We are leading the field of covalent drug discovery, expanding the druggable proteome and unlocking the next generation of high value small molecule therapeutics across disease areas.
The approval of drugs with covalent mechanisms of action predates rational drug discovery and is historically rooted on serendipitous findings, and reverse elucidation of covalent mechanisms of action and safety profile assessment. At Roivant we are challenging convention and making history.
Decades of accumulated expertise in each of these technical disciplines allow us to accelerate covalency-driven drug discovery across protein target classes, reactive side chains and disease indications.
Our differentiating capabilities to prosecute targets via covalency rest on four cornerstones:
Proprietary Human “Reactome” Database Built on Translational Disease Models
We are expanding covalency by building a proprietary human Disease Reactome database enabled by translational disease models. Using state-of-the-art mass spectrometry (MS) and machine learning (ML), our proprietary chemo-proteomic methods access the “dark” protein sequence space. We capture side chain post-translational status, functionality, reactivity, and expression levels across disease-relevant biosamples.
Activity-Based Proteomic Probes and Proprietary Covalent Small Molecule Library
Our tunable electrophilic biocompatible probes and bespoke screening library are tailored to harness nucleophile diversity and centered on drug-like chemical matter, with efficient representation of high impact ‘reactive warheads’ and proprietary diverse building blocks across size and shape. Together with our Disease Reactome Database, our library is our core differentiated asset. We curate and enhance our collection informed by experimentation and informatics, with flexible assembly to enable screening optionality.
Versatile Chemoproteomics-Based Hit Finding Engine
We screen proteins in their native biological state, to preserve key functional post-translational modification states and interactome. Our flexible and exhaustive screening paradigm allows us to interrogate whole disease proteomes, sub-proteomes, and single proteins of interest, to identify reactive residues, map binding sites, and functional covalent hits, and measure target engagement selectivity in native settings.
Computational Platform for Rational Design and Optimization of Covalent Inhibitors
We combine experimental data with physics-based computational methods within QUAISAR to assess relative pocket ligandability and protein high order structure to predict optimal covalent adduct structures. Leveraging our QUAISAR platform with state-of-the-art ‘Design-Predict-Make-Test-Analyze’ capabilities, we optimize molecules with desired reactivity to advance leads for challenging targets across disease areas.