Working title under construction
In course of the BioDB Labcourse project we planned a framework to compare disease based on pathway and phenotype overlap. For this 3 databases need to be downloaded, structured and turned into easily queriable SQLite databases. A python package should provide the necessary functions.
The preliminary chosen databases are:
- Uniprot (Team 3)
- Wikipathways, Genenames(Team 2)
- Monarch (Team 1)
We as team 3 need to design and implement a database, that connects diseases (OMIM ID's) to proteins (Uniprot accessions). The database should be easily queryable
1- Choose the biomedical database and obtain raw data (txt)
2- Design and implement a data model in python (sqlalchemy)
3- provide a python package that handles querying the database
4- Query the database
4- Document the project (wiki page, doctest, comments)
- Final decision on the protein database (Uniprot)
- Plan data model
- Download Uniprot data (CSV file)
- Implement a function to parse the disease column of uniprot data csv-file
- Parse: http://www.uniprot.org/docs/mimtosp.txt into a list
- Implement the data model in sqlalchemy
- Populate the database using sqlalchemy
- Setup packages
- Query functions
- Document the code (wiki page)
- Mohammed
- (5 other students, names are hiden for privacy)