Skip to main content

Table 1 Results of CIA of different subsets of gene expression datasets

From: Cross-platform comparison and visualisation of gene expression data using co-inertia analysis

Number of genes in each dataset

Matchminer results

Results of Coinertia Analysis on two datasets

Ross-cDNA*

Staunton-Affymetrix**

Number of "matched" genes ±

RV coefficient

% Inertia

Correlation of ordinations

    

F1

F2

F1

F2

5643

3144

1416

0.85

40

61

0.96

0.97

 

2455

1169

0.86

40

61

0.96

0.97

 

1517

776

0.88

42

63

0.96

0.98

3748

3144

786

0.86

30

49

0.96

0.97

 

2455

625

0.87

31

50

0.97

0.97

 

1517

388

0.86

32

51

0.97

0.97

1415

3144

-

0.83

38

62

0.95

0.96

 

2455

-

0.85

38

62

0.95

0.97

 

1517

-

0.86

40

64

0.95

0.97

1375

3144

433

0.83

38

62

0.95

0.96

 

2455

370

0.84

37

62

0.95

0.97

 

1517

269

0.86

40

64

0.95

0.97

  1. Gene expression data subsets from *spotted cDNA [16] and **Affymetrix [15] were subjected to CIA, where COA was performed on the Affymetrix dataset, and row weighted COA on spotted cDNA array dataset. Results of the co-inertia analysis show the RV co-efficient, accumulated inertia (% of total sum of eigenvalues of co-inertia analysis), and correlation between the coordinates on first pair (F1) and second pair (F2) of axes. ± Probes (sequence spots on each array) were matched using MatchMiner [22]. The 1415 cDNA subset contained the 1375 cDNA geneset and 40 extra genes for which no image identifier was given, thus matchminer counts for these 40 extra genes could not be determined, but the number of probes matched should be similar to the 1375 cDNA gene set results.