Autism is one of the most strongly inherited conditions in medicine, yet no single gene causes it. Most of the genetic risk comes from thousands of common variants that many people carry, each adding only a little.1PubMed. The Yin and Yang of Autism Genetics: How Rare De Novo and Common Variations Affect Liability In a smaller group of autistic people, one rare change in a single gene carries much of the weight on its own. The genetic story differs from family to family, which is why testing finds a clear answer for some children and not others.
How Much of Autism Comes From Genes
Twin studies give the clearest picture. Identical twins share all their DNA, while non-identical twins share about half, like ordinary siblings. When one identical twin is autistic, the other almost always shares autism-related traits, with a match near 98 percent across twin studies.2PubMed Central. Heritability of autism spectrum disorders: a meta‐analysis of twin studies For non-identical twins the match falls to roughly 53 to 67 percent.3PubMed Central. Heritability of autism spectrum disorders: a meta‐analysis of twin studies Combining these figures, about 64 to 91 percent of the differences in autism risk between people trace to genes.4PubMed Central. Heritability of autism spectrum disorders: a meta‐analysis of twin studies
Because identical and non-identical pairs share family environment equally but not DNA, the gap in how often they match points to genes rather than to the household. The range is wide, so read it as an estimate, not a precise measurement.
“Heritable” has a narrower meaning than it sounds. It describes differences across a population, not the cause in any one child. A high figure doesn’t mean a parent definitely carries a specific autism gene, and some genetic causes are brand-new in the child, so heritable doesn’t always mean inherited. Environment still plays a part, often interacting with genetic risk.
Genes shape traits even below the level of a diagnosis. A non-autistic identical twin of an autistic child tends to show more autistic traits than a non-autistic fraternal twin does.5PubMed Central. ASD Phenotype—Genotype Associations in Concordant and Discordant Monozygotic and Dizygotic Twins Stratified by Severity of Autistic Traits Parenting style does not cause autism; the genetic evidence points firmly elsewhere.
Common Variants Carry Most of the Risk
The largest portion of genetic liability for autism comes from the combined load of common variants.6PubMed. The Yin and Yang of Autism Genetics: How Rare De Novo and Common Variations Affect Liability Each effect is so small that the first five specific locations were confirmed only after studying over 18,000 autistic people.7PubMed. Identification of common genetic risk variants for autism spectrum disorder Most people, autistic or not, carry many of these variants.
A single common variant isn’t a risk gene you can point to, so finding one in a child doesn’t single that child out. The first confirmed locations were markers in regions of DNA, and working out which gene in a region is responsible is a further step that is often unresolved.
Rare and common risk add together rather than one replacing the other.8PubMed Central. Polygenic transmission disequilibrium confirms that common and rare variation act additively to create risk for autism spectrum disorders Children with autism inherit more of this common risk from their parents than their non-autistic siblings do.9PubMed Central. Polygenic transmission disequilibrium confirms that common and rare variation act additively to create risk for autism spectrum disorders That helps explain autism in families with no history of it. Two parents can each carry modest common risk that combines in a child, neither parent need be autistic, and siblings can inherit different amounts by chance.
A polygenic score adds up a person’s many small genetic risks into one number. For autism, that score has only a weak link to diagnosis, far too weak to predict whether one child is autistic.10PubMed Central. Systematic Review and Meta-Analysis: Phenotypic Correlates of the Autism Polygenic Score The same score is linked to social behavior, motor skills, and depression.11PubMed Central. Systematic Review and Meta-Analysis: Phenotypic Correlates of the Autism Polygenic Score Scores built mostly from European-ancestry data lose roughly 40 to 50 percent of their already modest accuracy for people of other ancestries.12PubMed. The impact of diverse ancestry on polygenic risk score-based prediction models for psychiatric and neurodevelopmental disorders Be wary of any company selling a polygenic autism risk test for an embryo or child.
Rare Gene Changes With a Large Effect
Some rare changes are new, appearing in the child but not in either parent’s DNA.13National Institute of Mental Health. Spontaneous Gene Glitches Linked to Autism Risk with Older Dads In families with only one autistic child, new coding mutations and DNA segment changes together account for about 30 percent of diagnoses.14PubMed Central. The contribution of de novo coding mutations to autism spectrum disorder In girls in such families the share rises to about 45 percent.15PubMed Central. The contribution of de novo coding mutations to autism spectrum disorder Rare inherited variants add to this, and large studies now include over 42,000 autistic people to find more of these genes.16PubMed Central. Integrating de novo and inherited variants in 42,607 autism cases identifies mutations in new moderate-risk genes
For a family, a new change is not something either parent did or could have prevented. A new change usually means a lower chance of the same change in a later child than an inherited one would. To check whether a change is new, a lab compares the child’s DNA with both parents’ DNA, which is why testing often asks for parent samples. A change found in the child but absent in both parents is classed as new, while one found in a parent is inherited. An inherited rare change can come from a parent with few or no noticeable traits, because the same change doesn’t affect everyone equally.
Some rare changes involve a stretch of DNA that is missing or duplicated. A missing or extra piece on chromosome 16, called 16p11.2, is one of the strongest single genetic risk factors for autism.17PubMed Central. 16p11.2 Copy Number Variations and Neurodevelopmental Disorders Even so, only about 16 to 26 percent of people missing it and 20 to 34 percent with an extra copy are diagnosed autistic.18PubMed Central. 16p11.2 Copy Number Variations and Neurodevelopmental Disorders It is also found in only about 1 in 132 autistic people tested.19PubMed. Copy number variation in the dosage-sensitive 16p11.2 interval accounts for only a small proportion of autism incidence: a systematic review and meta-analysis Background common risk and sex shift how strongly a change shows, and some carriers have developmental delay or epilepsy without autism.
Some new mutations arise after conception and are present in only some of a person’s cells.20PubMed Central. Somatic Mosaicism and Autism Spectrum Disorder These may explain about 3 to 5 percent of autism in families with one autistic child.21PubMed Central. Somatic Mosaicism and Autism Spectrum Disorder
Well-Known Autism Genes and What They Do
Several syndromes have a single known genetic cause. Fragile X syndrome, caused by silencing of the FMR1 gene, is the most common single-gene cause of autism.22PubMed Central. Shared functional defect in IP3R-mediated calcium signaling in diverse monogenic autism syndromes Tuberous sclerosis, from changes in TSC1 or TSC2, brings autism-like behaviors along with seizures and skin and brain growths.23PubMed Central. Shared functional defect in IP3R-mediated calcium signaling in diverse monogenic autism syndromes Rett syndrome, neurofibromatosis type 1, Noonan, Costello, and Cowden syndromes are other single-gene conditions linked to autism.24Russian Journal of Genetics: Applied Research. Molecular mechanisms of autism as a form of synaptic dysfunction
In these syndromes, autism is one part of a wider pattern, so doctors often notice other signs first, such as seizures, skin findings, or distinctive growth patterns. Most are identified by a specific genetic test rather than by the autism evaluation. Not everyone with one of these syndromes is autistic, so the syndrome diagnosis doesn’t settle whether a child meets autism criteria.
Other genes build connections between brain cells. SHANK3, neuroligins, and neurexins make proteins that hold together the junctions where brain cells signal to each other.25PubMed Central. Autism-associated mutations in ProSAP2/Shank3 impair synaptic transmission and neurexin-neuroligin-mediated transsynaptic signaling Autism-linked SHANK3 changes disrupt signaling on both sides of these junctions.26PubMed Central. Autism-associated mutations in ProSAP2/Shank3 impair synaptic transmission and neurexin-neuroligin-mediated transsynaptic signaling Still, SHANK3 changes appear in under 1 percent of autistic people, about 0.75 percent in one group.27PubMed Central. Contribution of SHANK3 mutations to autism spectrum disorder
Some genes control other genes. CHD8 helps switch other genes on and off during brain development, and many of the genes it controls are themselves autism genes.28PubMed Central. The autism-associated chromatin modifier CHD8 regulates other autism risk genes during human neurodevelopment People with CHD8 changes often have a larger head and brain size.29PubMed Central. Germline Chd8 haploinsufficiency alters brain development in mouse
SCN2A and SCN8A make channels that let nerve cells fire.30PubMed Central. Progress in understanding and treating SCN2A-mediated disorders Changes that weaken these channels are linked to autism or intellectual disability, while changes that overactivate them tend to cause infant seizures.31PubMed Central. Progress in understanding and treating SCN2A-mediated disorders32PubMed Central. Sodium channelopathies in neurodevelopmental disorders
Day to day, a named gene can point to linked health issues to watch, such as seizures, and families can connect with others who share the same gene change. Outcomes still vary widely even within one gene.
Why Boys Are Diagnosed More Often
Autism is diagnosed three to four times more often in boys than girls.33PubMed Central. The female protective effect against autism spectrum disorder Girls appear to need a heavier genetic load before autism shows.34PubMed. The inactive X chromosome as a female protector in autism and beyond Autistic girls carry more damaging new mutations on average than autistic boys.35PubMed Central. Genetic evidence of gender difference in autism spectrum disorder supports the female-protective effect Siblings of autistic girls are more likely to be autistic than siblings of autistic boys,36PubMed Central. The female protective effect against autism spectrum disorder and mothers of autistic children carry slightly more common autism risk than fathers.37PubMed Central. The female protective effect against autism spectrum disorder One idea is that girls’ second X chromosome adds extra activity of some genes that buffers risk.38PubMed. The inactive X chromosome as a female protector in autism and beyond Most brothers of an autistic girl are still not autistic.
For families, an autistic daughter may signal more overall risk worth discussing with a counselor. Under-diagnosis of girls also contributes to the gap, apart from genetics, and the second-X idea is a proposal still being tested.
Genetic Testing for Autism
Two main tests are used. Chromosomal microarray looks for missing or extra stretches of DNA, so it can catch something like a 16p11.2 change, but it cannot read the letters of single genes. Exome sequencing reads the instructions of nearly all genes to spot single changes, and in many labs it can also detect some larger missing or extra pieces, though methods vary. Neither test reads the DNA between genes, so some causes can be missed even by the broader one.
In one clinic, exome sequencing found a likely cause in 44 percent of children tested, against 7 percent with the microarray test.39PubMed Central. Yield of Genetic Testing in Children with Autism Spectrum Disorder – A Single-Center Experience Another group found genetic variants in about 35 percent, in line with earlier reports.40PubMed. Multilayered genetic dissection of autism: insights from whole-exome sequencing, molecular karyotyping, and cytogenetic analyses in a small Turkish cohort Both were small groups at single centers, and they differ partly because of who is referred for testing. Children with additional features, such as seizures or developmental delay, are more likely to have an identifiable cause than children with autism alone. Treat the numbers as a general range, not a forecast for your child.
To arrange testing, ask the child’s pediatrician or developmental doctor for a referral to clinical genetics. Expect a blood or saliva sample, sometimes from parents too. Results can take weeks, and insurance coverage varies.
A positive result may explain the cause and guide health checks. A negative result doesn’t rule out genetic causes, since common variants are not reported. Some results come back as “uncertain” and may be reclassified later. A genetic counselor can estimate the chance for future children.
Father’s Age and New Mutations
Sperm is made continuously throughout life, and each cell division can introduce new mistakes.41Nature. Fathers bequeath more mutations as they age Women, by contrast, are born with all their eggs,42Nature. Fathers bequeath more mutations as they age which go through far fewer divisions. New mutations in children therefore come about four times more often from the father, and more so as he ages.43National Institute of Mental Health. Spontaneous Gene Glitches Linked to Autism Risk with Older Dads The chance a child carries a harmful new mutation that could lead to autism rises with the father’s age.44PubMed Central. Rate of de novo mutations, father’s age, and disease risk The rise is gradual, and most children of older fathers are not autistic.
Can Knowing the Gene Lead to Treatment
Conditions caused by loss of one gene’s function, such as Rett syndrome from MECP2, could in theory be treated with gene therapy.45PubMed Central. Genes, Circuits, and Precision Therapies for Autism and Related Neurodevelopmental Disorders Gene therapy tries to supply a working copy of a gene that is missing or broken. Gene-silencing drugs do the opposite and turn down a gene that is overactive, which fits conditions caused by an extra copy. In mice with an extra MECP2 copy, a drug that lowers the gene’s activity reversed symptoms for about 10 weeks.46PubMed Central. Autism spectrum disorder: prospects for treatment using gene therapy Similar gene-silencing drugs already treat spinal muscular atrophy.47PubMed Central. Autism spectrum disorder: prospects for treatment using gene therapy These approaches suit specific single-gene conditions, not autism in general.
For families now, a known gene change may qualify a child for clinical trials, and gene-based groups track research and trial openings. Current support still relies on therapies and services, not gene treatments.