The multibillion-dollar deal is aimed at identifying novel targets and advancing medicines in the neuroscience and oncology areas
Clinical-stage biotechnology company Recursion has signed a multibillion-dollar AI drug discovery deal with Roche and its US subsidiary Genentech.
Using machine learning, the partnership will involve identifying novel targets and advancing medicines in key areas of neuroscience as well as in an oncology indication.
The collaboration will see Recursion working with research and development (R&D) units Roche and Genentech to leverage technology-enabled drug discovery through the Recursion Operating System (OS).
As per the terms of the agreement, Recursion will receive an upfront payment of $150m for the usage of its Recursion OS. The company is also eligible for additional performance-based research milestones.
Recursion OS is an integrated, multi-faceted system to generate, analyse and derive insight from its massive biological and chemical datasets.
Roche and Genentech are expected to initiate up to 40 programmes in neurology and cancer over the next ten years.
The partnership will allow Recursion to receive more than $300m for each project, if it is successfully developed and commercialised. The Salt Lake City-based company will also be eligible for tiered royalties on net sales.
Recursion co-founder and CEO Chris Gibson said: “We are excited to partner with Roche and Genentech to bring Recursion’s leading-edge, tech-enabled drug discovery platform, the Recursion OS, to bear against some of the most complex diseases impacting humanity.
“Technology-enabled drug discovery is here, Recursion is leading the space, and we are pursuing some of the most intractable areas of biology with the very best partners by our side.”
Recursion OS is designed to combine wet-lab and dry-lab biology at scale to industrialise and digitise drug discovery.
It will be deployed to phenomically capture chemical and genetic perturbations in neuroscience-related cell types and select cancer cell lines.
Recursion’s convolutional neural networks will analyse the resulting phenomics data to turn it into mathematical representations of biology to help advance therapeutic programmes.
The dataset will be enhanced by extensive single-cell perturbation screening data from Roche and Genentech.
The partnership will work on new machine learning algorithms to produce highly granular maps of human cellular biology.