Jonah Marks, who recently earned his PhD in chemical and biochemical engineering, created a better way for scientists to find “transition states” in chemical reactions, a key tool for unlocking new medicines, better catalysts, and advanced materials.

The innovation attracted the attention of Rowan Scientific, a Boston-based computational chemistry start-up, which has adopted Marks’ method as a go-to tool, freely available to scientists and researchers on Rowan’s cloud-based platform.

A transition state is the highest energy point that molecules must pass through as they transform from reactants to products in a chemical reaction. Transition state searches are crucial for understanding how chemical reactions occur, a foundational element for advancing medicine, manufacturing, and other fields.

“It’s incredibly rewarding to see this work recognized and made accessible to the broader research community,” Marks said. “The expanded impact is what really excites me; having the ML-FSM (machine learning freezing string method) incorporated in Rowan's platform enables many more scientists who don't specialize in computational methods to easily use this method to advance their own research.”

Mentored by Joe Gomes, UI assistant professor of chemical and biochemical engineering, Marks created an improved algorithm to make it easier to find these critical molecular structures.

Jonathon Vandezande of Rowan Scientific learned about the algorithm and implementation from a preprint of Marks and Gomes’ paper and contacted them. They soon partnered to adapt the academic code into a modern Python package and then made the tool widely available.

In a blog post, Rowan explained they chose the package developed by Marks and Gomes because “it provides significant improvements” on existing methods and is a great example of an industryacademia collaboration. Marks’ approach offers a better way to represent molecules, a powerful optimization algorithm, and a built-in check to keep calculations reliable.

Back to Iowa Engineer 2026