Scientists at the Lawrence Livermore National Laboratory (LLNL) are contributing to the global fight against COVID-19 by combining artificial intelligence, bioinformatics and supercomputing in an effort to discover new antibodies and pharmaceutical drugs to combat the disease.
Backed by its high-performance computing clusters and years of expertise in vaccine and countermeasure development, a COVID-19 response team of LLNL researchers used modeling and simulation, along with machine learning, to identify about 20 promising antibodies from a nearly infinite number of potential candidates, and to examine millions of small molecules that could have anti-viral properties.
Researchers caution that the antibody candidates will need to be synthesized and tested experimentally, which could take time, but progress is being made.
“For several decades, the laboratory has been at the forefront of protecting the country against biological threats of any type,” said Senior Science Adviser Dave Rakestraw, who formerly ran the laboratory’s biodefense programs and is now coordinating its COVID-19 technical response.
“We’ve been putting a large amount of focus for the last six years on applying the computational resources at LLNL to accelerate the timescales for developing a response to an emerging biological threat,” Rakestraw said. “We’ve done that by using our extensive computational capabilities, staff and computer infrastructure, and developing partnerships with universities, drug companies and tech companies.”
“That effort,” he said, “has put us in a position where we have tools now that are applicable to helping with the current response.”
When the COVID-19 outbreak began, LLNL’s Adam Zemla developed a predicted 3D protein structure of the virus that was downloaded and used by more than a dozen outside research groups. Since then, other researchers have determined the actual crystal structure of the key protein from the coronavirus that causes COVID-19, which closely matched the predictions.
Armed with the predicted 3D structure and a few antibodies known to be effective against the virus that causes a similar disease, SARS, a team led by Daniel Faissol and Thomas Desautels used high-powered computing and artificial intelligence to virtually screen antibodies capable of binding to the coronavirus that causes COVID-19.
The modeling was the first of its kind to integrate experimental data, structural biology, bioinformatics (the science of analyzing complex biological data such as genetic codes) and molecular simulations driven by a machine-learning algorithm to design potential antibodies.
“Our approach, while still being developed, is aimed at designing high-quality antibody therapeutics or vaccines in extremely rapid timescales for scenarios where waiting for rounds of time-consuming experimental steps is not an option,” said Faissol. “Experimental data and structural bioinformatics are important components to enable high-quality predictions, but integrating machine learning and molecular simulations with high-performance computing is the key to enabling the speed and scalability we need to search and evaluate huge numbers of possible antibody designs.”
The approach has not only sped up the process, narrowing down the number of candidate antibodies from 1,039 possibilities to a handful in a matter of weeks, but also pointed scientists at areas where they may not have otherwise looked.
“Now we’re not just searching blindly,” said Jim Brase, deputy associate director for data science. “We’re actually creating structures that we think are in the proper part of the design space. Then, we do our evaluations on those. We’ll get novelty, and we hope, a higher percentage of real validated answers out of this approach at the end.”
Researchers said they are just beginning to look at the data and are currently working to begin synthesis of the antibodies, as well as set up testing and evaluation of the designs, through both internal efforts and external collaborations.
Another component of the multi-pronged response involves antiviral drug design. A group of scientists led by Felice Lightstone and Jonathan Allen recently used the supercomputing cluster at LLNL to perform virtual screening of small molecules against two COVID-19 proteins.
Using software created by LLNL scientist Xiaohua Zhang, the team performed a large-scale computational run to screen 26 million molecules against four protein sites (totaling more than 100 million calculations) to identify compounds that could possibly prevent infection or treat COVID-19.
“Using the computational tools and data that we created from our American Heart Association’s Center for Accelerated Drug Discovery, we were able to computationally screen these molecules quickly and at such a large scale,” Lightstone said. “This is the first step toward finding a new antiviral. We developed a whole pipeline for drug design and plan to continue in the coming weeks, ending with experimental testing of the predicted molecules. This should speed up the drug design process.”
Some models being used to determine the safety of the antiviral molecules are derived from a system developed through the Accelerating Therapeutics for Opportunities in Medicine consortium, a project aimed at speeding up cancer drug discovery. That work has helped LLNL evaluate molecules quickly.
LLNL scientists called the COVID-19 pandemic a “wake-up call” signifying the need for longer-term investment and sustained government-wide effort, particularly in applying high performance computing to personalized medicine.
“The laboratory anticipated this kind of situation in pursuing a predictive biology initiative,” said Shankar Sundaram, director of LLNL’s Center for Bioengineering. “The reason we were able to jump onto this quickly was not just because we had the capabilities, but because we’ve been thinking about these scenarios for a long time.”
Source link Google News