Expanding the Frontier of Possible For Patients by Designing Better Therapeutics, Atom by Atom

Animated image of a dynamic protein.
Biologically relevant protein motion (BRPM) typically happens on the nanosecond (10-9 seconds) to millisecond (10-3 seconds) timescale.

A Moving Target

Proteins are dynamic, shapeshifting molecules. Traditional approaches treat protein targets as rigid structures because predicting protein motion is very challenging. However, modulating biological activity often relies on protein motion.

Doing the Previously Impossible

By predicting the fundamental relationship between protein motion and disease, we can create novel small molecules for targets with little or no known chemical matter—disease targets that previously have been considered “undruggable” with small molecules.

Discovering New Oral Treatments in Immunology and Inflammation

The TNF superfamily (TNFSF) represents a major opportunity for advancing new oral treatments for patients. Currently only injection-based biologics are approved for these targets. Our “conveyor-belt” superfamily strategy has progressed programs for multiple TNFSF targets where no known small molecule chemical matter previously existed. If these small molecules prove successful, they could be some of the first oral treatment options in their categories.

Say hello to the power of biologics in a pill.

Model of TL1A Protein

TL1A

Model of CD40L Protein

CD40L

Model of RANKL Protein

RANKL

Model of BAFF Protein

BAFF

Model of OX40L Protein

OX40L

  • Vast Chemical Space

    We explore vast chemical spaces with an emphasis on molecules that can be made in the lab using known building blocks and reactions as part of our design cycle. This ensures that the molecules we ultimately choose to make can be efficiently synthesized.

  • Diverse, Synthesizable Molecules

    The chemical space for each design cycle contains a diverse set of novel molecules that can be synthesized in the lab.

  • Drug-Like Molecules

    We profile each molecule from this vast chemical space for desirable ADME properties based on machine learning prediction algorithms. We select the most promising molecules for rigorous free energy simulations.

  • Enrichment

    We explore vast chemical spaces with an emphasis on molecules that can be made in the lab using known building blocks and reactions as part of our design cycle. This ensures that the molecules we ultimately choose to make can be efficiently synthesized.

Meet the People Making This Possible

Portrait of István Kolossváry
Portrait of Alex Liholips
Portrait of Fabio Trovato