Genetics-driven risk predictions leveraging the Mendelian randomization framework

(Downloading may take up to 30 seconds. If the slide opens in your browser, select File -> Save As to save it.)

Click on image to view larger version.

Figure 5.
Figure 5.

Application of PRiMeR to predict 5-year AD risk. (A) Comparative performance of PRiMeR against baseline MR models using average AUC for 5-year AD predictions using follow-up labels. (B) Heatmap of the signed Formula P-value of association between voxel intensities and the AD risk predictor scores, overlayed on the MNI152 template (Miller et al. 2016; Alfaro-Almagro et al. 2018; https://www.bic.mni.mcgill.ca/ServicesAtlases/ICBM152NLin6). Areas where increased risk predictor scores correlate with significant increased (decreased) voxel intensities are highlighted in red (blue) (Bonferroni-adjusted P < 0.05). (C) Spearman's correlation coefficients between the AD risk predictor and individual MRI traits in the validation set. Results for the top 10 associated regions are displayed, with associations for all analyzed regions available in Supplemental Figure A7.

This Article

  1. Genome Res. 34: 1276-1285

Preprint Server