Facilitating Drug Discovery with CWL

Presenter: Saul Acevedo


We translated tools developed for the NCI Predicative Oncology Model and Data Clearinghouse (MoDaC), a joint effort between the Joint Design of Advanced Computing Solutions for Cancer (JDACS4C), and the Accelerating Therapeutics for Opportunities in Medicine (ATOM Consortium) programs into CWL allowing for portability, reproducibility, and scalability. The release consists of chemo-informatics tools for integrating cancer treatment features in deep learning graph models. JDACS4C ML models ported to the Cancer Genomics Cloud (CGC) include classifiers, autoencoders, drug response predictors, and Multitask Convolutional Neural Networks. The porting of these tools to CWL on the CGC will allow for lower barriers to entry for new users who aim to use machine learning-driven drug discovery, as well as streamline the development process for more technical users. The accessibility and adaptability of these MoDaC toolsets will allow for the optimization of the drug discovery pipeline by supporting pre-clinical study evaluation, treatment identification, and experimental design.

Please leave your questions for the presenter below!

As an alternative to YouTube, this presentations is also available on ConfTube