Table 1.

Runtime (s) for growing a RF and computing a VIM on two gene expression and one GWAS dataset

Importance	Gene expr.		GWAS
	Leukemia	Breast cancer	Alzheimer
Impurity	3.7	4.2	439.3
Permutation	46.8	38.4	21 511.5
Holdout	112.5	89.1	48 772.5
AIR	3.9	5.5	535.6

Note: The RF is grown with 5000 trees using one computing thread.