"... Despite a lack of association between the 5-HTTLPR variant and ASD in a combined set of 18 TDT studies, we found the short variant to be associated with ASD in the subset of American samples and the long variant to be associated with ASD in the subset of samples from Asia. In the joint set, the variants mutually masked each otherís association with ASD..."
This is an old saw that we have pointed out over the years. Namely, if a condition is related to one of the NPA traits, then results of the study will be highly dependent on the habitancy in which the study is done.
It looks like ASD comes in two basic forms: 1) NP types, classical autistic child, 2) N types, Asperger syndrome, others.
If trait P is related to short variant, then American samples will naturally be mainly short-variant NP types, while Asian samples will be mostly long-variant N types.
Of course, if one mixes them together, then "...the variants the will mutually mask each otherís association with ASD..."
« Back to index