Correlation between gene mutation status and molecular subtypes
Overview
Introduction

This pipeline computes the correlation between significantly recurrent gene mutations and molecular subtypes.

Summary

Testing the association between mutation status of 47 genes and 6 molecular subtypes across 116 patients, 7 significant findings detected with P value < 0.05 and Q value < 0.25.

  • TP53 mutation correlated to 'CN_CNMF' and 'MIRSEQ_CHIERARCHICAL'.

  • PIK3CA mutation correlated to 'METHLYATION_CNMF' and 'MRNASEQ_CNMF'.

  • ARID1A mutation correlated to 'CN_CNMF',  'MRNASEQ_CNMF', and 'MIRSEQ_CHIERARCHICAL'.

Results
Overview of the results

Table 1.  Get Full Table Overview of the association between mutation status of 47 genes 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 		MRNASEQ
CNMF 		MRNASEQ
CHIERARCHICAL 		MIRSEQ
CNMF 		MIRSEQ
CHIERARCHICAL
nMutated (%) nWild-Type Fisher's exact test Fisher's exact test Chi-square test Fisher's exact test Fisher's exact test Fisher's exact test
ARID1A 22 (19%) 94 7.42e-06
(0.00201) 		0.00783
(1.00) 		0.000768
(0.204) 		0.154
(1.00) 		0.0203
(1.00) 		0.000551
(0.147)
TP53 52 (45%) 64 0.000278
(0.0748) 		0.011
(1.00) 		0.0949
(1.00) 		0.0923
(1.00) 		0.00228
(0.599) 		9.14e-06
(0.00247)
PIK3CA 24 (21%) 92 0.00256
(0.669) 		1.26e-06
(0.000344) 		0.000322
(0.0862) 		0.192
(1.00) 		0.0282
(1.00) 		0.0194
(1.00)
CBWD1 14 (12%) 102 0.0342
(1.00) 		0.0585
(1.00) 		0.00578
(1.00) 		0.132
(1.00) 		0.0482
(1.00) 		0.216
(1.00)
KRAS 14 (12%) 102 0.0906
(1.00) 		0.292
(1.00) 		0.368
(1.00) 		0.0849
(1.00) 		0.685
(1.00) 		0.578
(1.00)
PGM5 16 (14%) 100 0.0118
(1.00) 		0.00266
(0.694) 		0.192
(1.00) 		0.306
(1.00) 		0.158
(1.00) 		0.222
(1.00)
RPL22 9 (8%) 107 0.254
(1.00) 		0.123
(1.00) 		0.091
(1.00) 		0.299
(1.00) 		0.152
(1.00)
TRIM48 10 (9%) 106 0.00622
(1.00) 		0.174
(1.00) 		0.0438
(1.00) 		0.16
(1.00) 		0.0839
(1.00) 		0.0155
(1.00)
XPOT 6 (5%) 110 0.0703
(1.00) 		0.0277
(1.00) 		0.373
(1.00) 		0.34
(1.00) 		0.128
(1.00)
ACVR2A 13 (11%) 103 0.00214
(0.564) 		0.0128
(1.00) 		0.00158
(0.418) 		0.0808
(1.00) 		0.0253
(1.00) 		0.0203
(1.00)
RHOA 7 (6%) 109 0.354
(1.00) 		0.639
(1.00) 		0.476
(1.00) 		1
(1.00) 		0.0881
(1.00) 		1
(1.00)
OR8H3 10 (9%) 106 0.0238
(1.00) 		0.0884
(1.00) 		0.0274
(1.00) 		0.572
(1.00) 		0.221
(1.00) 		0.474
(1.00)
EDNRB 12 (10%) 104 0.112
(1.00) 		0.543
(1.00) 		0.868
(1.00) 		0.324
(1.00) 		0.475
(1.00) 		0.588
(1.00)
ZNF804B 18 (16%) 98 0.236
(1.00) 		0.047
(1.00) 		0.0795
(1.00) 		0.00277
(0.719) 		0.0127
(1.00) 		0.274
(1.00)
IRF2 8 (7%) 108 0.0405
(1.00) 		0.0511
(1.00) 		0.0225
(1.00) 		0.886
(1.00) 		0.176
(1.00) 		0.321
(1.00)
IAPP 4 (3%) 112 0.859
(1.00) 		0.736
(1.00) 		0.455
(1.00) 		0.147
(1.00) 		1
(1.00)
PCDH15 22 (19%) 94 0.634
(1.00) 		0.863
(1.00) 		0.928
(1.00) 		0.439
(1.00) 		0.594
(1.00) 		0.947
(1.00)
SPRYD5 8 (7%) 108 0.843
(1.00) 		0.0343
(1.00) 		0.441
(1.00) 		0.717
(1.00) 		0.263
(1.00) 		0.687
(1.00)
TUSC3 9 (8%) 107 0.666
(1.00) 		0.88
(1.00) 		0.968
(1.00) 		1
(1.00) 		1
(1.00) 		0.712
(1.00)
FGF22 3 (3%) 113 0.0257
(1.00) 		0.028
(1.00) 		0.327
(1.00) 		1
(1.00) 		0.573
(1.00)
HLA-B 9 (8%) 107 0.13
(1.00) 		0.00377
(0.977) 		0.123
(1.00) 		0.117
(1.00) 		0.299
(1.00) 		0.636
(1.00)
PTH2 3 (3%) 113 0.812
(1.00) 		0.302
(1.00) 		0.171
(1.00) 		0.161
(1.00) 		0.155
(1.00) 		0.384
(1.00)
C17ORF63 3 (3%) 113 0.435
(1.00) 		0.736
(1.00) 		1
(1.00) 		1
(1.00) 		0.773
(1.00)
SMAD4 7 (6%) 109 1
(1.00) 		0.859
(1.00) 		0.186
(1.00) 		0.477
(1.00) 		1
(1.00) 		0.753
(1.00)
POTEG 6 (5%) 110 0.294
(1.00) 		0.48
(1.00) 		0.469
(1.00) 		0.854
(1.00) 		1
(1.00) 		0.375
(1.00)
RNF43 13 (11%) 103 0.00576
(1.00) 		0.0168
(1.00) 		0.0478
(1.00) 		0.0789
(1.00) 		0.0253
(1.00) 		0.0592
(1.00)
WBSCR17 12 (10%) 104 0.288
(1.00) 		0.00419
(1.00) 		0.0813
(1.00) 		0.903
(1.00) 		0.0289
(1.00) 		0.412
(1.00)
PHF2 12 (10%) 104 0.006
(1.00) 		0.0128
(1.00) 		0.00746
(1.00) 		0.0311
(1.00) 		0.0289
(1.00) 		0.00602
(1.00)
TPTE 14 (12%) 102 0.563
(1.00) 		1
(1.00) 		0.464
(1.00) 		0.698
(1.00) 		0.0419
(1.00) 		0.856
(1.00)
CDH1 11 (9%) 105 0.14
(1.00) 		0.769
(1.00) 		0.0161
(1.00) 		0.121
(1.00) 		0.0919
(1.00) 		0.0862
(1.00)
CPS1 13 (11%) 103 0.731
(1.00) 		0.223
(1.00) 		0.518
(1.00) 		0.856
(1.00) 		0.906
(1.00) 		0.335
(1.00)
INO80E 5 (4%) 111 0.534
(1.00) 		0.798
(1.00) 		0.713
(1.00) 		0.369
(1.00) 		1
(1.00)
ELF3 5 (4%) 111 0.118
(1.00) 		0.163
(1.00) 		0.263
(1.00) 		0.358
(1.00) 		0.493
(1.00) 		0.221
(1.00)
PARK2 9 (8%) 107 0.235
(1.00) 		0.0548
(1.00) 		0.306
(1.00) 		0.0219
(1.00) 		0.869
(1.00) 		0.636
(1.00)
LARP4B 5 (4%) 111 0.0339
(1.00) 		0.174
(1.00) 		0.108
(1.00) 		0.358
(1.00) 		0.493
(1.00) 		0.221
(1.00)
OR6K3 6 (5%) 110 0.347
(1.00) 		0.358
(1.00) 		0.0808
(1.00) 		0.867
(1.00) 		0.434
(1.00) 		1
(1.00)
TM7SF4 7 (6%) 109 0.931
(1.00) 		0.223
(1.00) 		0.642
(1.00) 		0.782
(1.00) 		0.852
(1.00) 		0.376
(1.00)
UPF3A 6 (5%) 110 0.347
(1.00) 		0.174
(1.00) 		0.0905
(1.00) 		0.231
(1.00) 		0.434
(1.00) 		0.619
(1.00)
C7ORF63 5 (4%) 111 0.118
(1.00) 		0.108
(1.00) 		0.569
(1.00) 		0.493
(1.00) 		0.697
(1.00)
KDM4B 10 (9%) 106 0.0222
(1.00) 		0.0343
(1.00) 		0.211
(1.00) 		0.607
(1.00) 		0.221
(1.00) 		0.53
(1.00)
KIAA0748 7 (6%) 109 0.776
(1.00) 		0.545
(1.00) 		0.279
(1.00) 		0.473
(1.00) 		0.852
(1.00) 		0.432
(1.00)
OR8B4 3 (3%) 113 0.155
(1.00) 		0.00894
(1.00) 		1
(1.00) 		0.573
(1.00)
POM121L12 7 (6%) 109 0.0617
(1.00) 		0.122
(1.00) 		0.0988
(1.00) 		0.209
(1.00) 		0.852
(1.00) 		0.376
(1.00)
RASA1 9 (8%) 107 0.448
(1.00) 		0.551
(1.00) 		0.467
(1.00) 		0.0148
(1.00) 		0.066
(1.00) 		0.314
(1.00)
SLITRK6 10 (9%) 106 0.258
(1.00) 		0.88
(1.00) 		0.522
(1.00) 		0.717
(1.00) 		0.14
(1.00) 		0.307
(1.00)
TP53TG5 4 (3%) 112 0.024
(1.00) 		0.736
(1.00) 		1
(1.00) 		1
(1.00) 		0.641
(1.00)
LHCGR 7 (6%) 109 1
(1.00) 		0.813
(1.00) 		0.999
(1.00) 		0.713
(1.00) 		0.597
(1.00) 		0.275
(1.00)
'TP53 MUTATION STATUS' versus 'CN_CNMF'

P value = 0.000278 (Fisher's exact test), Q value = 0.075

Table S1.  Gene #5: 'TP53 MUTATION STATUS' versus Clinical Feature #1: 'CN_CNMF'

nPatients CLUS_1 CLUS_2 CLUS_3 CLUS_4
ALL 47 15 10 43
TP53 MUTATED 11 9 3 28
TP53 WILD-TYPE 36 6 7 15

Figure S1.  Get High-res Image Gene #5: 'TP53 MUTATION STATUS' versus Clinical Feature #1: 'CN_CNMF'

'TP53 MUTATION STATUS' versus 'MIRSEQ_CHIERARCHICAL'

P value = 9.14e-06 (Fisher's exact test), Q value = 0.0025

Table S2.  Gene #5: 'TP53 MUTATION STATUS' versus Clinical Feature #6: 'MIRSEQ_CHIERARCHICAL'

nPatients CLUS_1 CLUS_2 CLUS_3
ALL 26 23 67
TP53 MUTATED 22 6 24
TP53 WILD-TYPE 4 17 43

Figure S2.  Get High-res Image Gene #5: 'TP53 MUTATION STATUS' versus Clinical Feature #6: 'MIRSEQ_CHIERARCHICAL'

'PIK3CA MUTATION STATUS' versus 'METHLYATION_CNMF'

P value = 1.26e-06 (Fisher's exact test), Q value = 0.00034

Table S3.  Gene #8: 'PIK3CA MUTATION STATUS' versus Clinical Feature #2: 'METHLYATION_CNMF'

nPatients CLUS_1 CLUS_2 CLUS_3 CLUS_4
ALL 9 14 22 23
PIK3CA MUTATED 9 4 3 2
PIK3CA WILD-TYPE 0 10 19 21

Figure S3.  Get High-res Image Gene #8: 'PIK3CA MUTATION STATUS' versus Clinical Feature #2: 'METHLYATION_CNMF'

'PIK3CA MUTATION STATUS' versus 'MRNASEQ_CNMF'

P value = 0.000322 (Chi-square test), Q value = 0.086

Table S4.  Gene #8: 'PIK3CA MUTATION STATUS' versus Clinical Feature #3: 'MRNASEQ_CNMF'

nPatients CLUS_1 CLUS_2 CLUS_3 CLUS_4 CLUS_5
ALL 23 18 17 23 23
PIK3CA MUTATED 12 2 1 6 1
PIK3CA WILD-TYPE 11 16 16 17 22

Figure S4.  Get High-res Image Gene #8: 'PIK3CA MUTATION STATUS' versus Clinical Feature #3: 'MRNASEQ_CNMF'

'ARID1A MUTATION STATUS' versus 'CN_CNMF'

P value = 7.42e-06 (Fisher's exact test), Q value = 0.002

Table S5.  Gene #11: 'ARID1A MUTATION STATUS' versus Clinical Feature #1: 'CN_CNMF'

nPatients CLUS_1 CLUS_2 CLUS_3 CLUS_4
ALL 47 15 10 43
ARID1A MUTATED 15 2 5 0
ARID1A WILD-TYPE 32 13 5 43

Figure S5.  Get High-res Image Gene #11: 'ARID1A MUTATION STATUS' versus Clinical Feature #1: 'CN_CNMF'

'ARID1A MUTATION STATUS' versus 'MRNASEQ_CNMF'

P value = 0.000768 (Chi-square test), Q value = 0.2

Table S6.  Gene #11: 'ARID1A MUTATION STATUS' versus Clinical Feature #3: 'MRNASEQ_CNMF'

nPatients CLUS_1 CLUS_2 CLUS_3 CLUS_4 CLUS_5
ALL 23 18 17 23 23
ARID1A MUTATED 11 2 2 5 0
ARID1A WILD-TYPE 12 16 15 18 23

Figure S6.  Get High-res Image Gene #11: 'ARID1A MUTATION STATUS' versus Clinical Feature #3: 'MRNASEQ_CNMF'

'ARID1A MUTATION STATUS' versus 'MIRSEQ_CHIERARCHICAL'

P value = 0.000551 (Fisher's exact test), Q value = 0.15

Table S7.  Gene #11: 'ARID1A MUTATION STATUS' versus Clinical Feature #6: 'MIRSEQ_CHIERARCHICAL'

nPatients CLUS_1 CLUS_2 CLUS_3
ALL 26 23 67
ARID1A MUTATED 0 2 20
ARID1A WILD-TYPE 26 21 47

Figure S7.  Get High-res Image Gene #11: 'ARID1A MUTATION STATUS' versus Clinical Feature #6: 'MIRSEQ_CHIERARCHICAL'

Methods & Data
Input

  • Mutation data file = STAD-TP.mutsig.cluster.txt

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

  • Number of patients = 116

  • Number of significantly mutated genes = 47

  • Number of Molecular subtypes = 6

  • 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

Chi-square test

For multi-class clinical features (nominal or ordinal), Chi-square tests (Greenwood and Nikulin 1996) were used to estimate the P values using the 'chisq.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] Greenwood and Nikulin, A guide to chi-squared testing, Wiley, New York. ISBN 047155779X (1996)
[3] 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