Modeling gene interactions in polygenic prediction via geometric deep learning

(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 7.
Figure 7.

The ablation results for PRS-Net. (A) The performance comparison between PRS-Net and its multiple variations. The bar plot and error bar denote the mean and standard error, respectively. (B) The performance of PRS-Net with PPI dropout. The line plot and shaded area denote the mean and standard error, respectively. (C) Comparison results of PRS-Net with different GGI networks. (D) The prediction performance of PRS-Net with different extension lengths. (AD) Alzheimer’s disease, (MS) multiple sclerosis, (UC) ulcerative colitis, (AUROC) the area under the receiver operating characteristic curve, (PPI) protein–protein interaction.

This Article

  1. Genome Res. 35: 178-187

Preprint Server