Key findings
- The new automated workflow can enable real-time reaction analysis from a desktop computer instead of a lab.
- Unlike conventional benchtop methods, the automated workflow can identify new chemical reaction products within hours instead of days.
- It can accelerate the discovery of pharmaceutical drugs and the development of new chemical reactions.
A new automated workflow developed by scientists at Lawrence Berkeley National Laboratory (Berkeley Lab) has the potential to allow researchers to analyze the products of their reaction experiments in real time, a key capability needed for future automated chemical processes.
The developed workflow – which applies statistical analysis to process data from nuclear magnetic resonance (NMR) spectroscopy – can help accelerate the discovery of new pharmaceutical drugs and accelerate the development of new chemical reactions.
The Berkeley Lab scientists who developed the groundbreaking technique say the workflow can quickly identify the molecular structure of products formed by chemical reactions that have never been studied before. They recently reported their findings in Journal of Chemical Information and Modeling.
In addition to drug discovery and chemical reaction development, the workflow can also help researchers developing new catalysts. Catalysts are substances that facilitate a chemical reaction in the production of useful new products such as renewable fuels or biodegradable plastics.
“What excites people most about this technique is its potential for real-time reaction analysis, which is integral to automated chemistry,” said first author Maxwell S. Venetos, a former researcher in the lab’s Materials Science Division. Berkeley and a former student researcher in materials science at UC Berkeley. He completed his PhD last year. “Our workflow really allows you to start chasing the unknown. You are no longer limited by things you already know the answer to.”
“What excites people most about this technique is its potential for real-time reaction analysis, which is an integral part of automated chemistry.”
—Maxwell K. Venetos, first author and former researcher in the Materials Science Division
The new workflow can also identify isomers, which are molecules with the same chemical formula but different atomic arrangements. This can greatly speed up synthetic chemistry processes in pharmaceutical research, for example. “This workflow is the first of its kind where users can generate their own library and tune it to the quality of that library without relying on an external database,” Venetos said.
Advancement of new applications
In the pharmaceutical industry, drug developers are currently using machine learning algorithms to screen virtually hundreds of chemical compounds to identify potential new drug candidates that are more likely to be effective against specific types of cancer and other diseases. These screening methods look through online libraries or databases of known compounds (or reaction products) and match them to likely drug “targets” in cell walls.
But if a drug researcher is experimenting with molecules so new that their chemical structures don’t yet exist in a database, he usually has to spend days in the lab sorting out the molecular makeup of the mixture: First, by running the reaction products through a purification machine and then using one of the most useful characterization tools in a synthetic chemist’s arsenal, the NMR spectrometer, to identify and measure the molecules in the mixture one by one.
“But with our new workflow, you could do all that work within a few hours,” Venetos said. The time savings come from the workflow’s ability to quickly and accurately analyze the NMR spectra of crude reaction mixtures that contain multiple compounds, a task that is impossible through conventional NMR spectral analysis methods.
“I am very excited about this work as it applies new data-driven methods to the age-old problem of accelerating synthesis and characterization,” said senior author Christine Persson, senior faculty scientist in Berkeley Lab’s Materials Science Division and professor at UC Berkeley in Materials Science and Engineering, which also leads the materials project.
Experimental results
In addition to being much faster than bench-top purification methods, the new workflow has the potential to be just as accurate. NMR simulation experiments performed using the National Energy Research Science Computing Center (NERSC) at Berkeley Lab with support from the Materials Project showed that the new workflow can correctly identify compound molecules in reaction mixtures that produce isomers. and also to predict the relative concentrations of these compounds.
To ensure high statistical accuracy, the research team used a complex algorithm known as Hamiltonian Monte Carlo Markov Chain (HMCMC) to analyze the NMR spectra. They also performed advanced theoretical calculations based on a method called density functional theory.
Venetos designed the automated workflow as open source so that users can run it on a regular desktop computer. This convenience will be useful for anyone from industry or academia.
The technique emerged from conversations between the Persson group and experimental collaborators Masha Elkin and Connor Delaney, former postdoctoral fellows in John Hartwig’s group at UC Berkeley. Elkin is now a professor of chemistry at MIT, and Delaney is a professor of chemistry at the University of Texas at Dallas.
“In designing chemical reactions, we constantly spend time trying to figure out what the reaction is and in what ratio,” said John Hartwig, senior faculty scientist in Berkeley Lab’s Division of Chemical Sciences and professor of chemistry at UC Berkeley. “Some NMR spectrometry methods are precise, but if one is deciphering the contents of a crude reaction mixture containing a bunch of unknown potential products, these methods are too slow to be incorporated as part of a high-throughput experimental or automated workflow. And that’s where this new ability to predict the NMR spectrum can help,” he said.
Now that they’ve demonstrated the potential of the automated workflow, Persson and team hope to incorporate it into an automated lab that analyzes the NMR data of thousands or even millions of new chemical reactions at once.
“I am very excited about this work as it applies new data-driven methods to the age-old problem of accelerating synthesis and characterization.”
– Christine Persson, Senior Faculty Research Fellow, Department of Materials Science.
Other authors of the paper include Masha Elkin, Connor Delaney and John Hartwig of UC Berkeley.
NERSC is a DOE Office of Science user facility at Berkeley Laboratory.
The work was supported by the US Department of Energy’s Office of Science, the US National Science Foundation and the National Institutes of Health.
###
Lawrence Berkeley National Laboratory (Berkeley Lab) is committed to providing solutions for humanity through clean energy research, a healthy planet, and discovery science. Founded in 1931 with the belief that the biggest problems are best solved by teams, Berkeley Lab and its scientists have won 16 Nobel Prizes. Researchers from around the world rely on the Laboratory’s world-class scientific facilities for their own pioneering research. Berkeley Lab is a multi-program national laboratory operated by the University of California for the US Department of Energy’s Office of Science.
DOE’s Office of Science is the largest supporter of basic research in the physical sciences in the United States and works to address some of the most pressing challenges of our time. For more information, please visit energy.gov/science.
