Workflow for generating db from PubChem bioassay
- Set up mini-conda
- download and install miniconda https://docs.conda.io/en/latest/miniconda.html, choose yes to add it to .bashrc
- conda create -n bioassay
- conda activate bioassay
- conda install jupyter notebook numpy pandas matplotlib tqdm psycopg2 sqlalchemy
- conda install -c openbabel openbabel
- conda clean -a
- Set up jupyter notebook
- jupyter notebook --generate-config
- to set-up remote access edit ~/.jupyter/jupyter_notebook_config.py
- c.NotebookApp.allow_remote_access = True
- c.NotebookApp.ip = '*'
- c.NotebookApp.open_browser = False
- c.NotebookApp.password = 'sha1:...'
- c.NotebookApp.port = 8080
- Set up jupyter lab (optional)
- conda install -c conda-forge jupyterlab
- conda install -c conda-forge nodejs
- jupyter labextension install @jupyter-widgets/jupyterlab-manager
- Set up rdkit & postgres (though not using rdkit at the moment)
- conda install -c rdkit rdkit rdkit-postgresql
- apt-get install libfontconfig1 libxrender1
- adduser postgres
- su - postgres
- /bin/conda init bash
- exit
- su - postgres
- conda activate bioassay
- initdb -D /postgresdb
- pg_ctl start -D /postgresdb -l /postgresdb/log
- psql postgres
- \password postgres
- create extension rdkit;
- \q
- TODO: setup postgres as a service to start on boot
- finally run 'jupyter notebook' or 'jupyter lab'