Table 1

Number of genes called significant at 10% false discovery rate on isogenic biologic replicates

Statistic

Embryoid blast

Myeloid

4 hours

24 hours

4 hours

24 hours

Standard t (scheme 1)

32

8,038

7,154

331

Cyber t (scheme 2)

1,288

9,769

10,349

4,464

Scheme 4

90

87

268

38

SAM

4,954

14,239

16,644

9,392

Each entry in this table is the number of significant genes when comparing treatment (DMSO + drug) with control (just DMSO) for a cell type (embryoid blast or myeloid) and a time point (4 or 24 hours). All data were subject to quantile-quantile normalization and summarized using RMA. The Benjamini and Hochberg algorithm was used on P values generated by the standard t test, cyber t, and scheme 4 to calculate false discovery rate. The values for SAM were generated by the TIGR Multiple Array Viewer (see Materials and methods) with 100 permutations and the default parameters. DMSO, dimethyl sulfoxide; RMA, Robust Multichip Average; SAM, significance analysis of microarrays.

Fodor et al. Genome Biology 2007 8:R69   doi:10.1186/gb-2007-8-5-r69