Design of highly functional genome editors by modeling the universe of CRISPR-Cas sequences

Abstract

Gene editing has the potential to solve fundamental challenges in agriculture, biotechnology, and human health. CRISPR-based gene editors derived from microbes, while powerful, often show significant functional tradeoffs when ported into non-native environments, such as human cells. Artificial intelligence (AI) enabled design provides a powerful alternative with potential to bypass evolutionary constraints and generate editors with optimal properties. Here, using large language models (LLMs) trained on biological diversity at scale, we demonstrate the first successful precision editing of the human genome with a programmable gene editor designed with AI. To achieve this goal, we curated a dataset of over one million CRISPR operons through systematic mining of 26 terabases of assembled genomes and meta-genomes. We demonstrate the capacity of our models by generating 4.8x the number of protein clusters across CRISPR-Cas families found in nature and tailoring single-guide RNA sequences for Cas9-like effector proteins. Several of the generated gene editors show comparable or improved activity and specificity relative to SpCas9, the prototypical gene editing effector, while being 400 mutations away in sequence. Finally, we demonstrate an AI-generated gene editor, denoted as OpenCRISPR-1, exhibits compatibility with base editing. We release OpenCRISPR-1 publicly to facilitate broad, ethical usage across research and commercial applications.

Conditional language models enable the efficient design of proficient enzymes

“…1b ) and generating 20 sequences per parameter set. We tested a sampling generative procedure 26 , with a temperature of 1, max_length of 1024, top_p of 1, repetition penalty 1.2 and 1.3, and top_k for the values from 5 to 20, and 30, 50, 100, 200, and 458. Accuracy to match the natural distribution was computed as the sum of the absolute differences between all amino acid pairs.…”

“…Protein language models (pLMs) have likewise showcased impressive success, including predicting the effect of mutations 14 and protein structures 15,16 by learning statistics of coevolving residues 17 and allowing the step-wise design of de novo proteins 18 . In the context of enzyme design, pLMs and other generative models have provided artificial variants for several enzyme families 19,20 , including artificial TEM-1 β-lactamases 21 , lysozymes 22 , luciferases 2 , malate dehydrogenases 23 , superoxide dismutases 24 , chorismate mutases 25 , and CRISPR-Cas genome editors 26 in the last three years alone 27 . While these works highlight the potential of AI architectures in the protein realm, the implementation of a model that produces highly active enzymes without the need for further training and with high success rates remains a longstanding goal in the field.…”