I was following the Biostar Handbook up until where I was supposed to create an environment and install the Bioinformatics tools required for the rest of the project when I encountered the error that brought you here: PackagesNotFoundError
.
I had to surf the internet for solutions, and I found bits and pieces of them, which took quite some time. Don't worry; all of the research has been done to streamline the implementation process for you, allowing you to save more time.
Why the Error
The error is a result of the architecture/platform (osx-arm64) introduced to the M-chip MacBook systems, as opposed to the osx-64 platform in the preceding Intel versions. The packages obtained from channels like conda-forge and bioconda are not yet available on the osx-arm64 platform, hence the error.
The Solution
The solution is to find a way to set the platform for the environment created to osx-64 platform (which has all the required packages downloadable via the channels) by following the steps below in this order:
- Install miniconda.
- Add channels to fetch packages like bioconda and conda-forge to the base environment in that order using
conda config --add channels bioconda
andconda config --add channels conda-forge
. - Create an environment, e.g., bioinfo, using the code line
conda create --name bioinfo -y
. This is done to containerize the packages to avoid conflicts like version conflicts. - Activate the created environment using
conda activate bioinfo
. - After that, set your environment platform to osx-64 instead of the default osx-arm64 to enable the installation of packages, as they have not yet been migrated to the osx-arm64 platform. The CLI command used is:
conda config --env --set subdir osx-64
. - Then, install the required package, specifying the version if needed. For example, to install Python version 3.6 on the environment, use the command
conda install python=3.6
. - Finally, install all the tools required for bioinformatics in the environment using
curl http://data.biostarhandbook.com/install/conda.txt | xargs conda install -y
.
Top comments (0)