Building Blocks

Those goals may be within reach for Assistant Professor of Chemistry and Biochemistry Shuming Chen. 

Harnessing the computing power available through Oberlin’s supercomputing cluster, she uses quantum chemistry software to program computational models, creating virtual experiments and chemical reactions that in the past required extensive laboratory testing. She and her students build up complex molecules on a screen like virtual Legos and manipulate them in  different ways to test out hypotheses just like experimentalists do on a lab bench. 

“It’s still a challenge to predict the outcomes of chemical reactions,” Chen said. “It’s something we increasingly want to phase out because experimentation takes a lot of human and energy resources, along with environmental costs. The holy grail we’re working toward is replacing those initial exploratory experiments with simulated reactions.” 

Chen caught the chemistry bug early in her undergraduate career at Grinnell College, when she learned about valence shell electron pair repulsion (VSEPR) theory as a model to predict molecular structures. “VSEPR theory made me realize that molecules actually have characteristic three-dimensional shapes,” explained Chen. “If you know the number of electrons and bonds connected to central atoms, you can even predict the twists and sheets of something as complex as proteins, the bases of life.”

Chen’s fascination with molecular architecture led her to study the potential of manipulating those structures. Slight variations in molecules can make a drug safer and more effective—or, potentially, deadly—but as she noted, chemistry experiments are often a matter of serendipity with a lot of trial and error involved. Chemical reactions can require a significant amount of energy, and many reagents and solvents essential for chemical experimentation pose severe environmental or health hazards.

One focus of Chen’s lab is using metals as chemical catalysts. This has important real-world implications. For example, platinum and other metals in catalytic converters react with car exhaust to reduce harmful emissions. Biological organisms—including humans—also use metals in their bodies to enable basic metabolism and detoxification.

But metals can also catalyze reactions that wouldn’t normally take place in nature, leading to the creation of novel molecules that treat diseases. After using computational chemistry to identify chemical reactions of interest, scientists can use lab experiments to refine the most desirable molecules and manufacture better drug candidates.

For example, Chen recently published a paper in Organic Letters with undergraduate biochemistry and clarinet performance major Marisa Ih ’25 on using iron as a more sustainable catalyst to create amino acids that don’t occur in nature; this approach can be used to develop antibiotics that mimic the body's natural defenses against pathogens. “Unnatural amino acids allow us to make interesting modifications to protein structures that can dramatically alter their functions,” said Chen. “They actually underpin a lot of new and exciting developments in biological chemistry.”

Chen’s contributions to pharmaceutical development also include a recent project to synthesize rauvomine B, a molecule produced by the poison devil’s-pepper plant (Rauvolfia vomitoria). Total synthesis is an area of chemical research that focuses on assembling simpler, smaller building blocks to make larger and more complex molecules that already exist in nature. Through total synthesis, natural molecules that hold promise as new pharmaceutical drugs can be made in virtually any quantity on demand. In addition to facilitating pharmaceutical development, this also prevents the natural sources of such molecules—often vulnerable plant or animal populations—from being overexploited for global drug supplies. 

For a paper published in the Journal of the American Chemical Society, Chen and undergraduate chemistry student Gabriel Negrao de Morais ’25 collaborated with Myles Smith and colleagues at the University of Texas Southwestern Medical Center to synthesize rauvomine B. 

The Oberlin team contributed by conducting quantum chemical computations to reveal the key to making the ambitious synthetic endeavor work. Insights from these computations showed that the clever attachment of what chemists call a “protecting group” to a critical part of the molecule had the effect of directing the molecule to take its final, desired shape. Now, scientists can test rauvomine B’s potential to fight inflammation and pathogens while sparing the poison devil’s-pepper. 

“So many useful drugs already come from the plant family the devil’s-pepper is part of,” Chen explains. “If we can create more molecules like rauvomine B and even tweak their structures to make them more effective, we could produce them in much larger quantities and for a lower price than if they were extracted from the plants themselves.”