search

AstraZeneca / chemicalx

A PyTorch and TorchDrug based deep learning library for drug pair scoring. (KDD 2022)
https://chemicalx.readthedocs.io
Apache License 2.0
702 stars 89 forks source link

Add the DDI Dataset from DeepDDI #28

Closed benedekrozemberczki closed 2 years ago

benedekrozemberczki commented 2 years ago

The triples are here Dataset.

cthoyt commented 2 years ago

For the next dataset we add, I think it would be pretty important to include the preprocessing pipeline as well (e.g., using pystow to reproducibly download the data from its source)

Also for this one, I wonder if it's derived from DrugBank, and if we could re-write the pipeline ourselves using drugbank_downloader for preparing the dataset