Autism has a strong genetic basis, although the genetics of autism are complex and it is unclear whether ASD is explained more by rare mutations with major effects, or by rare multi-gene interactions of common genetic variants. Complexity arises due to interactions among multiple genes, the environment, and epigenetic factors which do not change DNA sequencing but are heritable and influence gene expression. Many genes have been associated with autism through sequencing the genomes of affected individuals and their parents. However, most of the mutations that increase autism risk have not been identified. Typically, autism cannot be traced to a Mendelian (single-gene) mutation or to a single chromosome abnormality, and none of the genetic syndromes associated with ASD have been shown to selectively cause ASD. Numerous candidate genes have been located, with only small effects attributable to any particular gene. Most loci individually explain less than 1% of cases of autism. As of 2018, it appeared that between 74% and 93% of ASD risk is heritable. After an older child is diagnosed with ASD, 7–20% of subsequent children are likely to be as well. If parents have one autistic child, they have a 2% to 8% chance of having a second child who is also autistic. If the autistic child is an identical twin the other will be affected 36 to 95 percent of the time. If they are fraternal twins the other will only be affected up to 31 percent of the time. The large number of autistic individuals with unaffected family members may result from spontaneous structural variation, such as deletions, duplications or inversions in genetic material during meiosis. Hence, a substantial fraction of autism cases may be traceable to genetic causes that are highly heritable but not inherited: that is, the mutation that causes the autism is not present in the parental genome.

As of 2018, understanding of genetic risk factors had shifted from a focus on a few alleles to an understanding that genetic involvement in ASD is probably diffuse, depending on a large number of variants, some of which are common and have a small effect, and some of which are rare and have a large effect. The most common gene disrupted with large effect rare variants appeared to be CHD8, but less than 0.5% of autistic people have such a mutation. The gene CHD8 encodes the protein chromodomain helicase DNA binding protein 8, which is a chromatin regulator enzyme that is essential during fetal development, CHD8 is an ATP dependent enzyme. The protein contains an Snf2 helicase domain that is responsible for the hydrolysis of ATP to ADP. CHD8 encodes for a DNA helicase that function as a transcription repressor by remodeling chromatin structure by altering the position of nucleosomes. CHD8 negatively regulates Wnt signaling. Wnt signaling is important in the vertebrate early development and morphogenesis. It is believed that CHD8 also recruits the linker histone H1 and causes the repression of β-catenin and p53 target genes. The importance of CHD8 can be observed in studies where CHD8-knockout mice died after 5.5 embryonic days because of widespread p53 induced apoptosis. Some studies have determined the role of CHD8 in autism spectrum disorder (ASD). CHD8 expression significantly increases during human mid-fetal development. The chromatin remodeling activity and its interaction with transcriptional regulators have shown to play an important role in ASD aetiology. The developing mammalian brain has a conserved CHD8 target regions that are associated with ASD risk genes. The knockdown of CHD8 in human neural stem cells results in dysregulation of ASD risk genes that are targeted by CHD8. Recently CD8 has been associated to the regulation of long non-coding RNAs (lncRNAs), and the regulation of X chromosome inactivation (XCI) initiation, via regulation of Xist long non-coding RNA,[ambiguous] the master regulator of XCI,[ambiguous] though competitive binding to Xist regulatory regions.

Some ASD is associated with clearly genetic conditions, like fragile X syndrome; however, only around 2% of autistic people have fragile X. Hypotheses from evolutionary psychiatry suggest that these genes persist because they are linked to human inventiveness, intelligence or systemising.

Current research suggests that genes that increase susceptibility to ASD are ones that control protein synthesis in neuronal cells in response to cell needs, activity and adhesion of neuronal cells, synapse formation and remodeling, and excitatory to inhibitory neurotransmitter balance. Therefore, despite up to 1000 different genes thought to contribute to increased risk of ASD, all of them eventually affect normal neural development and connectivity between different functional areas of the brain in a similar manner that is characteristic of an ASD brain. Some of these genes are known to modulate production of the GABA neurotransmitter which is the main inhibitory neurotransmitter in the nervous system. These GABA-related genes are under-expressed in an ASD brain. On the other hand, genes controlling expression of glial and immune cells in the brain e.g. astrocytes and microglia, respectively, are over-expressed which correlates with increased number of glial and immune cells found in postmortem ASD brains. Some genes under investigation in ASD pathophysiology are those that affect the mTOR signaling pathway which supports cell growth and survival.

All these genetic variants contribute to the development of the autistic spectrum; however, it cannot be guaranteed that they are determinants for the development.

ASD may be under-diagnosed in women and girls due to an assumption that it is primarily a male condition, but genetic phenomena such as imprinting and X linkage have the ability to raise the frequency and severity of conditions in males, and theories have been put forward for a genetic reason why males are diagnosed more often, such as the imprinted brain hypothesis and the extreme male brain theory.