search

flyark / AFM-LIS

Local Interaction Score (LIS) Calculation from AlphaFold-Multimer (Enhanced Protein-Protein Interaction Discovery via AlphaFold-Multimer)
https://www.flyrnai.org/tools/fly_predictome
MIT License
44 stars 6 forks source link

Having trouble installing the tested versions of NumPy and Pandas in a Jupyter Notebook custom environment #1

Closed Franz4272 closed 7 months ago

Franz4272 commented 7 months ago

Hi,

I hope this message finds you well. I am having trouble installing the tested versions of NumPy and Pandas in a Jupyter Notebook custom environment with the tested version of python. The tested version of biopython was installed without a problem.

!conda create --name AFM_LIS python=3.6.7 ipykernel -y !conda install numpy=1.19.4 pandas=1.1.4 biopython=1.78 -y Channels: - defaults Platform: linux-64 Collecting package metadata (repodata.json): done Solving environment: failed PackagesNotFoundError: The following packages are not available from current channels: - numpy=1.19.4* - pandas=1.1.4* Current channels: - defaults

Can I run AFM-LIS with the most up-to-date versions of python, NumPy, and Pandas? Do I need to install openpyxl as well for the excel files? Please let me know the best way to install the required packages when you have time.

Thank you, Franz

Franz4272 commented 7 months ago

AFM_LIS appears to be working after installing openpyxl=3.0.9 in a custom environment with python=3.6.7, biopython=1.78, numpy=1.19.2, and pandas =1.1.5. I am using the script for inputs from ColabFold. I am not sure If there is something better you would recommend. Please let me know what you think.

Thank you, Franz

flyark commented 7 months ago

Hi Franz,

Thanks for letting me know this!

Since I have been using the jupyter notebook in the school server, I haven't notice such issue. I will mention the version of openpyxl I am using to the requirement in README.

I wish this notebook is helpful for your PPI prediction!

Best, Ah-Ram

Franz4272 commented 7 months ago

Yes, it definitely will be! Thank you for making it!