Figure 1 | Scientific Reports

Figure 1

From: Exploring the contribution of lifestyle to the impact of education on the risk of cancer through Mendelian randomization analysis

Figure 1

Study design. Network Mendelian randomization (MR) was conducted to identify the mediation proportion through lifestyle factors of the effect of education on the risk of cancer. Genetic instrumental variables were selected for each exposure based on their association below the genome-wide significance threshold, p < 5 × 10–8. To obtain robust signals, we used a less stringent threshold for fruit and vegetable consuming in the UKB (p < 10–5). Potentially causal estimates were produced using Mendelian randomization inverse variance weighted (MR-IVW) method as our main approach. MR sensitivity analyses were also conducted (MR-Egger, MR-weighted median, MR-PRESSO) to assess the robustness of the results. The simultaneous effects of each lifestyle factor on cancer were estimated using multivariable Mendelian randomization (MVMR-IVW). Mediation percentage through a lifestyle factor was obtained by dividing the indirect effect over the total effect.

Back to article page