Automated peak alignment for 1H-NMR metabolomics using machine learning.
For portability, we will use conda environments. We will also use Jupyter notebooks for exploratory coding prior to converting code to scripts, functions, Classes, etc., so a Jupyter kernel will need to be associated with the conda environment.
First, clone the repository. When you use git clone, a new folder that has the same name as the repository is generated in the current directory that you are navigated to within your terminal/commandline:
> git clone https://github.com/medlocklab/nmr_autoprocessing.git
Create the conda environment; we'll use a conda environment yaml file, which specifies the name of the environment we're creating and all dependencies:
> conda env create environment.yaml
activate the conda environment; you'll need to do this whenever you want to use or modify the environment (e.g., install new packages):
> conda activate nmr_autoprocessing
The environment.yaml configuration file includes the nb_conda package, which improves access of conda environments from Jupyter; after activating the environment, you should be able to start notebooks containing a kernel for the nmr_autoprocessing environment from within Jupyter. Start JupyterLab to check this and run/modify any notebooks:
> jupyter lab