Fig. 4
From: Non-coding variability at the APOE locus contributes to the Alzheimer’s risk
Functional implications of PVRL2 and APOC1 haplotypes in an APOE-ε4–independent manner. a–e. Associations between PVRL2 minor haplotype alpha, and cognitive performance and biomarker expression in an APOE-ε4-independent manner. a, b Associations between PVRL2 alpha haplotype dosage with (a) cognitive performance indicated by total ECog score (scored between 0−4; higher scores represent more severe disability in functioning) reported by study partners (n = 527, T = 3.71, ***p < 0.001, Beta = 0.25) and (b) memory performance indicated by ECog memory score reported by study partners (n = 527, T = 3.60, ***p < 0.001, Beta = 0.29). c Association between PVRL2 alpha haplotype with hippocampal volume (n = 1,121, T = −2.31, *p < 0.05, Beta = −165.60 [mm2]). d, e Associations between PVRL2 alpha haplotype with (d) total Aβ1–42 in plasma (n = 226, T = −3.098, **p < 0.01, Beta = −4.113 [pg/mL]) and (e) ICAM-1 in cerebrospinal fluid; n = 298, T = −3.361, ***p < 0.001, Beta = −0.199 [log ng/mL]). Individuals harboring two copies of haplotypes were not included due to the small samples size. f, g Association between APOC1 gamma haplotype with levels of (f) plasma free Aβ1–40 (n = 226, T = −4.823, ***p < 0.001, Beta = −40.231 [pg/mL]) and (g) plasma MCP3 (CCL7) (n = 537, T = −3.665, ***p < 0.001, Beta = −0.229 [log ng/mL]). Aβ amyloid-beta, ECog everyday cognition. Data are presented in box plots, with boxes extending from the 25th to 75th percentiles and whiskers specifying the 10th and 90th percentiles; the line in the middle of the box denotes the median
