Closed James-Brundage closed 9 months ago
I believe the issue was the python version for the install. Issue was resolved with the following:
conda create -n jnb_env python=3.11
conda activate jnb_env
pip install --upgrade pip setuptools==57.5.0
pip install numpy scipy pandas matplotlib reportlab biopython pyfaidx pysam pyvcf
pip install cnvkit
Prior to this, I needed to install the required R scripts as well.
I am attempting to install CNVKit in a clean conda environment, but the enviornment will not solve. My commands are as follows:
conda create -n jnb_env
conda activate jnb_env
conda install cnvkit
However, I get the following output:
At first glance, this looked like a dependency installation issue. When I installed Miniconda, as suggested by the cnvkit documentation I added the bioconda, conda-forge and defaults via these commands:
I double checked that these channels were actually available via
conda config --show channels
which gave:I then tried to manually install the cnvkit dependencies via
conda install
in jnb_env. I was able to install all of them except for pysam and pyvcf. This remained true even after I specified the bioconda channel for installation. I was able to pip install pysam but not pyvcf.I have tried these approaches on both MacOS (macOS Ventura 13.0) and Linux (CentOS Linux 7 (Core)) each with python 3.7 in the conda environment.
What am I doing wrong in the installation process? I have not seen this issue posted, so I assume I am making an error somewhere! Thank you for your help!