Not a page about autism, and not a page about schizophrenia — a page about the genetic overlap between the two. Seventeen variants, sixteen of which sit within one crowded region on chromosome 6 — the same MHC/HLA region already known as schizophrenia's single strongest signal — which likely means one real association is being counted many times over, not seventeen separate findings.
Autism spectrum disorder (ASD) and schizophrenia are separately diagnosed, clinically distinct conditions. They also share more genetic risk than chance would predict — a finding this page is specifically about, rather than about either condition on its own. This site does not yet have a standalone autism page; this one exists because of what a combined analysis found, not as a general ASD reference.
The Psychiatric Genomics Consortium ran a large ASD-only meta-analysis first — a discovery sample of 7,387 cases and 8,567 controls, with two further replication sets adding thousands more of each. That analysis found one genome-wide significant ASD-specific locus, at 10q24.32, and — separately — a statistically significant genetic correlation between ASD and schizophrenia (Rg = 0.23, P = 9×10⁻⁶): a real, quantified answer to how much shared genetic architecture exists between the two, not just an impression that they sometimes co-occur.
Having established that overlap, the researchers combined the ASD data with the Psychiatric Genomics Consortium's own schizophrenia GWAS and searched for loci associated with the two conditions together. That combined analysis found 12 new genome-wide significant loci.
This site's catalogue carries seventeen variants under this combined trait. Checked one by one against their actual chromosome position — not just their assigned gene name — sixteen of the seventeen sit within about 6 million bases of each other on chromosome 6 (roughly 26.3 to 32.4 Mb), squarely inside the extended MHC/HLA region. That region is already schizophrenia's own single best-established genetic signal, and it is also famous across human genetics for exceptionally dense linkage disequilibrium — long stretches of DNA inherited together, which means one true causal signal there routinely gets tagged by many nearby genes at once. Sixteen "different" genes here — including several named for the HLA system directly (HLA-DRB1, HLA-C) and others nearby (BTNL2, HFE, ABCF1, CCHCR1, AGER, ABT1, DDR1, GPX5 among them) — most plausibly reflect one strong regional association, most likely schizophrenia's own MHC signal reappearing in the combined analysis, rather than sixteen independent discoveries.
The seventeenth, near LINC01360 on chromosome 1, sits well outside that region — a genuinely separate signal from the rest.
The paper names other genes from the same combined search with clearer developmental biology behind them and no MHC connection — FOXP1, ATP2B2, a "neurodevelopmental hub" region on chromosome 8, and others — none of which are in this site's own catalogue yet. This page covers only the loci we can actually show a variant for, and says plainly which of those are probably one signal counted many times.
Both ASD and schizophrenia are diagnosed clinically, through developmental history and behavioural assessment for ASD and through psychiatric evaluation for schizophrenia — not by genotype. A shared genetic correlation between two conditions does not mean a genetic test can distinguish, predict, or diagnose either one. Nothing on this page is used that way in any clinical setting.
A genetic correlation of 0.23 is real and worth stating precisely rather than rounding up to "linked": it means a meaningful share of the common-variant genetic influences on the two conditions overlap, not that they are genetically the same thing, and not that most of either condition's heritability is explained by this shared component.
This site's Schizophrenia page covers that condition's own, much larger body of genetics (around 8,300 associated positions). There is no standalone autism page on this site yet — this page exists specifically for the shared-risk finding, not as a substitute for one.
What a 23andMe/AncestryDNA export or raw VCF can and can't tell you about Autism Spectrum Disorder and Schizophrenia (Shared Genetic Risk) comes down to these specific, well-studied positions — not a diagnosis.
Databases, guidelines and references
No. Autism spectrum disorder is diagnosed through developmental history and behavioural assessment; schizophrenia through psychiatric evaluation. These variants come from a genome-wide association study and are not used diagnostically for either.
No. They remain separately diagnosed, clinically distinct conditions. The finding is a genetic correlation (Rg=0.23) — a real, partial overlap in common-variant genetic influences, not genetic identity.
Because sixteen of the seventeen variants sit within one crowded region on chromosome 6 — the MHC/HLA region, already schizophrenia's strongest known signal — where dense linkage disequilibrium means many nearby genes get statistically tagged by what is most plausibly one real, underlying association. This page states that explicitly rather than presenting seventeen genes as seventeen separate discoveries.
Because that is not what it covers. This site has no standalone autism page yet; the variants here are specifically from an analysis of ASD and schizophrenia combined, and would not necessarily be genome-wide significant findings for autism studied on its own.
Free to reuse. This page's text is original writing from freely-available research, licensed CC BY 4.0 — reuse it, including commercially, with attribution to MyGeneLog. It's general research-derived information, not medical advice or a diagnosis — see Terms of Use.