UC Davis researchers run molecular dynamics and machine learning models twice as fast as before using Oracle High-Performance Computing.
“We use Oracle’s high-performance computing (HPC) platform to help us run atomic-resolution investigations of ion channel functions and ion channel drug interactions, we’re also able to more accurately predict the safety and efficacy of preclinical drugs.”
For decades, researchers at the University of California Davis School of Medicine have been working with drug makers and regulators to help them bring new pharmaceuticals to market quickly, while preventing potential health risks.
One of the most significant risks is drug-induced cardiac health risks, known as cardiotoxicity. Through their work, UC Davis researchers Igor Vorobyov and Colleen Clancy began challenging the medical industry’s standard, two-dimensional model of testing drugs to find more heart-safe options.
By running molecular dynamics and machine learning models on Oracle Cloud Infrastructure (OCI), the UC Davis research team can more quickly assess risk factors for drugs by analyzing a person’s entire physiology, including protein molecules, cells, tissues, organs, gender, and any pre-existing heart conditions.
OCI’s HPC platform helps us run 50 different simulations at once, which allows us to test all sorts of conditions and ensure that our research is not limited by the speed of our simulations.
Why University of California at Davis Chose Oracle
UC Davis researchers Igor Vorobyov and Colleen Clancy, needed far more computing power than they could access from the university’s on-premises server clusters and other compute resources.
They turned to Oracle for Research, which supported Vorobyov, Clancy and their team with a one-year Oracle for Research grant, providing access to free Oracle Cloud and technical collaboration. The UC Davis research team can now run multiscale molecular dynamics simulations, including 500 million energy and force computations on more than 100,000 different atoms using high-performance computing (HPC) on Oracle Cloud Infrastructure.
Oracle High-Performance Computing is helping UC Davis researchers analyze more variables at a much larger scale than they could when they ran their simulations using the university’s on-premises resources.
By provisioning hundreds of compute cores and Nvidia Tesla P100 and V100 GPUs on OCI, UC Davis researchers can more quickly spin up their drug testing environments, in the size and shape they require, without the overhead of procuring and maintaining a changing mix of hardware.
And, with Oracle Cloud Infrastructure’s distributed architecture, it’s now much easier to process multiple tasks simultaneously.
Using HPC on OCI, the researchers can now run microsecond-long molecular dynamics simulations of atomic-resolution structures to test drugs for cardiotoxicity, and then link these “atomistic” models to millisecond- and second-long simulations of “functional” models, including channels, cells, and tissues.
To improve the level of performance while linking these molecular and functional scale models, UC Davis researchers run their simulations on an OCI bare metal compute instance, using 12-core Intel Xeon CPUs.
With a preinstalled virtual environment on OCI Data Science, which has a Jupyter Notebook integrated development environment and all the Python programming language and machine learning libraries, including PyTorch, NumPy, Pandas, and scikit-learn, the researchers are also able to complete machine learning simulations in about 700 seconds, which is twice as fast as when they were running it locally.
While the researchers have focused their work on determining drug-induced cardiotoxicity, they believe that by running their models on OCI HPC and OCI Data Science, they now have the compute power and scale to help drug makers and clinicians tackle all sorts of diseases, from cancer to metabolic disorders to inflammation.
Since the university began running its molecular dynamics and machine learning models on OCI HPC and OCI Data Science, their models are more easily repeatable, and they can be applied more broadly and translated to different ages, male and female sexes, and animal species.