Discovery of eQTL Alleles Associated with Autism Spectrum Disorder: A Case-Control Study.
Allison R Hickman, Bradley Selee, Rini Pauly and 3 others
PMID 35739433WHAT IT FOUND
Thirty variants tied to gene expression in different tissues differed between ASD cases and outside controls.
But no variants were significant when cases were compared with their unaffected siblings, so this is not a clinical test.
Key findings
01Thirty variants were significant in both comparisons with outside controls, and they aligned with 174 tissue-specific eQTLs, 34 genes and 49 tissues.
02In the secondary analysis with unaffected siblings as controls, no variants were significant after correction.
03A neural network classified ASD-affected and unaffected individuals with 90 to 100% accuracy using the 30 significant variants, compared with 30 to 70% for random or all tested variants.
STILL TO COME
How it was doneWhat they found
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
The positive discovery used external controls, but the same variants were not significant when compared with unaffected siblings of ASD-affected individuals. The study excluded ASD-affected individuals with possible environmental contributions and used only one person per family, so it does not address families with multiple affected members or ASD with those exposures. The African American classification subset had 96 individuals, so the classification result was less stable than in the larger Caucasian subsets. The classification accuracy came from selected research datasets and known group labels, not from a clinical test in an unselected population. The paper did not show that the variants cause ASD or change gene expression in a way that affects function; it reported associations with tissue-specific eQTLs. There are no therapy outcomes, motor measures, communication measures or occupational performance measures, so it does not tell therapists what to do differently.
Declared interests
The study was funded by the National Science Foundation (US). The supplied text does not report commercial sponsorship or author conflicts of interest.
The easy way to misread this
Do not read the neural network accuracy as evidence that these variants can diagnose ASD in practice. The accuracy came from selected research datasets, and the variants were not significant when compared with unaffected siblings of people with ASD.