Skip to main content

Table 1 Summary of the PCA models (Nsamples × Kvariables) obtained from the four control subsets of the single Rat experiment

From: Exploring the use of internal and externalcontrols for assessing microarray technical performance

Data subset

(control-type)

N samples

(hybridizations)

K variables

(probesets)

No. of PCs

% Variance

Cumulative

RMSEC a

RMSECV b

hybridization

137

18

4

94.9

0.087

0.680

polyA+

137

27

4

97.3

0.117

1.48

RNAd

137

12

3

94.6

0.143

1.05

endogenous

137

100

7

70.8

0.125

1.52

  1. aRoot mean square error of calibration (model)
  2. bRoot mean square error of cross-validation employing the venetian blinds approach