Skip to main content

Table 1 Summary of comparing the duplicate tumour samples as a repeated dataset (A and B) to assess the reproducibility of gene-lists.

From: Correcting for intra-experiment variation in Illumina BeadChip data is necessary to generate robust gene-expression profiles

   A B A & B overlap consensus (%)
Limma QN 192 30 23 11.6
  MC 225 222 59 15.2
  CB 260 211 188 66.4
SAM QN 214 40 30 13.4
  MC 240 238 65 15.7
  CB 265 218 193 66.6
Limma + UHRR QN 205 31 24 11.3
  MC 8 92 7 7.5
  CB 144 119 112 74.2
SAM + UHRR QN 224 42 32 13.7
  MC 17 100 12 11.4
  CB 149 125 117 74.5
  1. Differentially expressed genes were identified using Limma and SAM as described in the text with quantile-normalisation (QN), mean-centring (MC), and ComBat (CB). The UHRR was used as an inter-batch calibrator.