Comparison of autism domains across thirty rare variant genotypes
Rare gene mutations hike autism risk, yet behavior profiles stay mostly alike—so assess the individual, not the gene.
01Research in Context
What this study did
The team looked at thirty rare gene changes tied to autism. They asked whether each gene gives its own special symptom picture. Kids and adults with these mutations filled out the same autism tests.
The study grouped people by gene type and compared their scores side-by-side.
What they found
Every rare gene raised autism risk about forty-three times compared with typical controls. Some groups scored lower than others, yet most symptoms looked alike across genes.
In short, risk goes up with any rare variant, but the day-to-day profile stays surprisingly similar.
How this fits with other research
Chezan et al. (2019) also saw little clinical difference between kids with and without harmful chromosome chunks. Both papers warn that gene tests alone can’t predict how a child will behave.
Ajmone et al. (2022) drilled into one single-gene syndrome and found clear links between mutation type and outcome. That fine detail seems to fade when you zoom out to thirty genes at once.
Qiao et al. (2025) used brain scans, not DNA, to split autism into two stable subtypes. Their imaging groups line up better with symptom patterns than the gene groups in NMStagnone et al. (2025).
Why it matters
For BCBAs the message is simple: order the genetic report if you want, then set it aside. Run full VB-MAPP, AFLS, and sensory checks on every learner. Write goals from the child in front of you, not from the gene name on paper.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Score full adaptive and social batteries on your next learner with a rare variant; skip assumptions based on the gene alone.
02At a glance
03Original abstract
<h4>Background</h4>A number of Neurodevelopmental risk Copy Number Variants (ND-CNVs) and Single Gene Variants (SGVs) are strongly linked to elevated likelihood of autism. However, few studies have examined the impact on autism phenotypes across a wide range of rare variant genotypes.<h4>Methods</h4>This study compared Social Communication Questionnaire (SCQ) scores (total and subdomains: social, communication, repetitive behaviour) in 1314 young people with one of thirty rare variant genotypes (15 ND-CNVs; n = 1005, 9.2 ± 3.5 years and 15 SGVs; n = 309, 8.3 ± 4.0 years). Comparisons were also conducted with young people without known genetic conditions (controls; n = 460, 10.6 ± 3.4 years) and with idiopathic autism (n = 480, 8.6 ± 3.2 years).<h4>Findings</h4>The prevalence of indicative autism (SCQ ≥ 22) was higher in those with a rare variant genotype compared to controls (32% vs 2%; OR = 43.1, CI = 6.6-282.2, p < 0.001) and in those with SGVs compared to ND-CNVs (53% vs 25%; OR = 4.00, CI = 2.2-7.3, p = 0.002). The prevalence of indicative autism varied considerably across the 30 rare variant genotypes (range 10-85%). SGVs were associated with greater impairment in total, social, communication and repetitive behaviour subdomains than ND-CNVs. However, genotype explained limited variation in these scores (η<sup>2</sup> between 11.8 and 21.4%), indicating more convergence than divergence in autism phenotype across rare variant genotypes. Comparisons with young people with idiopathic autism indicated no differences compared to those with ND-CNVs, whereas those with SGVs showed greater communication and less repetitive behaviour.<h4>Interpretation</h4>The likelihood of autism was higher across all rare variant genotypes, with individuals with SGVs showing higher prevalence and greater impairment compared to those with ND-CNVs. Despite subdomain-specific patterns, there was no strong evidence for specific genotype-phenotype associations. This suggests that rare variant genotypes alone may have limited predictive value for autism phenotypes and that other factors like polygenic risk and the environment are likely to play a role. Further research is needed in order to understand these influences, improve risk prediction and inform genetic counselling and interventions.<h4>Funding</h4>This work was funded by the Tackling Multimorbidity at Scale Strategic Priorities Fund programme (MR/W014416/1) (van den Bree) delivered by the Medical Research Council and the National Institute for Health Research in partnership with the Economic and Social Research Council and in collaboration with the Engineering and Physical Sciences Research Council. NIMH U01 MH119738-01 (van den Bree), IMAGINE study (Medical Research Council UK: MR/T033045/1; MR/N022572/1; and MR/L011166/1) (van den Bree) and Medical Research Council UK Centre Grant (MR/L010305/1) (Owen). SJRAC is funded by a Medical Research Foundation Fellowship (MRF-058-0015-F-CHAW). We would also like to acknowledge NIH 1R01MH110701-01A1 (PI Mulle), U01MH119736 (CEB), R21MH116473 (CEB), and R01MH085953 (CEB), and the Simons Foundation (SFARI Explorer Award to CEB).
, 2025 · doi:10.1016/j.ebiom.2024.105521