We are building a computational-experimental platform to design therapeutics that can target several disease-causing proteins at once. By designing multi-specific drugs, we are able to create therapeutics that are more efficacious and safer than existing medicines
We focus on creating a new class of drugs targeted against the human kinome, a family of 500 proteins associated with diseases such as cancer, auto-immunity and neurodegeneration
We combine genetic screening, biochemical assays and bioinformatics to predict kinase targets that synergize when inhibited together and kinase anti-targets that when inhibited cause toxicity
Given a set of kinase targets and anti-targets, we use generative machine learning, chemical informatics and medicinal chemistry to rationally design a molecule with the desired binding profile
Our generative machine learning model is powered by our best-in-class kinome binding predictor, our machine learning model to predict the entire kinome profile of a molecule
From initial hits, our computational platform can rapidly iterate through designs for the next series of compounds, optimizing for binding profile, selectivity and ADME properties