AI designs DNA 'switch' that turns specific genes on, off
In a groundbreaking study, researchers from The Jackson Laboratory (JAX), the Broad Institute of MIT and Harvard, and Yale University have successfully used artificial intelligence (AI) for gene control. The innovative approach involves using AI to create synthetic DNA switches or cis-regulatory elements (CREs). These CREs can accurately regulate gene activity in specific tissues or cell types, solving a long-standing challenge in genetic engineering.
Synthetic DNA switches: A solution to gene control
The biggest challenge in genetic engineering has been controlling where and when genes are expressed in living organisms. Despite advancements in gene-editing technologies such as CRISPR, ensuring these modifications only impact desired tissues or cells continues to be complicated. To tackle this, the research team created synthetic DNA sequences that serve as precise on-off switches for gene expression. This offers more control over genetic interventions.
Synthetic DNA switches show remarkable specificity
Ryan Tewhey, co-senior author of the study, emphasized the uniqueness of these synthetically designed elements. He said, "What is special about these synthetically designed elements is that they show remarkable specificity to the target cell type they were designed for." Tewhey added that this specificity "creates the opportunity for us to turn the expression of a gene up or down in just one tissue without affecting the rest of the body."
AI helps decode DNA grammar of CREs
Every cell in our body has the same genetic code, but not all genes are active in every cell. Regulatory DNA elements such as CREs act as control switches, ensuring the right genes are activated at the right time. Using deep learning, a branch of AI, the research team trained a model to predict CRE activity. This helped them understand how sequences within CREs affect gene activation.
AI-designed CREs tested in animal models
The researchers employed a platform named CODA (Computational Optimization of DNA Activity) to design thousands of novel CREs capable of controlling gene expression in selected cell types. To confirm the functionality of these AI-designed CREs, experiments were performed in animal models, including zebrafish and mice. In one case, a synthetic CRE activated a fluorescent protein only in the livers of developing zebrafish, leaving other body parts unaffected.
AI-designed CREs could revolutionize gene therapies
The precision of these AI-designed CREs could revolutionize therapies that require gene expression in one specific tissue, without impacting others. They offer more control over where and when genes are activated, potentially benefiting various therapeutic applications from treating genetic diseases to optimizing tissue regeneration. As this AI-powered approach to designing CREs matures, it could be used in biomanufacturing or developing advanced treatments for a range of conditions.
