Gene Editing
We are initially pointing our platform towards CRISPR and gene editing, as we believe there is a significant and persistent unmet need that AI is uniquely positioned to address.
The transformation of microbial CRISPR-based enzymes like SpCas9 into gene editing tools has spurred a revolution in genetic medicine. However, current technologies still fall short in many key areas, limiting the promise that gene editing has to dramatically improve human health. Indeed, gene editing systems that are mixed and matched with components lifted from nature often have functional tradeoffs that significantly limit their reach. For example, PAM relaxation expands edit flexibility at the expense of precision, complex gene writing systems expand the scope of potential edits at the expense of efficiency, and large integration systems offer the promise of whole gene insertion but struggle with deliverability.
Our ability to read and write proteins from scratch offers an opportunity to reimagine gene editing systems from the ground up, breaking free of evolutionary constraints or labor-intensive, single-attribute protein engineering approaches. In stark contrast to current approaches, our AI platform enables us to start with final applications and work backward to custom-design solutions. This includes everything from precisely tuning existing CRISPR scaffolds to creating entirely novel editing systems.
OpenCRISPR™
Profluent is committed to advancing innovation and development in the gene editing community, to bring new treatments to patients with major unmet needs. Under the OpenCRISPR initiative, we are releasing the world’s first open-source, AI-generated gene editor. With this launch, Profluent demonstrates the first successful precision editing of the human genome with customizable gene editors designed from scratch with AI.
Through the OpenCRISPR training process, Profluent’s large language models (LLMs) learned from a massive scale of protein sequences overlaid with biological context to generate millions of diverse CRISPR-like proteins that do not occur in nature, thereby exponentially expanding virtually all known CRISPR families.
We have decided to launch OpenCRISPR-1 (one of our AI-created gene editors) as an initial open-source release, making it freely available to license for ethical research and commercial uses. OpenCRISPR-1 maintains the prototypical architecture of a Type II Cas9 nuclease but is >400 mutations away from SpCas9 and nearly 200 mutations away from any other known natural CRISPR-associated protein. We also used our LLMs to generate a synthetic guide RNA that could be assembled with OpenCRISPR-1.
In our recently released manuscript, we describe both the generation and characterization of OpenCRISPR-1. When delivered via plasmids in HEK293T cells, OpenCRISPR-1 showed comparable on-target editing efficiency and higher specificity relative to SpCas9, across a wide range of genomic on- and off-targets. Furthermore, when combined with AI-generated deaminases, OpenCRISPR-1 was able to assume functionality of a base editing architecture, showing robust A-to-G editing at a panel of target sites.
Studies are ongoing to examine genome-wide specificity and behavior of OpenCRISPR-1 as a purified Ribonucleoprotein (RNP) complex. OpenCRISPR-1 is our first release. We encourage folks to use it, test it, and provide feedback so we can iterate and advance the initiative forward.
Frequently Asked Questions
Reach out to info@profluent.bio