Correlation between copy number variation genes (focal events) and molecular subtypes
Overview
Introduction

This pipeline computes the correlation between significant copy number variation (cnv focal) genes and molecular subtypes.

Summary

Testing the association between copy number variation 40 focal events and 6 molecular subtypes across 73 patients, 7 significant findings detected with P value < 0.05 and Q value < 0.25.

  • 3p cnv correlated to 'MIRSEQ_CHIERARCHICAL' and 'MIRSEQ_MATURE_CHIERARCHICAL'.

  • 3q cnv correlated to 'CN_CNMF',  'METHLYATION_CNMF',  'MIRSEQ_CNMF',  'MIRSEQ_CHIERARCHICAL', and 'MIRSEQ_MATURE_CHIERARCHICAL'.

Results
Overview of the results

Table 1.  Get Full Table Overview of the association between significant copy number variation of 40 focal events and 6 molecular subtypes. Shown in the table are P values (Q values). Thresholded by P value < 0.05 and Q value < 0.25, 7 significant findings detected.

Clinical
Features 		CN
CNMF 		METHLYATION
CNMF 		MIRSEQ
CNMF 		MIRSEQ
CHIERARCHICAL 		MIRSEQ
MATURE
CNMF 		MIRSEQ
MATURE
CHIERARCHICAL
nCNV (%) nWild-Type Fisher's exact test Fisher's exact test Fisher's exact test Fisher's exact test Fisher's exact test Fisher's exact test
3q 46 (63%) 27 1.48e-06
(0.000354) 		8.06e-05
(0.0191) 		0.000616
(0.144) 		7.73e-05
(0.0185) 		0.0041
(0.948) 		7.73e-05
(0.0185)
3p 54 (74%) 19 0.0297
(1.00) 		0.0134
(1.00) 		0.00183
(0.425) 		0.000202
(0.0477) 		0.0211
(1.00) 		0.000202
(0.0477)
1p 23 (32%) 50 0.745
(1.00) 		0.927
(1.00) 		1
(1.00) 		1
(1.00) 		0.142
(1.00) 		1
(1.00)
1q 36 (49%) 37 0.0298
(1.00) 		0.534
(1.00) 		0.272
(1.00) 		0.183
(1.00) 		0.061
(1.00) 		0.183
(1.00)
2p 32 (44%) 41 0.206
(1.00) 		0.59
(1.00) 		0.181
(1.00) 		0.536
(1.00) 		0.0307
(1.00) 		0.536
(1.00)
2q 27 (37%) 46 0.529
(1.00) 		0.605
(1.00) 		0.662
(1.00) 		0.634
(1.00) 		0.219
(1.00) 		0.634
(1.00)
4p 48 (66%) 25 0.108
(1.00) 		0.615
(1.00) 		0.405
(1.00) 		0.0809
(1.00) 		0.658
(1.00) 		0.0809
(1.00)
4q 40 (55%) 33 0.0166
(1.00) 		0.518
(1.00) 		0.722
(1.00) 		0.321
(1.00) 		0.915
(1.00) 		0.321
(1.00)
5p 51 (70%) 22 0.872
(1.00) 		0.339
(1.00) 		0.0705
(1.00) 		0.258
(1.00) 		0.292
(1.00) 		0.258
(1.00)
5q 42 (58%) 31 0.936
(1.00) 		0.253
(1.00) 		0.295
(1.00) 		0.184
(1.00) 		0.275
(1.00) 		0.184
(1.00)
6p 27 (37%) 46 0.0918
(1.00) 		0.481
(1.00) 		0.717
(1.00) 		0.634
(1.00) 		0.377
(1.00) 		0.634
(1.00)
6q 24 (33%) 49 0.0442
(1.00) 		0.373
(1.00) 		0.452
(1.00) 		0.362
(1.00) 		0.779
(1.00) 		0.362
(1.00)
7p 53 (73%) 20 0.616
(1.00) 		0.467
(1.00) 		0.434
(1.00) 		0.187
(1.00) 		0.546
(1.00) 		0.187
(1.00)
7q 44 (60%) 29 1
(1.00) 		1
(1.00) 		0.558
(1.00) 		0.244
(1.00) 		0.181
(1.00) 		0.244
(1.00)
8p 56 (77%) 17 0.491
(1.00) 		0.152
(1.00) 		0.612
(1.00) 		0.128
(1.00) 		0.536
(1.00) 		0.128
(1.00)
8q 51 (70%) 22 0.71
(1.00) 		0.267
(1.00) 		0.757
(1.00) 		0.329
(1.00) 		0.888
(1.00) 		0.329
(1.00)
9p 49 (67%) 24 0.581
(1.00) 		0.373
(1.00) 		0.559
(1.00) 		0.186
(1.00) 		0.589
(1.00) 		0.186
(1.00)
9q 40 (55%) 33 0.642
(1.00) 		0.114
(1.00) 		0.319
(1.00) 		0.0913
(1.00) 		0.251
(1.00) 		0.0913
(1.00)
10p 36 (49%) 37 0.868
(1.00) 		0.234
(1.00) 		0.0916
(1.00) 		0.553
(1.00) 		0.319
(1.00) 		0.553
(1.00)
10q 35 (48%) 38 0.112
(1.00) 		0.00141
(0.328) 		0.0292
(1.00) 		0.00932
(1.00) 		0.0494
(1.00) 		0.00932
(1.00)
11p 37 (51%) 36 0.348
(1.00) 		0.431
(1.00) 		0.55
(1.00) 		0.553
(1.00) 		0.573
(1.00) 		0.553
(1.00)
11q 40 (55%) 33 0.072
(1.00) 		0.801
(1.00) 		0.932
(1.00) 		0.747
(1.00) 		0.402
(1.00) 		0.747
(1.00)
12p 46 (63%) 27 0.468
(1.00) 		0.529
(1.00) 		0.154
(1.00) 		0.428
(1.00) 		0.641
(1.00) 		0.428
(1.00)
12q 33 (45%) 40 0.382
(1.00) 		0.921
(1.00) 		0.429
(1.00) 		0.719
(1.00) 		0.667
(1.00) 		0.719
(1.00)
13q 46 (63%) 27 0.603
(1.00) 		0.511
(1.00) 		0.752
(1.00) 		0.397
(1.00) 		0.781
(1.00) 		0.397
(1.00)
14q 40 (55%) 33 0.82
(1.00) 		0.619
(1.00) 		0.5
(1.00) 		0.615
(1.00) 		0.884
(1.00) 		0.615
(1.00)
15q 32 (44%) 41 0.77
(1.00) 		0.0619
(1.00) 		0.0929
(1.00) 		0.116
(1.00) 		0.366
(1.00) 		0.116
(1.00)
16p 34 (47%) 39 0.612
(1.00) 		1
(1.00) 		0.932
(1.00) 		1
(1.00) 		0.716
(1.00) 		1
(1.00)
16q 32 (44%) 41 0.59
(1.00) 		0.848
(1.00) 		0.187
(1.00) 		0.819
(1.00) 		0.424
(1.00) 		0.819
(1.00)
17p 41 (56%) 32 0.209
(1.00) 		0.0591
(1.00) 		0.0642
(1.00) 		0.0347
(1.00) 		0.0718
(1.00) 		0.0347
(1.00)
17q 29 (40%) 44 0.691
(1.00) 		0.735
(1.00) 		0.283
(1.00) 		1
(1.00) 		0.325
(1.00) 		1
(1.00)
18p 49 (67%) 24 0.0188
(1.00) 		0.462
(1.00) 		0.476
(1.00) 		0.492
(1.00) 		0.587
(1.00) 		0.492
(1.00)
18q 49 (67%) 24 0.557
(1.00) 		0.104
(1.00) 		0.0929
(1.00) 		0.0973
(1.00) 		0.0457
(1.00) 		0.0973
(1.00)
19p 37 (51%) 36 0.131
(1.00) 		0.0708
(1.00) 		0.018
(1.00) 		0.553
(1.00) 		0.014
(1.00) 		0.553
(1.00)
19q 36 (49%) 37 0.573
(1.00) 		0.124
(1.00) 		0.0525
(1.00) 		0.681
(1.00) 		0.0368
(1.00) 		0.681
(1.00)
20p 48 (66%) 25 0.668
(1.00) 		0.741
(1.00) 		0.11
(1.00) 		0.174
(1.00) 		0.779
(1.00) 		0.174
(1.00)
20q 45 (62%) 28 0.138
(1.00) 		0.23
(1.00) 		0.0394
(1.00) 		0.0153
(1.00) 		0.18
(1.00) 		0.0153
(1.00)
21q 52 (71%) 21 0.623
(1.00) 		0.0884
(1.00) 		0.228
(1.00) 		0.0665
(1.00) 		0.408
(1.00) 		0.0665
(1.00)
22q 43 (59%) 30 0.355
(1.00) 		0.841
(1.00) 		0.418
(1.00) 		0.11
(1.00) 		0.662
(1.00) 		0.11
(1.00)
xq 36 (49%) 37 0.717
(1.00) 		0.415
(1.00) 		0.915
(1.00) 		1
(1.00) 		0.712
(1.00) 		1
(1.00)
'3p' versus 'MIRSEQ_CHIERARCHICAL'

P value = 0.000202 (Fisher's exact test), Q value = 0.048

Table S1.  Gene #5: '3p' versus Molecular Subtype #4: 'MIRSEQ_CHIERARCHICAL'

nPatients CLUS_1 CLUS_2 CLUS_3
ALL 1 28 43
3P MUTATED 0 15 39
3P WILD-TYPE 1 13 4

Figure S1.  Get High-res Image Gene #5: '3p' versus Molecular Subtype #4: 'MIRSEQ_CHIERARCHICAL'

'3p' versus 'MIRSEQ_MATURE_CHIERARCHICAL'

P value = 0.000202 (Fisher's exact test), Q value = 0.048

Table S2.  Gene #5: '3p' versus Molecular Subtype #6: 'MIRSEQ_MATURE_CHIERARCHICAL'

nPatients CLUS_1 CLUS_2 CLUS_3
ALL 1 28 43
3P MUTATED 0 15 39
3P WILD-TYPE 1 13 4

Figure S2.  Get High-res Image Gene #5: '3p' versus Molecular Subtype #6: 'MIRSEQ_MATURE_CHIERARCHICAL'

'3q' versus 'CN_CNMF'

P value = 1.48e-06 (Fisher's exact test), Q value = 0.00035

Table S3.  Gene #6: '3q' versus Molecular Subtype #1: 'CN_CNMF'

nPatients CLUS_1 CLUS_2 CLUS_3
ALL 15 34 24
3Q MUTATED 13 11 22
3Q WILD-TYPE 2 23 2

Figure S3.  Get High-res Image Gene #6: '3q' versus Molecular Subtype #1: 'CN_CNMF'

'3q' versus 'METHLYATION_CNMF'

P value = 8.06e-05 (Fisher's exact test), Q value = 0.019

Table S4.  Gene #6: '3q' versus Molecular Subtype #2: 'METHLYATION_CNMF'

nPatients CLUS_1 CLUS_2 CLUS_3
ALL 34 15 24
3Q MUTATED 30 7 9
3Q WILD-TYPE 4 8 15

Figure S4.  Get High-res Image Gene #6: '3q' versus Molecular Subtype #2: 'METHLYATION_CNMF'

'3q' versus 'MIRSEQ_CNMF'

P value = 0.000616 (Fisher's exact test), Q value = 0.14

Table S5.  Gene #6: '3q' versus Molecular Subtype #3: 'MIRSEQ_CNMF'

nPatients CLUS_1 CLUS_2 CLUS_3
ALL 30 14 28
3Q MUTATED 25 10 10
3Q WILD-TYPE 5 4 18

Figure S5.  Get High-res Image Gene #6: '3q' versus Molecular Subtype #3: 'MIRSEQ_CNMF'

'3q' versus 'MIRSEQ_CHIERARCHICAL'

P value = 7.73e-05 (Fisher's exact test), Q value = 0.018

Table S6.  Gene #6: '3q' versus Molecular Subtype #4: 'MIRSEQ_CHIERARCHICAL'

nPatients CLUS_1 CLUS_2 CLUS_3
ALL 1 28 43
3Q MUTATED 0 10 35
3Q WILD-TYPE 1 18 8

Figure S6.  Get High-res Image Gene #6: '3q' versus Molecular Subtype #4: 'MIRSEQ_CHIERARCHICAL'

'3q' versus 'MIRSEQ_MATURE_CHIERARCHICAL'

P value = 7.73e-05 (Fisher's exact test), Q value = 0.018

Table S7.  Gene #6: '3q' versus Molecular Subtype #6: 'MIRSEQ_MATURE_CHIERARCHICAL'

nPatients CLUS_1 CLUS_2 CLUS_3
ALL 1 28 43
3Q MUTATED 0 10 35
3Q WILD-TYPE 1 18 8

Figure S7.  Get High-res Image Gene #6: '3q' versus Molecular Subtype #6: 'MIRSEQ_MATURE_CHIERARCHICAL'

Methods & Data
Input

  • Copy number data file = transformed.cor.cli.txt

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

  • Number of patients = 73

  • Number of significantly focal cnvs = 40

  • Number of molecular subtypes = 6

  • Exclude genes that fewer than K tumors have alterations, 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.

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)
Made with Nozzle