Researchers developed a deep learning model, TIGER, that accurately predicts on- and off-target activity of RNA-targeting CRISPR tools. This new approach allows the fine-tuning of gene activity in human cells.
Researchers from New York University, Columbia Engineering, and the New York Genome Center have combined a deep learning model with CRISPR screens to develop precise gene controls. This technology allows the expression of human genes to be controlled in various ways, offering potential for the development of new CRISPR-based therapies. RNA-targeting CRISPRs, which use the enzyme Cas13, have a wide range of applications, including RNA editing, gene expression regulation, and drug candidate screening.
The researchers created a platform for RNA-targeting CRISPR screens to better understand RNA regulation and identify non-coding RNA functions. The study aims to maximize the activity of RNA-targeting CRISPRs on the intended target RNA while minimizing activity on other RNAs to avoid detrimental side effects. They developed a deep learning model called TIGER, trained on data from CRISPR screens, that accurately predicts both on-target and off-target activity of RNA-targeting CRISPRs, outperforming previous models.
TIGER’s predictions have the potential to modulate gene dosage and may be applicable in treating conditions such as Down syndrome, schizophrenia, Charcot-Marie-Tooth disease, and certain cancers. The combination of artificial intelligence and RNA-targeting CRISPR screens holds promise for the development of new RNA-targeting therapies, enabling precise control and avoiding undesired off-target effects.