By Tom Garlinghouse for the Office of the Dean for Research
Ellen Zhong has received funding through the Eric and Wendy Schmidt Transformative Technology Fund. Her work is one of four Princeton research projects selected for funding this year.
The goal of the Schmidt Transformative Technology Fund is to enable researchers to make leaps rather than incremental advances in the natural sciences and engineering. It supports projects that lead to the invention of a disruptive new technology that can have a major impact on a field of research, or to the development of equipment or an enabling technology that will transform research in a field.
The fund was created in 2009 through a gift from Eric and Wendy Schmidt. Eric Schmidt is the former Chief Executive Officer of Google and former Executive Chairman of Alphabet Inc., Google’s parent company. He earned his bachelor’s degree in electrical engineering from Princeton in 1976 and served as a Princeton Trustee from 2004 to 2008.
Zhong, an assistant professor of computer science, is working with Mohammad Seyedsayamdost, professor of chemistry, to develop a new AI system to transform how scientists identify small molecules.
Determining the 3-D structure of small molecules, a class that includes hormones, vitamins, and most FDA-approved drugs, is essential for understanding how they function and interact in cells. However, current techniques require painstaking and time-consuming experimentation. In this project, the researchers seek to develop an algorithm that can automate the process of determining small molecule structure from nuclear magnetic resonance (NMR) spectral analysis.
In preliminary work, the researchers developed a machine learning algorithm that reliably translated one-dimensional NMR spectra of five-residue peptides into precise molecular structures. The current project will expand this approach to include broad classes of small molecules of various sizes and structure types. The project has three aims: to compile a database of NMR spectra for large-scale deep learning, to develop an algorithm that incorporates context-dependent considerations, and to apply the algorithm to the discovery of novel small molecules.
Zhong and Seyedsayamdost anticipate that the open-source release of this algorithm will significantly benefit scientists in many disciplines and transform drug discovery.
The winning proposals were selected by an anonymous panel of faculty reviewers. Learn more about all four technologies that received funding.