We observed a strong positive correlation between the pBLS value and the JIL-1 binding intensity, though, after correcting by multiple testing, only remains statistically significant for the coding regions (Spearman's rank correlation coefficient: ρ = 0.617; P = 2e−4; fig. 4 B).
In summary, in the present study, because the risk genotype is rare, some analytic approaches including the goodness-of-fit test might not have sufficient power in detecting associations; how conservative the correction for multiple testing should be remains disputed; and the function of the risk marker remains unclear.
While the idea to test a group of genes simultaneously instead of multiple testing for each single gene remains the same in Chuang's work, we would like to put attention on how the groups were defined.
This association remains significant even after correction for multiple testing (P = 0.036).
The significant comparison remains significant after Bonferroni correction for multiple testing.
Correction for multiple testing was not applied for g.66493737C>T; however, the association remains stronger (PBonf. = 4.18 × 10-6 ) applicationcatiof of a correction factor.
Although the results are in line with some previous findings in the literature, none of the associations remains statistically significant after correction for multiple testing.
Further, the association of the T-T-C haplotype with the VEGF levels was stronger compared to that of the C-C-T haplotype and it remains statistically significant also after correction for multiple testing (Table 6).
The reason for this finding remains largely unexplained, but could be related to multiple testing.
This finding remains significant after a conservative Bonferroni correction for multiple testing (P value = 0.0085), and the HR is not significantly attenuated after adjusting for stage, grade, and treatment.
