Modeling the cumulative genetic risk for multiple sclerosis from genome-wide association data

Table 7 The percentage of variance (R²) explained by predictors in the regression model

	Center	Gender	DRB115:01*	12 genes^a	350 genes^b
The discovery dataset (n = 8,844)	15%	4%	7%	10%	57%
The validation dataset (n = 3,606)	NA	2%	9%	11% (AUC^c = 0.68)	27% (AUC^c = 0.769)

^aThe 12-gene set includes HLA-DRB1 and 11 additional validated genes. ^bThe 350-gene set includes HLA-DRB1 and 349 additional genes identified in the genetic profile. ^cAUC, area under curve from ROC analysis results.

ISSN: 1756-994X