Autism-related traits, including those historically labeled Asperger’s syndrome, are not inherited predominantly from one parent. Both mothers and fathers contribute genetic risk, though they do so through partly different mechanisms. Twin studies estimate the heritability of autism spectrum conditions at roughly 64 to 91 percent, confirming a strong genetic foundation, but the genetics involved are scattered across hundreds of common variants and occasional rare mutations rather than a single gene handed down from one side of the family. What makes the question interesting is that maternal and paternal contributions look different under the microscope, and understanding those differences matters more than picking a single parent to “blame.”
Asperger’s, Autism, and Shared Genetics
Asperger’s syndrome was folded into the broader autism spectrum disorder (ASD) diagnosis in 2013, but the question of whether it represents a genetically distinct subtype has lingered. A 2024 gene-network study found that all of the gene co-expression patterns seen in autism were also present in Asperger’s syndrome, but three modules unique to Asperger’s were not found in the broader autism group.1Research in Autism Spectrum Disorders. Genetic networks suggest Asperger’s syndrome as a distinct subtype of autism spectrum disorders That suggests Asperger’s shares the core genetic architecture of autism while adding a few wrinkles of its own. For practical purposes, most of the inheritance research applies across the spectrum, and the parent-of-origin findings discussed here are drawn from autism studies broadly because that is where the data are strongest.
A meta-analysis of twin studies placed the heritability of autism spectrum conditions between 64 and 91 percent.2PubMed Central. Heritability of autism spectrum disorders: a meta-analysis of twin studies That range is high compared to most psychiatric conditions, and it tells you that genes are a major driver, even though the environment and random developmental variation still play a role. Critically, that heritability is not concentrated in one parent’s genome. It is spread across many small-effect variants inherited from both sides, plus a sprinkling of new mutations that arise during the formation of eggs and sperm.
What Fathers Contribute
Fathers get a lot of attention in autism genetics research, largely because of two findings: older fathers have a higher chance of having an autistic child, and the father’s sperm accumulates new mutations with age at a measurable rate.
A large Israeli cohort study found that advancing paternal age was associated with increased risk of ASD, with the authors pointing to new spontaneous mutations and possible changes in how genes are chemically tagged as likely explanations.3PubMed. Advancing paternal age and autism Subsequent work confirmed this at the molecular level. An analysis using whole-genome data from family trios found that new single-letter DNA changes pile up at a rate of about 3 percent per year of paternal age.4Nature Communications. Paternal-age-related de novo mutations and risk for five disorders A separate review estimated roughly two additional new mutations per year of a father’s age.5PubMed Central. Paternal age, de novo mutations, and offspring health? New directions for an ageing problem Most of these mutations are harmless, but the more you accumulate, the higher the chance one lands in a gene that matters for brain development.
Beyond outright mutations, the chemical packaging around a father’s DNA appears to matter too. Sperm carries an epigenetic signature, essentially a set of on/off switches layered on top of the DNA sequence. Researchers have identified a set of over 800 regions in sperm DNA where the methylation pattern differs between fathers who go on to have autistic children and those who do not.6PubMed Central. Sperm DNA methylation epimutation biomarker for paternal offspring autism susceptibility A separate study found directionally consistent methylation differences in paternal sperm and in the brain tissue of individuals with autism, suggesting these epigenetic marks in sperm may carry functional consequences.7PubMed Central. Paternal sperm DNA methylation associated with early signs of autism risk in an autism-enriched cohort Some of the genes flagged in these studies, including ones involved in neural signaling and synaptic function, overlap with known autism risk genes.8Urology. Infertility Epigenetic Signatures in Family Clusters: Links Between Male Fertility and Autism Risk
One more paternal contribution worth noting: small insertions and deletions in DNA, known as indels, that arise spontaneously in offspring tend to originate on the father’s chromosome. A study of nearly 800 families found that these frameshift indels increased autism risk by about 60 percent, were more common in affected girls, and arose predominantly from the paternal side.9PubMed Central. De novo insertions and deletions of predominantly paternal origin are associated with autism spectrum disorder
What Mothers Contribute
Mothers contribute risk through a different set of channels, and the picture has grown more complex in recent years. Advancing maternal age independently raises the odds of having an autistic child, even after accounting for the father’s age. One study found that mothers aged 40 and older had about 50 percent higher odds compared with mothers in their late twenties.10PubMed Central. Independent and dependent contributions of advanced maternal and paternal ages to autism risk A large international study covering over five million children confirmed the independent effects of both parental ages, though the relative risk bump for very old fathers (50 and older) was somewhat larger than for older mothers.11Molecular Psychiatry. Autism risk associated with parental age and with increasing difference in age between the parents
A uniquely maternal channel of genetic transmission runs through mitochondrial DNA, which is passed exclusively from mother to child. Researchers studying family groups found that children with autism had more potentially damaging mitochondrial DNA variants than their unaffected siblings, and that the variants transmitted only to the autistic child tended to be more pathogenic than those shared with siblings.12Nature Communications. Association of mitochondrial DNA content, heteroplasmies and inter-generational transmission with autism An earlier study likewise found that the pattern of mitochondrial DNA transmission differed between mother-autistic child pairs and mother-unaffected sibling pairs, pointing to a possible role for mitochondrial variation during fetal development.13PLoS Genetics. Genetic Evidence for Elevated Pathogenicity of Mitochondrial DNA Heteroplasmy in Autism Spectrum Disorder This is a genuinely mother-only pathway: fathers do not pass along mitochondrial DNA.
An unexpected finding muddies the neat narrative of fathers providing most new mutations. In families with autism, a whole-genome sequencing study found that clusters of new mutations were more likely to have originated on the maternal chromosome, the opposite of what was seen in population controls.14npj Genomic Medicine. Genome-wide characteristics of de novo mutations in autism The researchers linked this partly to the presence of structural DNA changes near these mutation clusters, suggesting that some maternal contributions might come through a different mutational mechanism than the steady year-by-year accumulation seen in sperm.
The Female Protective Effect and Hidden Maternal Risk
One of the more striking findings in autism genetics is the so-called female protective effect. Autism is diagnosed roughly three to four times more often in boys than in girls, and the genetic explanation appears to be that females can tolerate a higher load of autism-related genetic variants before crossing the threshold into a clinical diagnosis. A study of over 13,000 parents of autistic children found that mothers of affected kids carried significantly more common autism risk variants than fathers of affected kids.15Cell Genomics. The female protective effect against autism spectrum disorder
This has a direct consequence for inheritance. A mother can carry a substantial genetic load for autism traits and never be diagnosed herself, then pass those variants along to a son who, without the same biological buffer, crosses the diagnostic threshold. It does not mean mothers contribute “more” autism genes in any absolute sense, but it does mean a larger share of maternally inherited risk may be invisible in the family history. When you see a family where “nobody on Mom’s side has autism,” that does not rule out the maternal line as the source. It may just mean the women in that line were protected from reaching the diagnostic threshold themselves.
Sibling recurrence data support this interpretation. A large consortium study found that about 20 percent of younger siblings of an autistic child went on to receive an autism diagnosis themselves, and siblings of female autistic children were more likely to develop autism than siblings of male autistic children.16PubMed Central. Familial Recurrence of Autism: Updates From the Baby Siblings Research Consortium The logic tracks: a girl who is autistic likely carried a higher genetic burden to overcome the female protective effect, meaning her family’s overall risk load is higher than in families where only boys are affected.
Tracing Risk Through Extended Family
A Swedish population study of over 847,000 children offered a direct look at whether maternal or paternal relatives matter more. Children whose mothers had a sibling diagnosed with ASD had roughly three times the risk of autism compared with the general population. Children whose fathers had an affected sibling also had elevated risk, about twice the population rate.17PubMed. Inherited Risk for Autism Through Maternal and Paternal Lineage The maternal side showed a slightly higher relative risk, but the study authors noted this was not greater than what you would expect for second-degree relatives in general, and the difference between the two sides was modest. The takeaway is that inherited risk flows through both family lines, with the maternal side perhaps carrying a slight edge that may reflect the hidden-carrier effect discussed above.
Broader Autism Traits in Parents
Another angle on the “which parent” question is to look at the parents themselves for subclinical autism traits, sometimes called the broader autism phenotype. You might expect that whichever parent contributes more genetic risk would show more of these traits. The picture here is surprisingly mixed. A systematic review found that a higher percentage of fathers showed broader autism traits compared to mothers, consistent with the overall male skew in autism prevalence, but this was not consistent across every study.18PubMed Central. Broader autism phenotype in parents of children with autism: a systematic review of percentage estimates A separate study found that fathers of autistic children scored significantly higher on a measure of autistic traits than control fathers, while mothers and siblings did not differ from controls.19PubMed. Quantitative autism traits in first degree relatives: evidence for the broader autism phenotype in fathers, but not in mothers and siblings
At first glance, this seems to point toward fathers. But it is easily explained by the female protective effect: mothers may carry just as much genetic risk while showing fewer outward traits. The fact that fathers’ traits are more visible does not mean their genetic contribution is larger. It means the same genetic load looks different in a male body versus a female one.
Assortative Mating and Combined Parental Risk
There is a reason both parents’ genetics matter so much: people tend to pair up with partners who share similar cognitive and personality traits, and autistic traits are no exception. A study and accompanying meta-analysis across 16 datasets found a significant positive correlation between partners’ levels of autistic traits.20Scientific Reports. Evidence of partner similarity for autistic traits, systemizing, and theory of mind via facial expressions When both parents carry above-average polygenic risk, their child inherits a combined load from both sides, which may push the child past the diagnostic threshold even if neither parent would meet diagnostic criteria alone. This is one reason it often seems like autism “comes out of nowhere” in a family. Both parents may have subtle traits that, on their own, were never recognized as unusual.
Imprinting and Parent-of-Origin Effects
Some genes behave differently depending on which parent they came from, a phenomenon called genomic imprinting. In most of your genome, both the copy from your mother and the copy from your father are active. But for a small subset of genes, only one parent’s copy is switched on. A study examining empathy and systemizing, two traits linked to the autism spectrum, found evidence that weak empathy was influenced by paternally expressed genes, while strong empathy was equally or maternally influenced.21Evolution and Human Behavior. Testing the imprinted brain: parent-of-origin effects on empathy and systemizing This is preliminary work, and the overall contribution of imprinted genes to autism risk appears to be small compared with the hundreds of ordinary variants involved. But it does mean that for certain traits, the same gene variant could have a different effect depending on whether it came from Mom or Dad.
Non-Genetic Maternal Factors That Look Genetic
Part of what makes the “mother or father” question confusing is that mothers contribute something fathers cannot: the prenatal environment. Several immune-related pathways during pregnancy have been linked to autism risk, and these can look hereditary even though they are not strictly genetic.
Multiple research groups have identified antibodies in the blood of mothers of autistic children that react against fetal brain proteins. These maternal autoantibodies could potentially interfere with brain development during pregnancy.22PubMed Central. Maternal autoantibodies in autism One study found that reactivity to specific protein bands was more common in mothers of autistic children compared with control mothers, and a particular combination of reactive antibodies appeared exclusively in mothers whose children developed early-onset autism.23PubMed Central. Maternal mid-pregnancy autoantibodies to fetal brain protein: the early markers for autism study
The maternal gut microbiome during pregnancy is another area of active research. Changes in a mother’s gut bacteria can alter the metabolites reaching the developing fetus, and animal studies have shown that specific microbial metabolites can promote nerve fiber growth in the fetal brain.24PubMed Central. Maternal microbe-specific modulation of the offspring microbiome and development during pregnancy and lactation Whether these pathways play a meaningful role in human autism remains unclear, but they illustrate why studies sometimes detect a maternal signal that is not purely genetic. A mother who has two autistic children might appear to be passing something on genetically when part of the contribution is actually her immune profile or metabolic environment during pregnancy.
What Genetic Testing Can and Cannot Tell You
For families looking for concrete answers, clinical genetic testing is increasingly available but still has significant limitations. A large study that returned genetic results to over 21,000 autistic individuals found identifiable pathogenic or likely pathogenic variants in about 8.6 percent of cases.25PubMed. Return of genetic research results in 21,532 individuals with autism That means for more than nine out of ten autistic people, no single “smoking gun” variant will show up on a clinical test. The genetic risk in those cases is polygenic, meaning it is smeared across many common variants of tiny individual effect, making it essentially invisible to standard testing.
Research-grade approaches using whole-genome sequencing and statistical models can detect the combined effects of both new mutations and inherited common variants, but these are not yet standard clinical tools. A major study integrating new and inherited variants across over 42,000 autism cases identified several new moderate-risk genes, but even these collectively explain only a fraction of the overall heritability.26Nature Genetics. Integrating de novo and inherited variants in 42,607 autism cases identifies mutations in new moderate-risk genes If you walk into a genetics clinic hoping to learn whether your child’s autism “came from” you or your partner, the honest answer in most cases is that current technology cannot trace it to one side.
When Family History Gives Misleading Signals
Families often do their own informal detective work, scanning both sides for anyone who seems “a little autistic.” This exercise is understandable but unreliable for several reasons. The female protective effect means women in the family can carry high genetic loads with few visible traits. Diagnostic criteria have shifted dramatically over the past few decades, so older relatives who would meet current criteria were never evaluated. And because autism-related traits like preference for routine, intense focus on narrow interests, or social awkwardness exist on a continuum in the general population, spotting them in a relative says very little about whether that person is actually carrying rare high-impact variants versus ordinary personality variation.
The most common pattern researchers see is not a clear maternal or paternal signal but a combination: both parents contributing small-effect common variants, occasionally amplified by a new mutation that arose in the egg or sperm of one parent, against a backdrop of prenatal environmental factors that may have nudged the child’s brain development in a particular direction. Trying to assign credit or blame to one parent misrepresents how polygenic traits work. Both lineages are involved, and the specific mix differs from one family to the next.