Genesis Therapeutics Closes Oversubscribed $200M Series BRead more ->
Our Platform
GEMS: Genesis Exploration of Molecular Space
Genesis has created the industry's most advanced molecular AI platform – GEMS – which integrates deep learning and molecular simulations for property prediction, and language models for molecular generation.

GEMS allows Genesis to create first- and best-in-class small molecule drugs with extremely high potency and selectivity, to address challenging and previously undruggable targets.
Our technology
Machine Learning + Physics
Breakthrough in potency + selectivity predictions via novel integration of ML + physics.

GEMS integrates structure-based deep learning methods with molecular simulations and quantum calculations, offering a breakthrough approach to predict potency and selectivity. This enables Genesis to find drug candidates for challenging and previously undruggable targets that lack on-target training data.
Diffusion Models for Docking
Fast and accurate structure predictions for protein-ligand complexes.

Genesis has pioneered state-of-the-art AI methods to predict protein-ligand structures, using diffusion models to efficiently sample the critical 3D structures for protein-ligand binding. GEMS leverages rapid computational generation of highly accurate poses to further enable potency predictions via GEMS' 3D deep learning methods.
Language Models + Chemistry
ADME-conditioned language models generate new ideas for drug candidates.

GEMS can generate millions to billions of drug-like yet diverse molecules. During Lead Optimization, GEMS can also be prompted by our chemists to explore a more targeted region of chemical space. Multi-task ADME models predict key properties, which are integrated into molecular generation.
Publications
Our academia-leading research was the initial component of our expanding portfolio of machine learning and molecular simulation technologies
1
Our AI platform expands on the initial discoveries in PotentialNet — field-leading, peer-reviewed methods for molecular property prediction that co-founder Evan Feinberg invented in Stanford's acclaimed Pande lab.
Key components of our platform have been rigorously tested against current state-of-the-art methods in a collaboration between Stanford and a top-five pharma company. We achieved a step-change improvement predicting 20+ different ADMET properties.