## Table 4 |
|||||

Simulation Example 4 |
|||||

False patterns |
|||||

Methods |
Main effects average^{∗} |
X_{100}X_{102} |
X_{300}X_{302} |
X_{50}X_{250} |
(Variables) |

pLPS | 96 (100) | 98 (100,100) | 98 (98,100) | 99 (100,100) | 320 |

RF | NA (99) | NA (96,100) | NA (87,72) | NA (94,89) | (1268) |

SPLR | 100(100) | 97 (100,100) | 82 (100,100) | 97 (100,100) | 1017 |

*n *= 700 and *p *= 200, with positive correlations among neighbors in the first 200 variables and negative
correlations among neighbors in the next 200 variables. Tabulated numbers show the
number of tests (out of 100) in which the pattern was detected by each algorithm.
The number outside the parentheses is the number of times the given pattern was selected;
the numbers inside the parentheses shows how many times the variables in the pattern
are detected in the model, as a main effect or in some interaction. The final column
shows the total number of times (in 100 tests) that the algorithms selected patterns
(variables for RF) that are not in the true model.

^{∗}The average of *X*_{1}, *X*_{3}, *X*_{10}, *X*_{201}, *X*_{210}, *X*_{220} and *X*_{230}.

Shi * et al.*

Shi * et al.* *BMC Bioinformatics* 2012 **13**:98 doi:10.1186/1471-2105-13-98