Generate Biomedicines Uses AI to Create Proteins That Have Never Existed

Generate Biomedicines Uses AI to Create Proteins That Have Never Existed

New model programs entirely novel proteins, not descendant from natural examples, with the potential to design towards any biological function

The ability to design proteins across a wide range of modalities will enable therapeutic development on a scale previously unseen

CAMBRIDGE, Mass.–(BUSINESS WIRE)–#artificialintelligence–Generate Biomedicines today published a preprint describing a breakthrough in protein science that unlocks the ability to generate novel proteins, at will, towards desired functional or other properties. This is enabled by their new machine learning (ML) model, Chroma, which Generate believes will usher in a new age in which proteins—the most functional molecules of life—can become truly programable, with a transformative impact on human health.

The work reinforces that the immense biodiversity of our planet, encoded by protein sequences resulting from ~3.5 billion years of evolution, is but a drop in the ocean of all possible protein concepts. Generate has demonstrated how these concepts can now be accessed, through their generative ML-based model Chroma, and used in service of a powerful method for designing novel proteins.

“We believe our model will have revolutionary implications. It is akin to learning how to ‘write’ in the mysterious language of proteins. Now that we have this ability, the possibilities of what we can produce are endless.” said Gevorg Grigoryan, Ph.D., Co-Founder and Chief Technology Officer of Generate Biomedicines. “The transition from conventional drug discovery to generative biology will mean that we no longer discover suboptimal medicines that nature evolved for its own purposes; instead, we will be creating, or generating, purpose-built, highly efficacious medicines.”

The capability to program proteins has promising and important implications for designing novel therapeutics, bringing new medicines to patients faster, cheaper, and better tailored to their specific conditions. As Generate’s platform is constantly learning from each new protein it produces and validates in the laboratory, it will enable therapeutic development on a scale previously unseen.

Read the preprint and the blog post to learn more about Chroma.

As a next step, the manuscript will be submitted for peer review. The preprint describing Chroma has also been submitted to bioRxiv.org.

About Generate Biomedicines

Generate Biomedicines is the first drug generation company, pioneering a machine learning-powered multimodality generative biology platform with the ability to generate new drugs on demand across a wide range of biologic modalities. The platform can drastically improve the speed at which targets, and therapeutics are identified and validated, the specificity of target engagement by generated proteins, and the cost of identifying and developing clinical candidates. The company’s platform represents a potentially fundamental shift in what’s possible in the field of therapeutic development, addressing key challenges of drug discovery and drastically expanding the available search space for novel biomedicines. Generate Biomedicines was founded by Flagship Pioneering after two years of foundational research in its Labs unit and launched in 2020. Learn more about Generate Biomedicines by visiting https://generatebiomedicines.com/ or following the company on Twitter and LinkedIn.

Contacts

Media
Megan McLaughlin

[email protected]

Subscribe on LinkedIn

Get the free newsletter

Subscribe to MedicaEx for top news, trends & analysis

Businesswire is solely responsible for the content of the above news submissions. If there are any violations of laws, violations of the membership terms of this website, or the risk of infringing on the rights of third parties, businesswire will be solely responsible for legal and damage compensation. Responsibility has nothing to do with MedicaEx.

Are you in?

Stay up-to-date with the latest Newsletters, free of charge.