AI-Powered Gene Editing Tool Open Sourced for Researchers

Biotechnology company Profluent has open sourced an AI-powered gene editing system that would help researchers customize and develop genetic modifications across health care and agriculture.

Profluent developed the OpenCRISPR-1 system by training large language models on CRISPR-based gene editing data. The models would use the data to generate representations of proteins, which scientists could use to assist research into developing disease resistance in crops or improving medical treatments.

Gene editors are tough to develop, requiring vast troves of data and expert knowledge in genetics, computer science and an understanding of how multi-domain proteins, DNA and RNA interact with one another.

Profluent is attempting to democratize gene editing, enlisting the help of generative AI to enable researchers to improve protein engineering at scale.

The tool focuses on editing CRISPR-Cas9, a technique for editing genetic sequences found in bacteria. The CRISPR gene editing process is painstaking, with Profluent claiming its AI-powered gene-editing tool can explore possible sequence variations in what would take “many lifetimes” in just a matter of hours.

The open source release of Profluent’s AI gene editing tool provides wider access to a potentially powerful research tool. It can also be licensed for use in designing related commercial applications, according to the company.

Related:AI Boosts Gene Mutation Discovery In Search For Cures

“Attempting to edit human DNA with an AI-designed biological system was a scientific moonshot,” said Ali Madani, Profluent’s CEO and co-founder. “Our success points to a future where AI precisely designs what is needed to create a range of bespoke cures for disease. To spur innovation and democratization in gene editing, with the goal of pulling this future forward, we are open-sourcing the products of this initiative.”

Researchers previously demonstrated AI’s potential in the world of analyzing genes and related DNA structures.

Google DeepMind created AlphaFold which predicted structures for almost every protein known to science back in 2022.

Meta also developed a protein-folding model, ESMFold, which was 60 times faster than DeepMind’s AlpahFold, though the Facebook parent has since shuttered its entire genetic-focused project as the company refocused its generative AI efforts towards more commercially lucrative efforts.

Founded in 2022, California-based Profluent is developing generative AI models to design and validate proteins for use in biomedicine and drug discovery.

Profluent claims OpenCRISPR-1 is the first AI gene editor freely available to license for both ethical research and commercial uses.

Related:Meta Lays Off Team Behind Revolutionary Protein-folding Model

The company opted to open source the model with hopes to lower gene editing development costs and reduce the barrier to entry for researchers looking to develop CRISPR-based technologies.

“[The] announcement is a watershed moment and the beginning of what we hope will be an iterative process as we embark on this next generation of building genetic medicines,” said Peter Cameron, Profluent’s head of gene editing. “We encourage the gene editing community to pressure test OpenCRISPR-1.”