Page 236 - Special Topic Session (STS) - Volume 4

STS582 Mariza de A.

OR (95% CI) P

Association between genetically predicted BMI and
VTE using 95 SNPs from multi-ethnic BMI meta-
analysis 5.8 × 10
−6
1.59 (1.30-1.93)
Association between genetically predicted BMI and
BMI Study VTE using 75 SNPs from European ancestry BMI
meta-analysis −5
1.58 (1.28-1.95) 2.0 × 10
Association between genetically predicted BMI (95
SNPs) and VTE based on MR Egger regression
1.90 (1.17-3.08) 0.01

4. Discussion and Conclusion
The advantages of MR is that one can add more than one mediator in the
analysis to assess the IV. There are several papers recently that use the
principle of MR to learn causal biological networks as well as multivariate MR.
The following MR software are available:
Bioconductor 3.8: GMRP:GWAS-based Mendelian Randomization and Path
Analyses.
R package: Mendelian Randomization ver:0.4.0 date: 2019/03/06- repository:
CR

225 | I S I W S C 2 0 1 9