This simple analysis of a unique data set shows evidence that there is assortative mating in regard to CAG length of HTT. That is, the length of CAG repeat in the maternal allele of HTT strongly correlates with the length of the CAG repeat of the paternal allele of HTT, suggesting that the male and female pair mated based on the common feature of having a similar genotype (length of CAG repeat).
One interpretation of these findings would be that variability of CAG length is manifest by variation in phenotype of brain structure and function. This notion supports a report in which measures of mitochondrial energy metabolism (ATP/ADP) directly correlated to HTT CAG repeat lengths below disease threshold . Although in that study the genetic variation was associated with metabolic phenotypic variation, brain structure/function phenotypic variation also may be associated with CAG repeat length in HTT. As is seen with other SSRs, variation in multiple types of behavior and cognitive functions have been shown to be associated with variations in repeat sizes of these genetic elements. With expansion of HTT CAG repeat length beyond 36 repeats, disease is manifested and the brain region most heavily affected is that of the basal ganglia. Basal ganglia circuits include widespread connections from and to multiple cortical regions, including frontal lobes. These frontal circuits influence numerous complex functions including cognitive and personality traits [27–32]. Potential influence of CAG repeat length on these basal ganglia frontal circuits could influence behaviors that provide the basis for assortative mating.
An alternative explanation for the current findings could be that a post-meiotic recombination between CAG repeat domains of the two HTT alleles 'equilibrate' the 2 allele sizes, thereby producing a correlation between the sizes of the HTT alleles . To distinguish whether the mechanism responsible for the reported correlation is assortative mating or post-meiotic recombination, future studies will need to analyze parental DNA along with the proband's DNA.
Racial or ethnic relationships within the sample is an important consideration in the current study since there are differences in the distribution of normal HTT allele sizes in different ethnic groups [34–38]. Furthermore, different haplotypes associated with different distribution of the normal HTT allele sizes may influence the prevalence of HD in certain regions of the world [39, 40]. Therefore, individuals choosing mates based on race or ethnicity may explain our current findings of relationships consistent with assortative mating. However, this explanation requires that a substantial number of subjects represent more than one racial or ethnic group. In the current sample, the vast majority (85%) are Caucasian with a small numbers of African Americans (n = 4) or multiracial subjects (n = 3). Moreover, if the correlation between the ranks of Allele1 and Allele2 are calculated within the 40 Caucasians, the relationship remains significant (Spearman's r = 0.408, p = 0.009). Thus, it seems unlikely that the current findings represent assortative mating based on ethnic group. Yet, recent reports of distribution of normal HTT allele sizes suggest that the prevalence of modifier genes may be different even among sub-groups of Caucasians . Again, this explanation requires multiple subjects within several discreet sub-groups within this sample of 40 Caucasians which, although possible, seems less likely. Thus, assortative mating based not on ethnic group but on some other human feature remains a viable explanation for the findings reported here. Nevertheless, given the preliminary nature of the findings, follow-up in larger samples and further exploration of the functions of the variance of normal CAG length in HTT are warranted.