Table 2

Median model sizes and the corresponding interquartile ranges (IQR) as well as the average true positives for phenotypes Q1 and Q2 for all investigated methods summarized across the 200 repetitions (first three columns)
Results Comparisons
Method Model size TP TP TP TP
Median (IQR) Method CAR COR RND
Q1
CAR 51 (53) 5.85 5.85 5.42 0.23
COR 176 (108) 8.06 8.99 8.06 0.88
NEG 1390 (118) 15.31 17.57 14.38 6.60
MCP 20 (5) 4.11 4.19 3.95 0.12
BOOST 53 (5) 5.84 5.91 5.50 0.25
LASSO 37 (31) 5.19 5.21 4.89 0.18
Q2
CAR 31 (38) 2.93 2.93 2.85 0.29
COR 1 (7) 0.38 0.21 0.38 0.00
NEG 1632 (755) 20.21 28.08 25.90 14.50
MCP 29 (5) 2.75 2.82 2.76 0.28
BOOST 59 (6) 3.92 4.34 3.82 0.59
LASSO 15 (36) 1.50 1.88 1.97 0.14

For comparison, the last three columns show the average true positives at the specified model size for CAR, COR and RND. The best performing method is shown in bold, the second best in italic.

Zuber et al.

Zuber et al. BMC Bioinformatics 2012 13:284   doi:10.1186/1471-2105-13-284

Open Data