Chromosome 18 markers: Linked or not linked to bipolar affective disorders in the Old Order Amish? A reply to Gershon et al

We appreciate the careful review of our paper by Gershon et al., examining linkage of bipolar affective disorder (BAD) to markers on chromosome 18. These authors have raised several issues concerning our article and specifically challenge our conclusion concerning the presence or absence of a major susceptibility locus for BAD on chromosome 18 in the Old Order Amish sample. Gershon et al. state that we have interpreted our data as a "failure to replicate" a previous positive study reported by Berrettini et al. in the right extension of Old Order Amish pedigree 110. While we did not provide definitive data supporting the presence of a disease locus on chromosome 18 in the Amish, we were careful not to conclude that our findings constituted a "failure to replicate." On the contrary, we stated that the results in the right extension could represent a locus conferring risk for affective disorder in this population. Gershon et al. also state that we implied that the false-positive rate was too high. Our simulation data suggest that the right extension of Old Order Amish pedigree 110 was more likely to yield false-positive results when compared with the other Amish pedigrees, but we did not state that the false-positive rate was too high. We simply stated that it was significantly higher in the right extension than in the other Amish pedigrees. It was our conclusion that "these [Berrettini's] findings could not be extended to the larger sample of Amish kindreds". This statement was based on a P< .001 criterion. It is true that, as Gershon et al. point out, our results did not entirely contradict Berrettini's findings. We reported that there was sharing of alleles at locus D18S40 (P = .0048) in BP I individuals for all families. We suggested that the linkage findings in the right extension could reflect the presence of two or more loci in this population that confer risk for BAD: one that occurred primarily in the right extension (the chromosome 18 locus) and others that did not.