Correlation between copy number variations of arm-level result and molecular subtypes
Esophageal Carcinoma (Primary solid tumor)
23 September 2013  |  analyses__2013_09_23
Maintainer Information
Citation Information
doi:10.7908/C1HX1B0J
Maintained by TCGA GDAC Team (Broad Institute/MD Anderson Cancer Center/Harvard Medical School)
Cite as Broad Institute TCGA Genome Data Analysis Center (2013): Correlation between copy number variations of arm-level result and molecular subtypes. Broad Institute of MIT and Harvard. doi:10.7908/C1HX1B0J
Overview
Introduction

This pipeline computes the correlation between significant arm-level copy number variations (cnvs) and subtypes.

Summary

Testing the association between copy number variation 71 arm-level results and 2 molecular subtypes across 63 patients, 5 significant findings detected with Q value < 0.25.

  • 3q gain cnv correlated to 'CN_CNMF' and 'METHLYATION_CNMF'.

  • 12p gain cnv correlated to 'CN_CNMF'.

  • 11q loss cnv correlated to 'CN_CNMF'.

  • 17p loss cnv correlated to 'METHLYATION_CNMF'.

Results
Overview of the results

Table 1.  Get Full Table Overview of the association between significant copy number variation of 71 arm-level results and 2 molecular subtypes. Shown in the table are P values (Q values). Thresholded by Q value < 0.25, 5 significant findings detected.

Molecular
subtypes 		CN
CNMF 		METHLYATION
CNMF
nCNV (%) nWild-Type Fisher's exact test Fisher's exact test
3q gain 0 (0%) 44 0.000262
(0.037) 		3.6e-05
(0.00511)
12p gain 0 (0%) 42 0.000606
(0.0842) 		0.00639
(0.856)
11q loss 0 (0%) 47 0.000492
(0.0689) 		0.00757
(1.00)
17p loss 0 (0%) 49 0.174
(1.00) 		0.000683
(0.0943)
1p gain 0 (0%) 56 0.589
(1.00) 		0.668
(1.00)
1q gain 0 (0%) 46 0.55
(1.00) 		0.816
(1.00)
2p gain 0 (0%) 52 0.0965
(1.00) 		0.236
(1.00)
2q gain 0 (0%) 56 0.0224
(1.00) 		0.158
(1.00)
3p gain 0 (0%) 54 0.0188
(1.00) 		0.191
(1.00)
4p gain 0 (0%) 59 0.346
(1.00) 		1
(1.00)
5p gain 0 (0%) 43 0.579
(1.00) 		0.645
(1.00)
6p gain 0 (0%) 58 0.611
(1.00) 		1
(1.00)
6q gain 0 (0%) 60 0.0115
(1.00) 		1
(1.00)
7p gain 0 (0%) 35 0.175
(1.00) 		0.413
(1.00)
7q gain 0 (0%) 44 0.573
(1.00) 		0.329
(1.00)
8p gain 0 (0%) 46 0.318
(1.00) 		0.139
(1.00)
8q gain 0 (0%) 37 0.898
(1.00) 		0.538
(1.00)
9p gain 0 (0%) 60 0.108
(1.00) 		0.308
(1.00)
9q gain 0 (0%) 55 0.243
(1.00) 		0.379
(1.00)
10p gain 0 (0%) 57 1
(1.00) 		0.391
(1.00)
10q gain 0 (0%) 60 0.243
(1.00) 		1
(1.00)
11p gain 0 (0%) 60 0.0115
(1.00) 		0.591
(1.00)
11q gain 0 (0%) 60 0.0115
(1.00) 		0.792
(1.00)
12q gain 0 (0%) 54 0.0952
(1.00) 		0.191
(1.00)
13q gain 0 (0%) 55 0.382
(1.00) 		0.014
(1.00)
14q gain 0 (0%) 54 0.813
(1.00) 		0.108
(1.00)
15q gain 0 (0%) 59 0.669
(1.00) 		0.676
(1.00)
16p gain 0 (0%) 58 1
(1.00) 		0.234
(1.00)
16q gain 0 (0%) 59 0.669
(1.00) 		0.431
(1.00)
17p gain 0 (0%) 56 0.0032
(0.435) 		0.0812
(1.00)
17q gain 0 (0%) 56 0.0898
(1.00) 		0.513
(1.00)
18p gain 0 (0%) 49 0.0147
(1.00) 		0.0725
(1.00)
18q gain 0 (0%) 60 0.604
(1.00) 		0.792
(1.00)
19p gain 0 (0%) 60 0.0115
(1.00) 		0.591
(1.00)
19q gain 0 (0%) 56 0.00496
(0.67) 		0.252
(1.00)
20p gain 0 (0%) 38 0.761
(1.00) 		0.602
(1.00)
20q gain 0 (0%) 36 0.804
(1.00) 		0.342
(1.00)
21q gain 0 (0%) 60 0.108
(1.00) 		0.308
(1.00)
22q gain 0 (0%) 55 1
(1.00) 		0.379
(1.00)
Xq gain 0 (0%) 53 0.221
(1.00) 		0.0289
(1.00)
3p loss 0 (0%) 37 0.581
(1.00) 		0.714
(1.00)
3q loss 0 (0%) 59 0.543
(1.00) 		0.676
(1.00)
4p loss 0 (0%) 40 0.0471
(1.00) 		0.0196
(1.00)
4q loss 0 (0%) 45 0.139
(1.00) 		0.0134
(1.00)
5p loss 0 (0%) 54 0.342
(1.00) 		0.214
(1.00)
5q loss 0 (0%) 46 0.722
(1.00) 		0.0196
(1.00)
6p loss 0 (0%) 53 0.299
(1.00) 		0.111
(1.00)
6q loss 0 (0%) 56 0.272
(1.00) 		0.0812
(1.00)
7p loss 0 (0%) 60 0.108
(1.00) 		0.308
(1.00)
8p loss 0 (0%) 49 0.801
(1.00) 		0.737
(1.00)
8q loss 0 (0%) 57 1
(1.00) 		0.547
(1.00)
9p loss 0 (0%) 40 0.4
(1.00) 		0.146
(1.00)
9q loss 0 (0%) 53 0.221
(1.00) 		0.167
(1.00)
10p loss 0 (0%) 50 0.216
(1.00) 		0.24
(1.00)
10q loss 0 (0%) 48 0.00218
(0.299) 		0.0178
(1.00)
11p loss 0 (0%) 52 0.128
(1.00) 		0.0936
(1.00)
12p loss 0 (0%) 55 0.635
(1.00) 		0.338
(1.00)
12q loss 0 (0%) 57 0.641
(1.00) 		0.0916
(1.00)
13q loss 0 (0%) 48 0.0272
(1.00) 		0.012
(1.00)
14q loss 0 (0%) 56 1
(1.00) 		0.582
(1.00)
15q loss 0 (0%) 54 0.469
(1.00) 		1
(1.00)
16p loss 0 (0%) 53 0.109
(1.00) 		0.0762
(1.00)
16q loss 0 (0%) 52 0.139
(1.00) 		0.525
(1.00)
18p loss 0 (0%) 51 0.307
(1.00) 		0.282
(1.00)
18q loss 0 (0%) 39 0.51
(1.00) 		0.0832
(1.00)
19p loss 0 (0%) 54 0.139
(1.00) 		0.898
(1.00)
19q loss 0 (0%) 57 0.178
(1.00) 		0.75
(1.00)
20p loss 0 (0%) 58 0.193
(1.00) 		1
(1.00)
21q loss 0 (0%) 36 0.764
(1.00) 		0.637
(1.00)
22q loss 0 (0%) 52 0.274
(1.00) 		0.0237
(1.00)
Xq loss 0 (0%) 60 0.784
(1.00) 		0.438
(1.00)
'3q gain' versus 'CN_CNMF'

P value = 0.000262 (Fisher's exact test), Q value = 0.037

Table S1.  Gene #6: '3q gain' versus Molecular Subtype #1: 'CN_CNMF'

nPatients CLUS_1 CLUS_2 CLUS_3
ALL 22 15 26
3Q GAIN CNV 11 7 1
3Q GAIN WILD-TYPE 11 8 25

Figure S1.  Get High-res Image Gene #6: '3q gain' versus Molecular Subtype #1: 'CN_CNMF'

'3q gain' versus 'METHLYATION_CNMF'

P value = 3.6e-05 (Fisher's exact test), Q value = 0.0051

Table S2.  Gene #6: '3q gain' versus Molecular Subtype #2: 'METHLYATION_CNMF'

nPatients CLUS_1 CLUS_2 CLUS_3
ALL 30 14 19
3Q GAIN CNV 17 1 1
3Q GAIN WILD-TYPE 13 13 18

Figure S2.  Get High-res Image Gene #6: '3q gain' versus Molecular Subtype #2: 'METHLYATION_CNMF'

'12p gain' versus 'CN_CNMF'

P value = 0.000606 (Fisher's exact test), Q value = 0.084

Table S3.  Gene #21: '12p gain' versus Molecular Subtype #1: 'CN_CNMF'

nPatients CLUS_1 CLUS_2 CLUS_3
ALL 22 15 26
12P GAIN CNV 12 7 2
12P GAIN WILD-TYPE 10 8 24

Figure S3.  Get High-res Image Gene #21: '12p gain' versus Molecular Subtype #1: 'CN_CNMF'

'11q loss' versus 'CN_CNMF'

P value = 0.000492 (Fisher's exact test), Q value = 0.069

Table S4.  Gene #55: '11q loss' versus Molecular Subtype #1: 'CN_CNMF'

nPatients CLUS_1 CLUS_2 CLUS_3
ALL 22 15 26
11Q LOSS CNV 12 2 2
11Q LOSS WILD-TYPE 10 13 24

Figure S4.  Get High-res Image Gene #55: '11q loss' versus Molecular Subtype #1: 'CN_CNMF'

'17p loss' versus 'METHLYATION_CNMF'

P value = 0.000683 (Fisher's exact test), Q value = 0.094

Table S5.  Gene #63: '17p loss' versus Molecular Subtype #2: 'METHLYATION_CNMF'

nPatients CLUS_1 CLUS_2 CLUS_3
ALL 30 14 19
17P LOSS CNV 2 2 10
17P LOSS WILD-TYPE 28 12 9

Figure S5.  Get High-res Image Gene #63: '17p loss' versus Molecular Subtype #2: 'METHLYATION_CNMF'

Methods & Data
Input

  • Mutation data file = broad_values_by_arm.mutsig.cluster.txt

  • Molecular subtypes file = ESCA-TP.transferedmergedcluster.txt

  • Number of patients = 63

  • Number of significantly arm-level cnvs = 71

  • Number of molecular subtypes = 2

  • Exclude genes that fewer than K tumors have mutations, K = 3

Fisher's exact test

For binary or multi-class clinical features (nominal or ordinal), two-tailed Fisher's exact tests (Fisher 1922) were used to estimate the P values using the 'fisher.test' function in R

Q value calculation

For multiple hypothesis correction, Q value is the False Discovery Rate (FDR) analogue of the P value (Benjamini and Hochberg 1995), defined as the minimum FDR at which the test may be called significant. We used the 'Benjamini and Hochberg' method of 'p.adjust' function in R to convert P values into Q values.

Download Results

In addition to the links below, the full results of the analysis summarized in this report can also be downloaded programmatically using firehose_get, or interactively from either the Broad GDAC website or TCGA Data Coordination Center Portal.

References
[1] Fisher, R.A., On the interpretation of chi-square from contingency tables, and the calculation of P, Journal of the Royal Statistical Society 85(1):87-94 (1922)
[2] Benjamini and Hochberg, Controlling the false discovery rate: a practical and powerful approach to multiple testing, Journal of the Royal Statistical Society Series B 59:289-300 (1995)
Copyright © 2013 Broad Institute TCGA GDAC as part of the TCGA Research Network. All rights reserved.
Made with Nozzle