Connexin 26 hearing loss: carrier status, the 35delG question, and what your file can say
Most people meet the words connexin 26 in an audiology clinic, not a genetics one. This page is for the two situations that bring people here: a child newly diagnosed with hearing loss, and an adult wondering what a carrier result in GJB2 means for a family.
Also known as: GJB2, DFNB1, connexin 26, 35delG.
- Gene
- GJB2
- Inheritance
- Autosomal recessive
- Carrier frequency
- roughly 1 in 30 in many European populations (the 35delG change is especially common)
What it is
GJB2 is the gene for connexin 26, a protein that forms tiny channels between the supporting cells of the inner ear. Those channels recycle the potassium that hair cells need to turn sound into a nerve signal. When both copies of GJB2 carry a disease-causing change, the recycling fails and hearing loss is present from birth. It is called non-syndromic because nothing else is involved: children with GJB2-related hearing loss are otherwise healthy, and the loss itself is usually stable rather than progressive, though the degree varies from mild to profound even within one family.
No single gene explains more inherited hearing loss than this one. In many populations changes in GJB2 account for a large share of hearing loss that is present from birth in otherwise healthy children, and the reason is a handful of ancient founder changes. One of them, 35delG, is common across European-ancestry populations; other populations have their own, such as 235delC in East Asia and 167delT in Ashkenazi Jewish ancestry. Because the carrier state is common and completely silent, most children with this form of hearing loss are born to two hearing parents who had no idea they were carriers.
How it is inherited
GJB2-related hearing loss of this form is autosomal recessive. A carrier has one changed copy and hears normally; hearing loss appears when a child inherits a changed copy from each parent, which carries a one in four chance in each pregnancy for a couple who are both carriers. This page is about that recessive form. A few rare GJB2 changes act in a dominant way and belong to a different clinical picture; they are not what a carrier result refers to.
Not every change in this gene behaves identically either. The classic founder changes, when paired, usually mean significant loss from birth, while a couple of common milder variants tend toward lesser degrees of loss with real variability from person to person. And it is worth saying directly: hearing loss is not a tragedy foretold. Children with GJB2-related deafness grow up healthy, options from early support to hearing aids and cochlear implants are well established, and many Deaf adults regard deafness as an identity rather than a defect. A carrier result is information about biology. What a family does with it is theirs to decide.
What your raw DNA file can and cannot tell you
Start with the chip file. 35delG is common enough that a consumer array may genuinely type that exact position, so a positive call there is a real signal worth confirming with a clinician. The trap is the quiet result. An array that typed one or two European founder changes has said nothing about the more than one hundred other pathogenic GJB2 variants, and nothing at all for someone whose ancestry points to different founder alleles entirely. Aimosti does not report carrier status from a chip file, for this gene or any other, and a typed position is a reason for more care in reading a quiet result, not less.
A whole-genome VCF or gVCF is what our carrier module reads, and it can find GJB2 changes anywhere the sequencing covered. Its blind spots are specific and worth knowing. Some DFNB1 hearing loss involves large deletions near GJB2 that remove a stretch of regulatory DNA and silence the gene without touching its letters; a short-variant file does not represent those, and they matter both alone and in combination with a single GJB2 change. A clear GJB2 result also says nothing about the dozens of other genes that cause hearing loss. This gene is the most common single cause, not the majority of all causes.
If the question is about a child who has hearing loss, a parent's raw data file is the wrong instrument altogether. The clinical route is an audiology evaluation and a clinician-ordered gene panel for the child, which covers the deletions and the wider gene set that a re-analysis of a variant file cannot. A re-analysis of data you already own is a way to look at what you paid for once. It is not a substitute for carrier screening, and not a diagnostic work-up for anyone, least of all a child.
A genotyping chip reads only a few hundred thousand pre-selected positions, about 0.02% of your genome, chosen for common variation. It does not sequence the rest, so it cannot find the rare or novel pathogenic variants the clinical and carrier modules look for: a “no finding” from chip data means “this chip never looked”, far more so than with whole-genome sequencing. Chip data suits common-variant traits, pharmacogenomic tag SNPs, and haplogroup ancestry, not clinical or carrier screening.
Common questions
My child was just diagnosed with hearing loss. Can my raw DNA data tell us if it is connexin 26?
Not in a way you should lean on. Your file could at most show whether you carry a GJB2 change, which is one half of one hypothesis. The answer for the child comes from the child's own clinical testing, ordered alongside the audiology work-up, which covers the large deletions and the many other hearing-loss genes that a consumer variant file does not.
Can 23andMe show the 35delG variant?
Arrays may type that position, and some vendors have included a small set of GJB2 changes in their carrier reports. A positive call there is worth taking to a clinician. A clear call is much weaker: it covers only the listed positions, tuned to European founder alleles, so it is not a rule-out. Aimosti does not issue carrier status from chip data.
My whole-genome file is clear for GJB2. Could a child still have genetic hearing loss?
Yes. The regulatory deletions near GJB2 are invisible to a short-variant file, dozens of other genes cause hearing loss, and residual risk remains after any negative result, clinical ones included. A clear file shifts the odds; it does not close the question.
I am a GJB2 carrier. What does that mean for me and for a pregnancy?
Your own hearing is unaffected. The result matters jointly: only if a reproductive partner also carries a GJB2 change does each pregnancy carry a one in four chance of a child with hearing loss, so testing the partner is the informative next step. If you both carry one, that is the conversation to have with a genetic counselor, with a validated clinical test to confirm anything a re-analysis found first.