With the analysis of five temperaments, the issue of multiple comparisons must be considered. However, as shown in our data and others’, four of the five temperaments are highly correlated (r= 0.44 to 0.68) and share a pattern of correlations with several clinical characteristics, indicating that these analyses are not entirely independent and complicating corrections for multiple comparisons (23). If we consider that the hyperthymic temperament appears to be relatively independent of the other four temperaments and correct for the analysis of two effectively independent temperament groups, two of our findings remain GW significant: hyperthymic with the FBLN1 gene and irritable with the INTS7 gene. These results will ultimately need to be validated in an independent sample.
