AI Meets CRISPR Unlocking the Genome: “Creme” AI Tool Accelerates Virtual Gene Editing

Cold Spring Harbor Laboratory's new AI tool, “Creme”, offers scientists a revolutionary way to explore the genome, enabling virtual gene-editing experiments at unprecedented scale. By simulating CRISPRi’s gene-modifying abilities, Creme allows researchers to predict and understand gene activity from countless mutations.

Imagine you’re looking at millions upon millions of mysterious genetic mutations. With CRISPR gene-editing technology, a select few of these mutations might have therapeutic potential. However, proving it would mean many thousands of hours of lab work. Just figuring out which ones are worth exploring further would take a lot of time and money. But what if you could do it in the virtual realm with artificial intelligence?

“Creme” is a new AI-powered virtual laboratory invented by Cold Spring Harbor Laboratory (CSHL) Assistant Professor Peter Koo and his team. It allows geneticists to run thousands of virtual experiments with the click of a button. Now, scientists can use it to begin identifying and understanding key regions of the genome.

The program is modeled after CRISPR interference (CRISPRi), a genetic perturbation technique based on CRISPR. CRISPRi allows biologists to turn down the activity of specific genes in a cell. “Creme” lets scientists make similar changes in the virtual genome and predicts their effects on gene activity. In other words, it’s almost like an AI version of CRISPRi.

“In reality, CRISPRi is incredibly challenging to perform in the laboratory. And you’re limited by the number of perturbations and the scale. But since we’re doing all our perturbations [virtually], we can push the boundaries. And the scale of experiments that we performed is unprecedented — hundreds of thousands of perturbation experiments," explains Koo.

DNA Editing

New Gene-Editing Tool for Treating Genetic Disorders

Koo and his team tested “Creme” on another AI-powered genome analysis tool called Enformer. They wanted to know how Enformer’s algorithm makes predictions about the genome. Questions like that are central to Koo’s work, he says.

“We have these big, powerful models. They’re quite compelling at taking DNA sequences and predicting gene expression. But we don’t really have any good ways of trying to understand what these models are learning. Presumably, they’re making accurate predictions because they’ve learned a lot of the rules about gene regulation, but we don’t actually know what their predictions are based off of.”

With “Creme”, Koo’s team uncovered a series of genetic rules that Enformer learned while analyzing the genome. That insight may one day prove invaluable for drug discovery. “Understanding the rules of gene regulation gives you more options for tuning gene expression levels in precise and predictable ways,” says Koo.

With further fine-tuning, “Creme” may soon set geneticists on the path to discovering new therapeutic targets. Perhaps most impactfully, it may even give scientists who do not have access to a real laboratory the power to make these breakthroughs.

Toneyan, S., et al., “Interpreting cis-regulatory Interactions from large-scale deep neural networks”, Nature Genetics, Sept 16, 2024. DOI: 10.1038/s41588-024-01923-3

(ID:50168149)