jcapels / boimmg Goto Github PK

Dockerize the tool to allow the speeding of the process and reducing the need for more infrastructure resources.

Add pipelines for the whole database.

This will include an apache airflow ETL pipeline that will integrate and update the whole database of LIPID MAPS (https://www.lipidmaps.org/), SwissLipids (https://www.swisslipids.org/), and ModelSEED (https://modelseed.org/).

This will have to include

• the cross-references
• SMILES string
• hierarchy
• the relationship between alcohols and fatty acids with lipids
• biosynthetic relationships (that will have to be generated by hand-made reaction rules of RHEA reactions)

After that, a cronjob have to be run every month to integrate new data and update the database.

Currently it is necessary to indicate wich path the data has to follow in the process, this demonstrate that the Apache Airflow code is not triggering the process automatically.