Back to Blog & Insights
Education

A quarter of a baby's brain cells have scrambled DNA. Then something clears them out

Qurexa Editorial Team1 November 20266 min read 0 0
A quarter of a baby's brain cells have scrambled DNA. Then something clears them out

What happened

Researchers have found that a surprisingly large share of brain cells in the developing human brain carry major DNA errors, and that most of those cells are gone shortly after birth. The study, 'Perinatal elimination of genetically aberrant human cortical neurons', was published in Nature on 28 October 2026. It came from Harvard Medical School and Boston Children's Hospital. The team sequenced DNA from individual nerve cells, one at a time. They examined more than 2,300 single neurons, with one report putting the figure at 2,357. The samples came from post-mortem brain tissue held in tissue banks, mostly the NIH NeuroBioBank. The central finding is a before-and-after. In the brain before birth, an average of 25.9% of neurons carried widespread copy number variants, which are large duplications or deletions of chunks of DNA. In some samples the figure approached half of all cells. After birth, that average dropped to 6.2%. Not all of them go. A small percentage of cells carrying these variants persist into adult life, even in healthy brains.

Why this matters

Most of us were taught that every cell in your body carries the same DNA. That turns out to be roughly true and not exactly true. As cells divide, copying errors happen. When different cells in one person end up with different DNA, that is called somatic mosaicism. You are, in a small way, a patchwork. What this study suggests is that the developing brain produces a great many faulty cells, and then has a mechanism for getting rid of most of them around the time of birth. That is quality control on a scale nobody had measured before. Think about what that implies. Building a human brain is not a careful, error-free process. It is closer to mass production with heavy inspection at the end. Make plenty, discard the bad ones. That is a genuinely different mental picture of development, and it is the kind of finding that reshapes a textbook rather than producing a treatment. It also raises an obvious question. If this clear-out is important, what happens when it does not work properly? The researchers suggest the finding could bear on neurodevelopmental conditions.

What the evidence actually says

This is where some of the coverage has run ahead of the study, and it is worth slowing down. Several outlets reported that the brain 'eliminates billions' of mutant neurons. The study's own findings concern the proportion of cells carrying genomic abnormalities, not a verified total count of cells removed. 'Billions' is an extrapolation from a percentage, not a measurement. The more measured wording in the paper's own abstract is the better guide. The autism connection needs the same care. Coverage noted the findings could lead to better understanding of autism and other conditions where neural development goes awry. That is a hypothesis the authors raise. It is not a demonstrated result. This study did not show that faulty clear-out causes any condition, and nobody involved claims it did. There are real methodological limits too, and they are inherent rather than sloppy. The tissue was post-mortem. You cannot sample a living developing human brain, so every study of this kind is a series of snapshots from different individuals rather than one brain followed over time. The before-and-after comparison is across different samples. Sources also disagree slightly on the lead author's name, given variously as Shao or Hsiao, and on the exact neuron count. These are small discrepancies in secondary reporting rather than doubts about the study. What remains genuinely unknown: whether the cells that persist into adulthood matter functionally. The researchers have not settled that, and say so.

Practical advice

There is nothing here to act on medically, and it would be wrong to imply otherwise. But there are useful things to take from it. If you are a parent, this is not a cause for worry. It describes normal development working as it should. A process that clears out faulty cells is a feature, not a fault. Nothing in this study suggests anything you did or did not do during pregnancy. If you are studying biology, this is a good example of a concept that is easy to state and easy to get wrong: somatic mosaicism. Not every cell in your body has identical DNA. It is worth being able to explain why. It is also a useful case study in reading science news. Take the headline 'brain eliminates billions of mutant neurons', then look at what was actually measured: percentages in a sample of around 2,300 cells. The gap between those two statements is where most science misreporting lives. Practising on a harmless example like this builds a habit that matters later, when the subject is a medicine or a diet. If you want to follow this properly, the paper is in Nature and the abstract is freely readable even where the full text is not. Abstracts are usually more cautious than the press coverage of them, which is itself a useful thing to notice. For anything about your own or your child's development, your GP or health visitor is the right place to go, not a research paper.

What to know

A Nature study published on 28 October 2026 by Harvard Medical School and Boston Children's Hospital sequenced more than 2,300 individual human brain cells from post-mortem tissue. It found that an average of 25.9% of neurons in the prenatal cortex carried widespread copy number variants, approaching half in some samples, falling to an average of 6.2% after birth. A small share of affected cells persists into adult life even in healthy brains. This suggests the developing brain produces many genetically faulty neurons and clears most of them around birth. Headlines about 'billions' of neurons are extrapolations, not measurements, and the suggested link to autism and other neurodevelopmental conditions is a hypothesis the authors raise, not something the study demonstrated. Whether the surviving abnormal cells matter functionally is unknown. Sources: Nature, 'Perinatal elimination of genetically aberrant human cortical neurons', 28 October 2026, https://www.nature.com/articles/s41586-026-11190-3; News-Medical, 'Brain development eliminates flawed neurons to ensure genomic integrity', 28 October 2026, https://www.news-medical.net/news/20261028/Brain-development-eliminates-flawed-neurons-to-ensure-genomic-integrity.aspx; Neuroscience News, 'Developing Fetal Brain Weeds Out Genetically Flawed Neurons', October 2026, https://neurosciencenews.com/developing-brain-genetics-neurons-32352/ This article is for general information and does not replace advice from a doctor, pharmacist or other qualified healthcare professional.

#neuroscience#brain development#genetics#Nature#child development#research

Related articles