Class prediction with nominal gene set selected with the random varaince t-test (P=0.001):

Number of classes: 2  (SSc vs normal)

Based on 5000 random permutations,
the compound covariate predictor has p-value of 0.002
the 1-nearest neighbor classifier has p-value of < 2e-04
the 3-nearest neighbors classifier has p-value of < 2e-04
the nearest centroid classifier has p-value of < 2e-04
the support vector machines classifier has p-value of < 2e-04
the linear discriminant analysis classifier has p-value of 0.003
Note: t-values used for the compound covariate predictor were truncated at abs(t)=10 level.


Performance of classifiers during cross-validation:


Pair ID Number of genes in classifier Compound
Covariate
Predictor
Correct?
Diagonal Linear
Discriminant
Analysis
Correct?
1-Nearest
Neighbor
Correct?
3-Nearest
Neighbors
Correct?
Nearest
Centroid
Correct?
Support
Vector
Machines
Correct?
1 1 22 YES YES YES YES YES YES
2 10 18 YES YES YES YES YES YES
3 11 23 YES YES YES YES YES YES
4 12 23 YES YES YES YES YES YES
5 13 23 YES YES YES YES YES YES
6 14 15 YES YES YES YES YES YES
7 15 19 YES YES YES YES YES YES
8 16 25 YES YES YES YES YES YES
9 17 22 YES YES YES YES YES YES
10 18 24 YES YES YES YES YES YES
11 19 19 YES YES YES YES YES YES
12 2 19 YES YES YES YES YES YES
13 20 25 YES YES YES YES YES YES
14 21 29 YES YES YES YES YES YES
15 22 22 YES YES YES YES YES YES
16 23 19 YES YES YES YES YES YES
17 24 25 YES YES YES YES YES YES
18 25 23 YES YES YES YES YES YES
19 3 24 YES YES YES YES YES YES
20 4 27 NO NO YES YES YES YES
21 5 18 YES YES YES YES YES YES
22 6 18 YES YES YES YES YES YES
23 7 17 YES YES YES YES YES YES
24 8 18 YES YES YES YES YES YES
25 9 28 YES YES YES YES YES YES
Percent
correctly
classified:


96 96 100 100 100 100


Composition of classifier (26 genes significant at the 1e-04 level):

Table - Sorted by t -value:



t-value Parametric p-value % CV support Geometric mean of ratios
(class Disease /class Normal )
Qiagen oligo ID
Description GB acc UG cluster Gene symbol
1 -6.61 p < 0.000001
100 0.621 H003528_01 Decay accelerating factor for complement (CD55, Cromer blood group system) M30142 1369 DAF
2 -5.92 1e-06 100 0.548 H003827_01 Serum/glucocorticoid regulated kinase AJ000512 296323 SGK
3 -5.91 2e-06 100 0.548 H016371_01 Hypothetical protein FLJ21212 NM_024642 47099 FLJ21212
4 -5.48 5e-06 100 0.584 H009347_01 Neuronal cell adhesion molecule AB002341 7912 NRCAM
5 -4.86 3.1e-05 100 0.811 H005438_01 Homo sapiens mRNA; cDNA DKFZp434B1620 (from clone DKFZp434B1620) AL137548 43112
6 -4.86 3.1e-05 100 0.728 H004078_01 Heme-binding protein NM_015987 108675 HEBP
7 -4.85 3.2e-05 100 0.465 H001509_01 Aldo-keto reductase family 1, member C3 (3-alpha hydroxysteroid dehydrogenase, type II) D17793 78183 AKR1C3
8 -4.8 3.8e-05 100 0.708 H002860_01 Inositol polyphosphate-1-phosphatase L08488 32309 INPP1
9 -4.79 3.8e-05 100 0.62 H010655_01 Hypothetical protein FLJ20546 AK000953 279896 FLJ20546
10 -4.78 3.9e-05 100 0.768 H002574_01 Alcohol dehydrogenase 5 (class III), chi polypeptide M81118 78989 ADH5
11 -4.78 3.9e-05 100 0.517 H003858_01 Aldo-keto reductase family 1, member C2 (dihydrodiol dehydrogenase 2; bile acid binding protein; 3-a U05598 201967 AKR1C2
12 -4.72 4.6e-05 76 0.751 H004767_01 Tetraspan 3 AK001326 100090 TSPAN-3
13 -4.53 8e-05 24 0.771 H016494_01 Hypothetical protein FLJ12436 NM_024661 69485 FLJ12436
14 -4.52 8.4e-05 36 0.721 H003965_01 Cellular repressor of E1A-stimulated genes AF084523 5710 CREG
15 -4.51 8.4e-05 32 0.724 H002462_01 Receptor tyrosine kinase-like orphan receptor 1 M97675 274243 ROR1
16 -4.47 9.6e-05 32 0.689 H000959_01 Glycophorin C (Gerbich blood group) NM_002101 81994 GYPC
17 -4.46 9.8e-05 36 0.742 H008089_01 KIAA0469 gene product AB007938 7764 KIAA0469
18 4.53 8.1e-05 36 1.689 H016341_01 Platelet derived growth factor C NM_016205 43080 PDGFC
19 4.53 8.1e-05 40 1.757 H002994_01 Collagen, type XVIII, alpha 1 AF018081 78409 COL18A1
20 4.55 7.7e-05 36 1.314 H003383_01 Ras-related C3 botulinum toxin substrate 2 (rho family, small GTP binding protein Rac2) Z82188 173466 RAC2
21 4.6 6.6e-05 36 1.249 H003776_01 Aldehyde dehydrogenase 2 family (mitochondrial) X05409 195432 ALDH2
22 4.64 5.9e-05 40 1.352 H000498_01 Desmoplakin (DPI, DPII) AL031058 74316 DSP
23 4.69 5.1e-05 60 1.251 H011854_01 Heterogeneous nuclear ribonucleoprotein C (C1/C2) M16342 182447 HNRPC
24 4.71 4.8e-05 60 1.422 H006183_01 Metallothionein 1X X65607 278462 MT1X
25 6.23 p < 0.000001
100 1.453 H007688_01 6-phosphofructo-2-kinase/fructose-2,6-biphosphatase 3 AF109735 195471 PFKFB3
26 7.26 p < 0.000001
100 1.553 H002655_01 Collagen, type VII, alpha 1 (epidermolysis bullosa, dystrophic, dominant and recessive) L02870 1640 COL7A1

Table - Sorted by mean difference:



t-value Parametric p-value % CV support Geometric mean of ratios
(class Disease /class Normal )
Qiagen oligo ID
Description GB acc UG cluster Gene symbol
19 4.53 8.1e-05 40 1.757 H002994_01 Collagen, type XVIII, alpha 1 AF018081 78409 COL18A1
18 4.53 8.1e-05 36 1.689 H016341_01 Platelet derived growth factor C NM_016205 43080 PDGFC
26 7.26 p < 0.000001
100 1.553 H002655_01 Collagen, type VII, alpha 1 (epidermolysis bullosa, dystrophic, dominant and recessive) L02870 1640 COL7A1
25 6.23 p < 0.000001
100 1.453 H007688_01 6-phosphofructo-2-kinase/fructose-2,6-biphosphatase 3 AF109735 195471 PFKFB3
24 4.71 4.8e-05 60 1.422 H006183_01 Metallothionein 1X X65607 278462 MT1X
22 4.64 5.9e-05 40 1.352 H000498_01 Desmoplakin (DPI, DPII) AL031058 74316 DSP
20 4.55 7.7e-05 36 1.314 H003383_01 Ras-related C3 botulinum toxin substrate 2 (rho family, small GTP binding protein Rac2) Z82188 173466 RAC2
23 4.69 5.1e-05 60 1.251 H011854_01 Heterogeneous nuclear ribonucleoprotein C (C1/C2) M16342 182447 HNRPC
21 4.6 6.6e-05 36 1.249 H003776_01 Aldehyde dehydrogenase 2 family (mitochondrial) X05409 195432 ALDH2
5 -4.86 3.1e-05 100 0.811 H005438_01 Homo sapiens mRNA; cDNA DKFZp434B1620 (from clone DKFZp434B1620) AL137548 43112
13 -4.53 8e-05 24 0.771 H016494_01 Hypothetical protein FLJ12436 NM_024661 69485 FLJ12436
10 -4.78 3.9e-05 100 0.768 H002574_01 Alcohol dehydrogenase 5 (class III), chi polypeptide M81118 78989 ADH5
12 -4.72 4.6e-05 76 0.751 H004767_01 Tetraspan 3 AK001326 100090 TSPAN-3
17 -4.46 9.8e-05 36 0.742 H008089_01 KIAA0469 gene product AB007938 7764 KIAA0469
6 -4.86 3.1e-05 100 0.728 H004078_01 Heme-binding protein NM_015987 108675 HEBP
15 -4.51 8.4e-05 32 0.724 H002462_01 Receptor tyrosine kinase-like orphan receptor 1 M97675 274243 ROR1
14 -4.52 8.4e-05 36 0.721 H003965_01 Cellular repressor of E1A-stimulated genes AF084523 5710 CREG
8 -4.8 3.8e-05 100 0.708 H002860_01 Inositol polyphosphate-1-phosphatase L08488 32309 INPP1
16 -4.47 9.6e-05 32 0.689 H000959_01 Glycophorin C (Gerbich blood group) NM_002101 81994 GYPC
1 -6.61 p < 0.000001
100 0.621 H003528_01 Decay accelerating factor for complement (CD55, Cromer blood group system) M30142 1369 DAF
9 -4.79 3.8e-05 100 0.62 H010655_01 Hypothetical protein FLJ20546 AK000953 279896 FLJ20546
4 -5.48 5e-06 100 0.584 H009347_01 Neuronal cell adhesion molecule AB002341 7912 NRCAM
3 -5.91 2e-06 100 0.548 H016371_01 Hypothetical protein FLJ21212 NM_024642 47099 FLJ21212
2 -5.92 1e-06 100 0.548 H003827_01 Serum/glucocorticoid regulated kinase AJ000512 296323 SGK
11 -4.78 3.9e-05 100 0.517 H003858_01 Aldo-keto reductase family 1, member C2 (dihydrodiol dehydrogenase 2; bile acid binding protein; 3-a U05598 201967 AKR1C2
7 -4.85 3.2e-05 100 0.465 H001509_01 Aldo-keto reductase family 1, member C3 (3-alpha hydroxysteroid dehydrogenase, type II) D17793 78183 AKR1C3


'Observed v. Expected' table of GO classes and parent classes, in list of 26 genes shown above:

Only GO classes and parent classes with at least 5 observations in the selected subset and with an 'Observed vs. Expected' ratio of at least 2 are shown.

Biological Process

GO id GO classification Observed in
selected subset
Expected in
selected subset
Observed/
Expected
0009887 organogenesis 5 1.24 4.02
0009653 morphogenesis 5 1.57 3.19
0007275 development 7 2.37 2.95