Overview

This documentation describes Augmented Cancer Drug Atlas (ACDA) drug synergy prediction methods. We augmented the drug synergy prediction modeling approach CDA described in Narayan et al. by applying a Random Forest Regression and optimization via cross-validation hyperparameter tuning. For ease of sharing and use we implemented it as a python package. The source code is located at https://github.com/TheJacksonLaboratory/drug-synergy.

Description of the package functionality

This documentation shows what data inputs are necessary and how to use the ACDA API (Application Programming Interface) to generate the synergy predictions and the corresponding visualizations.

Versions change log

0.0.1 Beta release

0.0.2 Added examples and adjusted the package codebase

0.0.3 Update of the train-test split function

0.0.4 Minor code edits to allow to lowercase drug names in preparation functions

0.0.5 Added stratified splitting for cross-validation, added method that averages predictions of ACDA and EN-ACDA

0.0.6 Added a function that averages prediction for Drug A/Drug B and Drug B/Drug A