“…The sensitivity analyses were performed after removing the aforementioned specific studies, such as studies with NOS scores < 7 [18,46], studies being not in HWE [15,40,69,74,75], studies with zero in frequency for genotype distribution [59,63], studies with all samples from neuropathological diagnosis [46], or studies with samples from the mixed neuropathological and clinical diagnoses [28,34,50,53]. The final results were not altered in all comparisons except for female group, in which, the analysis demonstrated a significant association of II homozygote and V allele with AD risk after excluding studies with samples from mixed neuropathological and clinical diagnosis [50] (For II: OR, 95%CI: 0.43, 0.26-0.73, pFDR = 0.003 and for V: OR, 95%CI: 2.15, 1.38-3.35, pFDR = 0.0024) (Supplementary Figures 1 and 2).…”