Pilot 1—Cellular Level Pilot: Predictive Modeling for Pre-Clinical Screening
The cellular level pilot (Pilot 1) of the Joint Design of Advanced Computing Solutions for Cancer (JDACS4C) program focused on developing predictive models, both computational and experimental, to improve pre-clinical therapeutic drug screening.
The pilot team used advanced computational technologies including machine and deep learning to rapidly develop, test, and validate predictive pre-clinical drug efficacy models in collaboration with NCI’s Patient-Derived Models Repository.
The goal of this pilot was to accelerate the identification of promising new treatment options for precision oncology.
Pilot 1 Leads
With shared expertise across the JDACS4C collaboration, this pilot was jointly led by:
- Dr. James Doroshow, Director, Division of Cancer Treatment and Diagnosis, NCI
- Dr. Yvonne Evrard, NCI's Frederick National Laboratory for Cancer Research
- Rick Stevens, Argonne National Laboratory, Department of Energy
Aims of the Pilot
- Develop reliable machine-learning-based predictive models of drug response that enable the projection of screening results from and among cell lines, cell-line xenografts, and patient-derived xenograft models
- Integrate uncertainty quantification and optimal experimental design to assert quantitative limits on predictions and to recommend experiments that will improve predictions
- Develop Hybrid Predictive Models that support the graded introduction of mechanistic models into the machine-learning framework for incorporation of molecular pathways and gene regulatory networks
Progress
The multidisciplinary team worked to develop machine learning-based predictive models trained on experimental data from many sources, including:
- Cancer cell lines
- Organoids
- Patient-derived xenografts
The team established a large database, focused primarily on RNAseq data, to train the deep learning models with over a dozen integrated NCI data sources, including:
Using gene expression data from the integrated database, the team developed deep learning models for predicting tumor response to single drugs and drug combinations.