Rowan Announces $2.1M to Build Machine Learning-Powered Computational Tools for Chemistry

Startup aims to replace expensive, slow, quantum mechanics simulations with inexpensive, fast machine-learned potentials.

BOSTON, Dec. 9, 2024 — Rowan, a company building molecular design and simulation tools for scientists, today announced $2.1 million in pre-seed funding. Investors include Pillar VC, AI Grant, and angel investors.

"Modern algorithms and machine learning models can dramatically accelerate scientific R&D by replacing resource-intensive experiments," said Corin Wagen, PhD, CEO and founder of Rowan. "As Rowan helps more scientists use modern computational techniques, our company will accelerate the speed of scientific progress across drug discovery and materials science."

Rowan makes it possible for any scientist to run modern simulation workflows, not just experts in computational chemistry. Users can design, simulate, and analyze molecules and materials through the company's web platform. Rowan's capabilities include:

"Rowan is now my first port of call for QM modeling," said Lewis Martin, chief scientific officer of OpenBench. "The attention paid to pragmatic and benchmarked methods means the software suite is practical to incorporate into our workflows."

"Today, biotech and pharma companies can invest hundreds of millions of dollars developing a single drug, and conventional simulations struggle to accelerate this process," said Tony Kulesa, partner at Pillar VC. "Rowan is building the next generation of powerful computational tools to save these companies an extraordinary amount of time and money as they advance their research."

About Rowan
Rowan is building machine learning-powered computational tools to accelerate chemical innovation. To learn more, visit rowansci.com.

Banner background image

What to Read Next

Hydrogen-Bond-Basicity Predictions for Scaffold Hopping in PDE2A Inhibitors

Hydrogen-Bond-Basicity Predictions for Scaffold Hopping in PDE2A Inhibitors

How new computational workflows can make it possible to design complex modifications to heterocyclic cores.
Feb 14, 2025 · Corin Wagen
Intrinsic Reaction Coordinates

Intrinsic Reaction Coordinates

expanding Rowan's reaction modeling toolkit; verifying transition states; reaction mechanism insights
Feb 6, 2025 · Jonathon Vandezande and Ari Wagen
Reactions from the Bottom Up

Reactions from the Bottom Up

Building up an understanding of how energy barriers and the potential energy surface affect the rate of a reaction.
Feb 4, 2025 · Jonathon Vandezande
A New RDKit-Native API

A New RDKit-Native API

cultural barriers in science; integrating RDKit with quantum chemistry; Rowan's new API; changes to billing
Jan 31, 2025 · Corin Wagen and Spencer Schneider
Hydrogen-Bond Basicity Prediction Made Easy

Hydrogen-Bond Basicity Prediction Made Easy

not all hydrogen-bond donors are created equal; the pKBHX scale; predicting pKBHX in Rowan; case studies & a preprint
Jan 24, 2025 · Corin Wagen
Efficient Black-Box Prediction of Hydrogen-Bond-Acceptor Strength

Efficient Black-Box Prediction of Hydrogen-Bond-Acceptor Strength

Here, we report a robust black-box workflow for predicting site-specific pKBHX values in organic molecules with minimal computational cost.
Jan 24, 2025 · Corin C. Wagen
Benchmarking NNPs, Orb-v2, and MACE-MP-0

Benchmarking NNPs, Orb-v2, and MACE-MP-0

benchmarking as driver of systematic methodological improvement; our new benchmarking website; new NNPs on Rowan; GPU-based inference coming to more users
Jan 17, 2025 · Ari Wagen
Density-Functional-Theory Functionals Quiz

Density-Functional-Theory Functionals Quiz

Ready to test your knowledge of density-functional-theory functionals in a multiple-choice game of "real or fake"?
Jan 10, 2025 · Jonathon Vandezande
Wiggle150: Benchmarking Density Functionals And Neural Network Potentials On Highly Strained Conformers

Wiggle150: Benchmarking Density Functionals And Neural Network Potentials On Highly Strained Conformers

We introduce Wiggle150, a benchmark comprising 150 highly strained conformations of adenosine, benzylpenicillin, and efavirenz, to validate computational protocols involving non-equilibrium systems and guide the development of new density functionals and neural network potentials.
Jan 8, 2025 · Joseph Gair, Corin Wagen, et al., ChemRxiv
The "Charlotte's Web" of Density-Functional Theory

The "Charlotte's Web" of Density-Functional Theory

A layman's guide to cutting your way through the web of DFT functionals, explaining GGAs, mGGAs, hybrids, range-separated hybrids, double hybrids, and dispersion corrections.
Dec 20, 2024 · Jonathon Vandezande