DrugCell Oracle
Interpretable Deep Learning for Drug Response
via Visible Neural Networks (VNNs)

What is DrugCell?

DrugCell is a “visible” neural network (VNN) that predicts anti-cancer drug responses by modeling the hierarchical organization of a human cancer cell. Genotypes and drug structures induce differential patterns of activity on cellular subsystems, enabling in silico investigations of the molecular mechanisms underlying cancer drug response. This web portal (DrugCell Oracle) provides access to the DrugCell model and can be used in two ways, via the VNN Browser and the Genotype Analyzer.

Code Availability

The DrugCell codebase, along with instructions to set up the required computational environment, is published in a GitHub repository at: https://github.com/idekerlab/DrugCell. This repository also contains the pre-trained DrugCell model used for the DrugCell Oracle analysis portal provided by this website.

Publications

Please cite the following when utilizing this application in your study:

  • Park J, Kuenzi BM, Otasek D, Churas C, Liu S, Pratt D, Kreisberg JF and Ideker T, DrugCell Oracle: A web portal for prediction and pathway interpretation of drug response and synergy (Publication In Process)
  • Kuenzi BM, Park J, Fong SH, Sanchez KS, Lee J, Kreisberg JF, Ma J, Ideker T. Predicting drug response and synergy using a deep learning model of human cancer cells. Cancer Cell. 2020 Nov 9;38(5):672-84.

© 2020, UC, San Diego / Built with Gatsby