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 7 genes and 6 molecular subtypes across 196 patients, 15 significant findings detected with P value < 0.05 and Q value < 0.25.

DNMT3A mutation correlated to 'METHLYATION_CNMF', 'MIRSEQ_CNMF', and 'MIRSEQ_CHIERARCHICAL'.
FLT3 mutation correlated to 'METHLYATION_CNMF', 'MRNASEQ_CNMF', 'MIRSEQ_CNMF', and 'MIRSEQ_CHIERARCHICAL'.
IDH2 mutation correlated to 'METHLYATION_CNMF'.
IDH1 mutation correlated to 'METHLYATION_CNMF'.
NPM1 mutation correlated to 'CN_CNMF', 'METHLYATION_CNMF', 'MRNASEQ_CNMF', 'MRNASEQ_CHIERARCHICAL', 'MIRSEQ_CNMF', and 'MIRSEQ_CHIERARCHICAL'.

Results

Overview of the results

Table 1. Get Full Table Overview of the association between mutation status of 7 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, 15 significant findings detected.


		Clinical Features	CN CNMF	METHLYATION CNMF	MRNASEQ CNMF	MRNASEQ CHIERARCHICAL	MIRSEQ CNMF	MIRSEQ CHIERARCHICAL
	nMutated (%)	nWild-Type	Chi-square test	Chi-square test	Fisher's exact test	Fisher's exact test	Fisher's exact test	Fisher's exact test
NPM1	54 (28%)	142	0.000852 (0.0259)	3.26e-20 (1.34e-18)	1.79e-13 (6.99e-12)	0.00649 (0.182)	6.58e-14 (2.63e-12)	1.07e-22 (4.51e-21)
FLT3	56 (29%)	140	0.0152 (0.39)	2.48e-05 (0.000844)	1.86e-07 (7.08e-06)	0.0209 (0.502)	0.000471 (0.0151)	4.22e-05 (0.00139)
DNMT3A	51 (26%)	145	0.977 (1.00)	4e-07 (1.48e-05)	0.0122 (0.329)	0.195 (1.00)	0.000837 (0.0259)	3.94e-06 (0.000138)
IDH2	20 (10%)	176	0.274 (1.00)	1.59e-06 (5.74e-05)	0.725 (1.00)	1 (1.00)	0.739 (1.00)	0.599 (1.00)
IDH1	19 (10%)	177	0.969 (1.00)	0.00563 (0.163)	0.423 (1.00)	0.268 (1.00)	0.284 (1.00)	0.436 (1.00)
U2AF1	8 (4%)	188	0.776 (1.00)	0.128 (1.00)	0.232 (1.00)	0.426 (1.00)	0.0716 (1.00)	0.234 (1.00)
NRAS	15 (8%)	181	0.015 (0.39)	0.892 (1.00)	0.645 (1.00)	0.539 (1.00)	0.825 (1.00)	0.765 (1.00)

'DNMT3A MUTATION STATUS' versus 'METHLYATION_CNMF'

P value = 4e-07 (Chi-square test), Q value = 1.5e-05

Table S1. Gene #1: 'DNMT3A MUTATION STATUS' versus Clinical Feature #2: 'METHLYATION_CNMF'

nPatients	CLUS_1	CLUS_2	CLUS_3	CLUS_4	CLUS_5
ALL	47	44	65	14	20
DNMT3A MUTATED	25	1	19	1	3
DNMT3A WILD-TYPE	22	43	46	13	17

Figure S1. Get High-res Image Gene #1: 'DNMT3A MUTATION STATUS' versus Clinical Feature #2: 'METHLYATION_CNMF'

'DNMT3A MUTATION STATUS' versus 'MIRSEQ_CNMF'

P value = 0.000837 (Fisher's exact test), Q value = 0.026

Table S2. Gene #1: 'DNMT3A MUTATION STATUS' versus Clinical Feature #5: 'MIRSEQ_CNMF'

nPatients	CLUS_1	CLUS_2	CLUS_3	CLUS_4
ALL	58	38	42	46
DNMT3A MUTATED	24	4	5	13
DNMT3A WILD-TYPE	34	34	37	33

Figure S2. Get High-res Image Gene #1: 'DNMT3A MUTATION STATUS' versus Clinical Feature #5: 'MIRSEQ_CNMF'

'DNMT3A MUTATION STATUS' versus 'MIRSEQ_CHIERARCHICAL'

P value = 3.94e-06 (Fisher's exact test), Q value = 0.00014

Table S3. Gene #1: 'DNMT3A MUTATION STATUS' versus Clinical Feature #6: 'MIRSEQ_CHIERARCHICAL'

nPatients	CLUS_1	CLUS_2	CLUS_3
ALL	36	80	68
DNMT3A MUTATED	2	34	10
DNMT3A WILD-TYPE	34	46	58

Figure S3. Get High-res Image Gene #1: 'DNMT3A MUTATION STATUS' versus Clinical Feature #6: 'MIRSEQ_CHIERARCHICAL'

'FLT3 MUTATION STATUS' versus 'METHLYATION_CNMF'

P value = 2.48e-05 (Chi-square test), Q value = 0.00084

Table S4. Gene #3: 'FLT3 MUTATION STATUS' versus Clinical Feature #2: 'METHLYATION_CNMF'

nPatients	CLUS_1	CLUS_2	CLUS_3	CLUS_4	CLUS_5
ALL	47	44	65	14	20
FLT3 MUTATED	26	12	8	2	7
FLT3 WILD-TYPE	21	32	57	12	13

Figure S4. Get High-res Image Gene #3: 'FLT3 MUTATION STATUS' versus Clinical Feature #2: 'METHLYATION_CNMF'

'FLT3 MUTATION STATUS' versus 'MRNASEQ_CNMF'

P value = 1.86e-07 (Fisher's exact test), Q value = 7.1e-06

Table S5. Gene #3: 'FLT3 MUTATION STATUS' versus Clinical Feature #3: 'MRNASEQ_CNMF'

nPatients	CLUS_1	CLUS_2	CLUS_3
ALL	73	52	45
FLT3 MUTATED	6	20	23
FLT3 WILD-TYPE	67	32	22

Figure S5. Get High-res Image Gene #3: 'FLT3 MUTATION STATUS' versus Clinical Feature #3: 'MRNASEQ_CNMF'

'FLT3 MUTATION STATUS' versus 'MIRSEQ_CNMF'

P value = 0.000471 (Fisher's exact test), Q value = 0.015

Table S6. Gene #3: 'FLT3 MUTATION STATUS' versus Clinical Feature #5: 'MIRSEQ_CNMF'

nPatients	CLUS_1	CLUS_2	CLUS_3	CLUS_4
ALL	58	38	42	46
FLT3 MUTATED	25	4	7	18
FLT3 WILD-TYPE	33	34	35	28

Figure S6. Get High-res Image Gene #3: 'FLT3 MUTATION STATUS' versus Clinical Feature #5: 'MIRSEQ_CNMF'

'FLT3 MUTATION STATUS' versus 'MIRSEQ_CHIERARCHICAL'

P value = 4.22e-05 (Fisher's exact test), Q value = 0.0014

Table S7. Gene #3: 'FLT3 MUTATION STATUS' versus Clinical Feature #6: 'MIRSEQ_CHIERARCHICAL'

nPatients	CLUS_1	CLUS_2	CLUS_3
ALL	36	80	68
FLT3 MUTATED	4	37	13
FLT3 WILD-TYPE	32	43	55

Figure S7. Get High-res Image Gene #3: 'FLT3 MUTATION STATUS' versus Clinical Feature #6: 'MIRSEQ_CHIERARCHICAL'

'IDH2 MUTATION STATUS' versus 'METHLYATION_CNMF'

P value = 1.59e-06 (Chi-square test), Q value = 5.7e-05

Table S8. Gene #4: 'IDH2 MUTATION STATUS' versus Clinical Feature #2: 'METHLYATION_CNMF'

nPatients	CLUS_1	CLUS_2	CLUS_3	CLUS_4	CLUS_5
ALL	47	44	65	14	20
IDH2 MUTATED	0	0	11	6	1
IDH2 WILD-TYPE	47	44	54	8	19

Figure S8. Get High-res Image Gene #4: 'IDH2 MUTATION STATUS' versus Clinical Feature #2: 'METHLYATION_CNMF'

'IDH1 MUTATION STATUS' versus 'METHLYATION_CNMF'

P value = 0.00563 (Chi-square test), Q value = 0.16

Table S9. Gene #5: 'IDH1 MUTATION STATUS' versus Clinical Feature #2: 'METHLYATION_CNMF'

nPatients	CLUS_1	CLUS_2	CLUS_3	CLUS_4	CLUS_5
ALL	47	44	65	14	20
IDH1 MUTATED	5	0	10	4	0
IDH1 WILD-TYPE	42	44	55	10	20

Figure S9. Get High-res Image Gene #5: 'IDH1 MUTATION STATUS' versus Clinical Feature #2: 'METHLYATION_CNMF'

'NPM1 MUTATION STATUS' versus 'CN_CNMF'

P value = 0.000852 (Chi-square test), Q value = 0.026

Table S10. Gene #6: 'NPM1 MUTATION STATUS' versus Clinical Feature #1: 'CN_CNMF'

nPatients	CLUS_1	CLUS_2	CLUS_3	CLUS_4	CLUS_5
ALL	138	15	15	18	1
NPM1 MUTATED	50	0	1	1	0
NPM1 WILD-TYPE	88	15	14	17	1

Figure S10. Get High-res Image Gene #6: 'NPM1 MUTATION STATUS' versus Clinical Feature #1: 'CN_CNMF'

'NPM1 MUTATION STATUS' versus 'METHLYATION_CNMF'

P value = 3.26e-20 (Chi-square test), Q value = 1.3e-18

Table S11. Gene #6: 'NPM1 MUTATION STATUS' versus Clinical Feature #2: 'METHLYATION_CNMF'

nPatients	CLUS_1	CLUS_2	CLUS_3	CLUS_4	CLUS_5
ALL	47	44	65	14	20
NPM1 MUTATED	36	0	4	9	4
NPM1 WILD-TYPE	11	44	61	5	16

Figure S11. Get High-res Image Gene #6: 'NPM1 MUTATION STATUS' versus Clinical Feature #2: 'METHLYATION_CNMF'

'NPM1 MUTATION STATUS' versus 'MRNASEQ_CNMF'

P value = 1.79e-13 (Fisher's exact test), Q value = 7e-12

Table S12. Gene #6: 'NPM1 MUTATION STATUS' versus Clinical Feature #3: 'MRNASEQ_CNMF'

nPatients	CLUS_1	CLUS_2	CLUS_3
ALL	73	52	45
NPM1 MUTATED	1	22	25
NPM1 WILD-TYPE	72	30	20

Figure S12. Get High-res Image Gene #6: 'NPM1 MUTATION STATUS' versus Clinical Feature #3: 'MRNASEQ_CNMF'

'NPM1 MUTATION STATUS' versus 'MRNASEQ_CHIERARCHICAL'

P value = 0.00649 (Fisher's exact test), Q value = 0.18

Table S13. Gene #6: 'NPM1 MUTATION STATUS' versus Clinical Feature #4: 'MRNASEQ_CHIERARCHICAL'

nPatients	CLUS_1	CLUS_2
ALL	57	113
NPM1 MUTATED	24	24
NPM1 WILD-TYPE	33	89

Figure S13. Get High-res Image Gene #6: 'NPM1 MUTATION STATUS' versus Clinical Feature #4: 'MRNASEQ_CHIERARCHICAL'

'NPM1 MUTATION STATUS' versus 'MIRSEQ_CNMF'

P value = 6.58e-14 (Fisher's exact test), Q value = 2.6e-12

Table S14. Gene #6: 'NPM1 MUTATION STATUS' versus Clinical Feature #5: 'MIRSEQ_CNMF'

nPatients	CLUS_1	CLUS_2	CLUS_3	CLUS_4
ALL	58	38	42	46
NPM1 MUTATED	37	2	1	12
NPM1 WILD-TYPE	21	36	41	34

Figure S14. Get High-res Image Gene #6: 'NPM1 MUTATION STATUS' versus Clinical Feature #5: 'MIRSEQ_CNMF'

'NPM1 MUTATION STATUS' versus 'MIRSEQ_CHIERARCHICAL'

P value = 1.07e-22 (Fisher's exact test), Q value = 4.5e-21

Table S15. Gene #6: 'NPM1 MUTATION STATUS' versus Clinical Feature #6: 'MIRSEQ_CHIERARCHICAL'

nPatients	CLUS_1	CLUS_2	CLUS_3
ALL	36	80	68
NPM1 MUTATED	0	51	1
NPM1 WILD-TYPE	36	29	67

Figure S15. Get High-res Image Gene #6: 'NPM1 MUTATION STATUS' versus Clinical Feature #6: 'MIRSEQ_CHIERARCHICAL'

Methods & Data

Input

Mutation data file = LAML-TB.mutsig.cluster.txt
Molecular subtypes file = LAML-TB.transferedmergedcluster.txt
Number of patients = 196
Number of significantly mutated genes = 7
Number of Molecular subtypes = 6
Exclude genes that fewer than K tumors have mutations, K = 3

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

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] Greenwood and Nikulin, A guide to chi-squared testing, Wiley, New York. ISBN 047155779X (1996)

[2] 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)

[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