while trying to run Epinano_Predict.py. I identified at least 5 third party python packages imported in the source files but not listed in the install script

• scikit-learn
• numpy
• biopython
• pysam
• matplotlib

However I have no idea which versions are required and so far I could not find a combination which works.

In addition there are a number of non-Python dependencies required to run the full pipeline, including R packages and extra tools such as nanopolish and minimap.

Would you be able to provide a detailed list of all the dependencies and the versions required to run the entire toolset ?

An ideal solution would be a conda package available on Anaoconda Cound, but a conda environment.yml file exported from a local working environment would also be a good solution (https://docs.conda.io/projects/conda/en/latest/user-guide/tasks/manage-environments.html#sharing-an-environment).

Thanks

[Huanle]

Hi @a-slide ,

The dependencies are mentioned in https://github.com/enovoa/EpiNano#pre-requisites. I did miss pysam (0.15.3) and matplotlib (3.3.2) though the foremer one is supposed to be removed because i did not to use it in the end, i will update the script.

All relevant R PKGs were installed for R v3.6.0. I will add them to the pre-requisites as well.

Thanks a lot for the conda pkg management suggection, i will give it a go.

[a-slide]

Thanks @Huanle