Pathway-Based Kernel Boosting for the Analysis of Genome-Wide Association Studies

Figure 7

Relative frequency of datasets in which a pathway was selected using (a) kernel boosting (, ) and (b) LKMT (, ) for sample sizes of 1000 individuals. Effect strength was set to relative risks of 1.1 per allele. Pathways including effect genes are labeled in bold; numbers in brackets denote the count of included influential genes within the pathway.

(a)

(b)