Genetic testing in autism: what UK results can and cannot tell you

Is my child eligible for testing?

A community paediatrician may request genomic testing, but this is not a walk-in order. For R27, the National Genomic Test Directory expects discussion with clinical genetics or a GLH-approved subspecialist. You do not always need a separate clinical-genetics clinic first, and a community paediatrician who is a GLH-approved requester can order the test — that is not the same as “no need to involve clinical genetics.”

NICE Clinical Guideline CG128 (Autism spectrum disorder in under 19s: diagnosis and management) recommends a medical/paediatric assessment alongside the behavioural assessment. That assessment looks for underlying biological factors or co-occurring conditions that might need specific support or testing.

How is it inherited?

Many parents ask, “Did I pass this on?” The answer is often complex. Autism genetics is rarely as simple as eye colour.

De novo (new)

A de novo change is one that is new in the child — it was not inherited from either parent. Origin is not the same as pathogenicity: a new change may or may not cause a condition. If a de novo change is the cause, the chance of recurrence for future children is often low (though slightly higher than the general population).

Polygenic / multifactorial

Most autism is the result of hundreds of tiny, common genetic variations inherited from both parents, interacting with environmental factors. No single test can find this. It helps explain why autism runs in families without a clear pattern.

Mendelian (rare)

Classic inheritance patterns: both parents healthy carriers (about 25% chance per pregnancy); a parent with the genetic change or a milder version (about 50% chance); or X-linked conditions such as Fragile X syndrome, where boys are generally more significantly affected.

Heritability vs recurrence chance

Autism has high heritability (estimated between 60% and 90%). That is a population statistic: genetic factors explain much of the difference between people. High heritability does not mean we can predict exactly what will happen in your family.

Even when microarray or whole genome sequencing (WGS) comes back “normal” (negative), genetics is often still involved through polygenic factors. We currently cannot test that background likelihood in a clinical setting.

• Single incidence (simplex) — one autistic child, no other affected relatives: estimated recurrence chance about 10% to 20% if no specific genetic cause is found (higher than the general population ~1–2%).
• Multiple incidence (multiplex) — two or more autistic children: estimated recurrence chance about 30% to 50%.

These figures are estimates from large studies and may vary by sex of the child and other factors.

Types of variants and whole genome sequencing

When the lab analyses the sample, they generally look for three main types of genetic change. Whole genome sequencing (WGS) is now the standard NHS test used to find all of these for eligible children.

• Chromosomal aneuploidies — whole chromosomes missing or extra (for example Down syndrome or sex chromosome variations such as XXY).
• Copy number variants (CNVs) — missing (deletion) or extra (duplication) pieces of DNA within a chromosome.
• Single gene variants (SNVs) — spelling mistakes in the DNA code of a specific gene.

Unlike older tests such as microarray that only looked at parts of the DNA, WGS reads the full genetic code — about 3 billion letters — so it can find both structural changes (CNVs) and small spelling mistakes (SNVs) in one go.

Virtual panels: labs often analyse WGS data against agreed gene lists (virtual panels) for the clinical indication. A “normal” report for that panel does not mean every letter of the genome was interpreted for every possible gene.

The official NHS Genomic Medicine Service target for WGS results is 84 days. Some families wait longer, including over a year, depending on the laboratory hub and the indication. That longer wait is reported in practice; it is not a single published national standard to set against the 84-day target. If you have not heard back in a few months, ask the requesting clinician or the laboratory hub.

What results can and cannot tell you

In eligible children, a specific genetic cause is found in a minority of cases (often cited around 10–30% depending on the cohort and how strictly “cause” is defined). A negative result does not rule out a genetic contribution.

A clear pathogenic finding can change medical surveillance, sibling counselling, and access to condition-specific support. Many results are variants of uncertain significance (VUS). Those need careful explanation and are not the same as a diagnosis.

This guide is for UK families and clinicians. It is aligned with the NHS Genomic Medicine Service pathway and is not sponsored by any testing company.