CANcer Distributed Learning Environment (CANDLE)
CANcer Distributed Learning Environment (CANDLE) is an open source, collaboratively developed software platform that provides deep learning methodologies for accelerating cancer research.
CANDLE supports the NCI-Department of Energy (DOE) Collaboration.
Driven by scientific challenges in cancer research defined early in the Collaboration, CANDLE capabilities enable advances in exascale computing through support from DOE’s Exascale Computing Project. DOE’s Argonne National Lab spearheads the project with participation across the NCI-DOE Collaboration national laboratories: Oak Ridge National Lab, Lawrence Livermore National Lab, Los Alamos National Lab, and NCI’s Frederick National Lab for Cancer Research. To further develop CANDLE, scientists across the Collaboration work in a genuine team-science manner with combined expertise in cancer and computational, data, and physical sciences.
Features of CANDLE
The CANDLE platform allows you access to key features at the forefront of the data science frontier shaped by advances in deep learning.
- Hyperparameter optimization efficiently identifies the most effective model implementations.
- Scalable data parallelism speeds machine learning where very large data are required.
As a scalable system, you can install and run the CANDLE platform on systems ranging from laptops to the largest supercomputers available for scientific research.
CANDLE leverages the best, emerging, open-source innovations developed by the deep learning community. You can access, adapt, and use these innovations to address your own challenges because CANDLE allows for the extension and delivery of a scalable platform for deep learning.
- ECP CANDLE GitHub Organization hosts the benchmark codes, documentation, tutorials, and database schema.
- ECP CANDLE FTP Site hosts all the public data sets for the benchmarks from the three pilots.
The CANDLE benchmarks deliver working examples of large-scale deep learning applied to different cancer research challenges. To provide a performance baseline to measure progress and improvement and implement deep learning architectures relevant to scientific challenges in cancer research and opportunities in exascale computing, the developed benchmark models were developed.
Through the deep learning capabilities of CANDLE, demonstrated in the benchmarks, you can produce faster results in several areas:
- Identification of key molecular interactions, based on molecular dynamic simulations of proteins, specifically RAS
- Predictions of tumor response to drug treatments, based on molecular features of tumor cells and drug descriptors
- Better characterization of cancer patient trajectories and outcomes using a growing compendium of clinical information
Connecting the Cancer Community
CANDLE helps bring the cancer research community together, with hands-on workshops that offer opportunities for researchers to share insights and benchmark examples that demonstrate how deep learning and CANDLE can be used broadly to accelerate cancer research.
The Future of CANDLE
The CANDLE project has produced software that you can use. Work is underway for new releases of CANDLE, including:
- model optimization joined with parallel computing to increase capabilities for even larger amounts of data, and
- development of new areas in which deep learning can accelerate cancer research.