There is no single “autism gene” carried by one parent. Both mothers and fathers contribute genetic risk for autism, and they do so in different ways. Hundreds of genetic variants spread across nearly every chromosome play a role, and a child’s risk comes from the combined contributions of both parents plus, in many cases, spontaneous mutations that neither parent carries at all.
That said, research does show that mothers and fathers make distinct genetic contributions, and understanding those differences helps explain why autism runs in some families and appears out of nowhere in others.
Why There’s No Single Autism Gene
Autism’s genetic architecture is nothing like a simple inherited condition such as cystic fibrosis or sickle cell disease, where one specific gene from each parent determines the outcome. Hundreds of DNA variants have been identified across the human genome’s roughly 20,000 genes, each adding a small amount of risk. A landmark study using Swedish population data estimated that about 54% of autism’s heritability traces to common genetic variants, the kind nearly everyone carries in some combination. Rare mutations account for a much smaller share, around 3% of the overall variation in risk, even though they can have a large effect in individual cases.
Some forms of autism are tied to a single gene. Fragile X syndrome, tuberous sclerosis, and Rett syndrome each involve a specific mutation and produce autism-like features alongside other medical issues. These are sometimes called “syndromic” autism and represent a minority of all cases. For most autistic people, the genetic picture is polygenic: many small-effect variants adding up rather than one clear cause.
What Mothers Contribute
Research published in Cell Genomics found that mothers of autistic children carry significantly more common autism-related genetic variants than fathers of autistic children. In a study of over 13,000 parents, mothers scored 0.09 standard deviations higher on a polygenic risk score for autism. That may sound small, but it points to a consistent biological pattern: women can carry a higher genetic load for autism without being autistic themselves.
This is known as the female protective effect. Something about female biology, possibly related to hormonal differences, X chromosome dosage, or other mechanisms not yet identified, allows women to tolerate more autism-associated variants before crossing the threshold into a clinical diagnosis. The practical result is that mothers can silently harbor and pass on more risk variants than fathers typically do.
The evidence shows up in sibling patterns too. Unaffected male siblings tend to inherit less polygenic risk than the midpoint of their parents’ combined scores, as if they need to fall below a certain threshold to remain undiagnosed. Female siblings, by contrast, don’t show this same pattern of under-inheritance, consistent with the idea that girls can carry the same genetic load without the same clinical outcome.
What Fathers Contribute
Fathers play a particularly important role through spontaneous mutations, also called de novo mutations. These are genetic changes that appear in a father’s sperm (or less commonly in a mother’s egg) and are not present in either parent’s own cells. A study in Nature Communications found that these spontaneous mutations accumulate in sperm at a rate of about 3.1% per year of the father’s age. A 40-year-old father, in other words, passes on meaningfully more new mutations than a 25-year-old father.
De novo mutations contribute to an estimated 30 to 39% of all autism cases, and that figure rises to 52 to 67% in families with no prior history of autism. This explains why autism so often appears “out of nowhere” in families where no one has a diagnosis. Neither parent carried the variant. It arose spontaneously during the formation of sperm or egg cells, with paternal age being the strongest known predictor.
Beyond DNA sequence, there’s also emerging evidence that chemical modifications to a father’s sperm DNA matter. A study of fathers in an autism-enriched cohort found 94 regions where chemical tags on sperm DNA (a process called methylation) were associated with autism-related traits in their 3-year-old children. Several of these regions overlapped with genes already implicated in autism and matched patterns found in brain tissue of autistic individuals. This suggests that a father’s environmental exposures or age may influence his child’s risk through pathways beyond the genetic code itself.
Why Boys Are Diagnosed More Often
Autism is diagnosed roughly three to four times more often in boys than in girls, and the female protective effect is a major reason why. Because girls can tolerate a higher genetic burden before showing clinical features, a girl who does receive a diagnosis typically carries more risk variants than a boy with the same diagnosis. This also means that when an autistic girl has siblings, those siblings face a higher recurrence risk, because the family’s overall genetic load is likely greater.
Data from the Baby Siblings Research Consortium confirms this: siblings of autistic girls are significantly more likely to develop autism than siblings of autistic boys. Overall, about 20% of younger siblings of an autistic child go on to receive a diagnosis themselves, a rate roughly 10 times higher than the general population.
Some autism-related genes do sit on the X chromosome, which could contribute to the sex difference. Boys have only one X chromosome, so a mutation there has no backup copy to compensate. Girls, with two X chromosomes, have a built-in redundancy. Fragile X syndrome is the clearest example: it’s an X-linked condition that causes autism-like features far more severely in boys.
Inherited Variants vs. Spontaneous Mutations
One of the most useful ways to think about autism genetics is the split between inherited and de novo causes. Common inherited variants, the ones passed down from both parents, account for the largest share of autism’s genetic basis, at least 41% of the total variability in risk. These are variants that individually have tiny effects but collectively shift a child’s likelihood. Both parents contribute these, with mothers typically contributing a somewhat larger share due to the female protective effect.
Rare inherited variants, the kind detectable through clinical genetic testing, explain a smaller fraction, no more than about 17%. De novo mutations, while present in roughly 14% of autistic individuals as detectable copy number changes or gene-disrupting mutations, account for only about 2.6% of the overall population-level variation. Their outsized importance in individual families, especially those with no autism history, makes them clinically significant even though their statistical contribution to the population is modest.
What Genetic Testing Can Tell Parents
If your child has been diagnosed with autism, genetic testing can sometimes identify a specific contributing variant, though it won’t in the majority of cases. The most commonly ordered test is a chromosomal microarray, which scans for deletions or duplications of DNA segments. This test picks up copy number variants and is typically the first-line genetic test recommended after an autism diagnosis, especially when there are additional features like intellectual disability or unusual physical characteristics.
For more detailed analysis, DNA sequencing can examine specific genes or the entire protein-coding portion of the genome. Clinical panels typically screen between 61 and 107 genes that have been linked to autism and related neurodevelopmental conditions. DNA is usually collected from a simple blood draw from the child, though prenatal testing through amniocentesis is also possible.
Results often fall into three categories: a clearly disease-causing variant is found, nothing significant is detected, or a “variant of uncertain significance” is identified, meaning a change was found but its role in autism isn’t yet clear. Parents who undergo testing alongside their child (a setup called trio testing) can help clarify whether a variant was inherited or arose spontaneously. If the variant is de novo, it’s reassuring for recurrence risk in future pregnancies. If it was inherited from a parent, siblings may carry the same variant, though that doesn’t guarantee they’ll be autistic.
The Bottom Line on Parental Contribution
Both parents contribute to autism risk, but in different proportions and through different mechanisms. Mothers, on average, pass on more common inherited risk variants, buffered by a female biology that lets them carry those variants without a diagnosis. Fathers contribute through inherited variants as well, but their unique role is in de novo mutations, which increase with age and account for a large share of autism in families with no prior history. Neither parent is “the one” who carries the autism gene, because autism doesn’t work that way. It’s the result of many genetic inputs from both sides, sometimes combined with spontaneous changes that neither parent could have predicted or prevented.