Conda vs Mamba
Mamba … the better snake?
As you have seen when you created the environments in the previous tutorials, conda can be a little slow when it tries to install tools. Especially the Solving dependencies step can take several minutes and sometimes even get stuck in a loop.
Mamba is a re-implementation, i.e., a re-write, of the Conda package manager using another programming language (C++ to be exact) that offers higher speed and more reliable dependency and environment resolution. It understands basically all conda commands and can even be installed inside an existing conda installation.
Install Mamba
To get Mamba we simply install it into the base environment of our existing conda installation using the following command
course_user> conda install mamba --name base -c conda-forge
Once the installation is finished let’s create another conda environment ont_mamba and install python 3.10 and porechop into it. However, this time we use the command mamba instead of conda
course_user> mamba create --name ont_mamba python==3.10 porechop
Follow the installation process on the screen, press Enter when asked to proceed.
Once the installation is done you can use the mamba environment exactly like a conda environment. You can even activate it using conda
course_user> conda activate ont_mamba
(ont_mamba) course_user>
The main difference is that Mamba is often faster and more reliable than Conda.
Congratulations, you are finally ready to do some Bioinformatics … but maybe get a tea and a cookie first :)