Python packages:
R packages (within anaconda/miniconda):
Drfeelgood is a package that takes a list of genes and enriches this list of genes against a few different biological levels (KEGG, reactome, GO and protein-protein interactions). It then enriches the lists that come out of the first enrichments against a dataset containing drugs with their corresponding targets. This enrichment is done with only the fisher's exact test. These enrichments are then ranked by their pvalue. When the lists have been ranked an average ranking is calculated. A list of drugs will be returned sorted by the average ranking.
- init.py : To tell Python this is a package.
- biomart.py : In case you want to turn ensembl protein id's into entrez gene id's.
- databases.py : Preps all the databases and does the first enrichment.
- drfeelgood.py : Calls the first enrichments and initiates the second enrichment.
- proteinset.py : Code for an enrichment.
- ranking.py : Contains code for the first and final ranking. The first ranking just ranks the enriched lists from the second enrichment. The final ranking calculated an average ranking and sorts the dataframe based of that ranking.
- aucroc.py : This piece of code calculates the AUC and can make an ROC (optional).
This is a class that takes a list of genes and uses heat diffusion to predict drugs that are related to the genes. A ranking of drugs will be returned based on the amount of heat that the drug ends up with.
The other two directories are test_codes and intervention-genes. Test_codes contain codes that were either used for the preprocessing of some datasets or for testing. Intervention-genes contains codes that were used to find the differentially expressed genes in different lifestyle intervention papers.