Skip to main content

Table 3 Wallclock run times for 2-LCs (pairs) for some of the data sets listed in Table. The P-LC program is written in C and uses double precision arithmetic. "Total time" (seconds) and "Time estimator" (seconds per 106 evaluations per sample) are execution times for the software on a 1.8 GHz AMD Athlon computer running Linux in an unloaded network configuration. An average time estimate is roughly 1 second per 106 evaluations per sample.

From: Many accurate small-discriminatory feature subsets exist in microarray transcript data: biomarker discovery

Data set

Number samples

Number genes

Number pairs

Total time

Time estimator

BreastER

47

3,389

5,700,000

251

0.94

BreastBRCA

22

3,226

5,200,000

161

1.41

LungBeer

96

4,966

12,300,000

1,500

1.27

Cutaneous

38

3,613

6,500,000

260

1.05

YeohALL

248

4,169

8,700,000

1,800

0.83