150 "Switches" Hiding in Junk DNA Turn Out to Run Brain Genes
A CRISPR screen tested a thousand candidate switches in the non-coding DNA once written off as junk. About 150 do real work, and many of them control genes tied to Alzheimer's. Function keeps turning up where the old story expected leftovers.
Published in Nature Neuroscience, December 18, 2025. Summarized August 22, 2026.
- What are these "switches"?
They are enhancers: short stretches of non-coding DNA that control when and how strongly other genes turn on. Researchers at the University of New South Wales tested about 1,000 candidate enhancers in human brain-support cells called astrocytes, and found roughly 150 with a real, measurable effect on gene activity.
- How did they prove a sequence actually does something?
They used CRISPR interference to silence each candidate switch one at a time, then read the result with single-cell RNA sequencing to see which genes changed. Only sequences that produced a measurable shift in gene activity counted as functional, which separates genuine switches from inert stretches of DNA.
- Why does this connect to Alzheimer's?
A large share of the roughly 150 confirmed switches control genes already linked to Alzheimer's risk. That points to some disease risk operating through the non-coding regulatory DNA, not only through the protein-coding genes, and the map is now being used to train AI models that predict which non-coding sequences do real regulatory work.
- Does this overturn the idea of "junk DNA"?
It adds to a long line of results chipping away at it. Roughly 98 percent of the human genome does not code for proteins, and much of it was once dismissed as junk. Findings like this keep showing precise regulatory function in that so-called junk.
Function keeps turning up where "junk" was predicted
The label "junk DNA" was not a neutral description. It was a prediction, rooted in the expectation that most of the genome is evolutionary debris. Design-friendly voices made the opposite prediction: that function would be the rule, not the exception, in a genome that reads more like layered, purposeful code than like clutter. Studies like this one keep landing on the second side of that bet.
Our inference, and we label it as ours: dense, organized, functional information across most of the genome is what you would expect if the genome is engineered, and is a steady surprise to a story that expected mostly leftovers. The UNSW team was hunting Alzheimer's mechanisms, not settling that debate. But their result is a data point in it, and an honest digest should say which way it points.
Sources
- The 98% mystery: Scientists just cracked the code on "junk DNA" linked to Alzheimer's Primary
ScienceDaily, on the Nature Neuroscience paper (UNSW; Green, Voineagu et al.) · December 19, 2025 - Junk DNA (topic archive) Commentary
Discovery Institute, Center for Science & Culture