Correlation between aggregated molecular cancer subtypes and selected clinical features

Pheochromocytoma and Paraganglioma (Primary solid tumor)

02 April 2015 | analyses__2015_04_02

Maintainer Information

Citation Information

Maintained by TCGA GDAC Team (Broad Institute/MD Anderson Cancer Center/Harvard Medical School)

Cite as Broad Institute TCGA Genome Data Analysis Center (2015): Correlation between aggregated molecular cancer subtypes and selected clinical features. Broad Institute of MIT and Harvard. doi:10.7908/C1KW5F4C

Overview

Introduction

This pipeline computes the correlation between cancer subtypes identified by different molecular patterns and selected clinical features.

Summary

Testing the association between subtypes identified by 10 different clustering approaches and 5 clinical features across 179 patients, one significant finding detected with P value < 0.05 and Q value < 0.25.

3 subtypes identified in current cancer cohort by 'Copy Number Ratio CNMF subtypes'. These subtypes do not correlate to any clinical features.
3 subtypes identified in current cancer cohort by 'METHLYATION CNMF'. These subtypes do not correlate to any clinical features.
CNMF clustering analysis on RPPA data identified 6 subtypes that do not correlate to any clinical features.
Consensus hierarchical clustering analysis on RPPA data identified 3 subtypes that do not correlate to any clinical features.
CNMF clustering analysis on sequencing-based mRNA expression data identified 4 subtypes that do not correlate to any clinical features.
Consensus hierarchical clustering analysis on sequencing-based mRNA expression data identified 4 subtypes that correlate to 'YEARS_TO_BIRTH'.
3 subtypes identified in current cancer cohort by 'MIRSEQ CNMF'. These subtypes do not correlate to any clinical features.
4 subtypes identified in current cancer cohort by 'MIRSEQ CHIERARCHICAL'. These subtypes do not correlate to any clinical features.
5 subtypes identified in current cancer cohort by 'MIRseq Mature CNMF subtypes'. These subtypes do not correlate to any clinical features.
4 subtypes identified in current cancer cohort by 'MIRseq Mature cHierClus subtypes'. These subtypes do not correlate to any clinical features.

Results

Overview of the results

Table 1. Get Full Table Overview of the association between subtypes identified by 10 different clustering approaches and 5 clinical features. Shown in the table are P values (Q values). Thresholded by P value < 0.05 and Q value < 0.25, one significant finding detected.


Clinical Features	Time to Death	YEARS TO BIRTH	GENDER	RACE	ETHNICITY
Statistical Tests	logrank test	Kruskal-Wallis (anova)	Fisher's exact test	Fisher's exact test	Fisher's exact test
Copy Number Ratio CNMF subtypes	0.551 (1.00)	0.813 (1.00)	0.939 (1.00)	0.418 (0.926)	0.678 (1.00)
METHLYATION CNMF	0.298 (0.926)	0.107 (0.873)	0.206 (0.873)	0.21 (0.873)	0.719 (1.00)
RPPA CNMF subtypes	100 (1.00)	0.591 (1.00)	0.887 (1.00)	0.257 (0.918)	0.357 (0.926)
RPPA cHierClus subtypes	100 (1.00)	0.188 (0.873)	0.134 (0.873)	0.835 (1.00)	1 (1.00)
RNAseq CNMF subtypes	0.826 (1.00)	0.459 (0.926)	0.42 (0.926)	0.238 (0.914)	0.0568 (0.71)
RNAseq cHierClus subtypes	0.814 (1.00)	0.00266 (0.133)	0.496 (0.954)	0.773 (1.00)	0.193 (0.873)
MIRSEQ CNMF	0.17 (0.873)	0.44 (0.926)	1 (1.00)	0.888 (1.00)	0.358 (0.926)
MIRSEQ CHIERARCHICAL	0.65 (1.00)	0.918 (1.00)	0.858 (1.00)	0.667 (1.00)	0.329 (0.926)
MIRseq Mature CNMF subtypes	0.443 (0.926)	0.0109 (0.273)	0.61 (1.00)	0.577 (1.00)	0.437 (0.926)
MIRseq Mature cHierClus subtypes	0.687 (1.00)	0.0453 (0.71)	0.463 (0.926)	0.857 (1.00)	0.114 (0.873)

Clustering Approach #1: 'Copy Number Ratio CNMF subtypes'

Table S1. Description of clustering approach #1: 'Copy Number Ratio CNMF subtypes'

Cluster Labels	1	2	3
Number of samples	60	63	39

'Copy Number Ratio CNMF subtypes' versus 'Time to Death'

P value = 0.551 (logrank test), Q value = 1

Table S2. Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #1: 'Time to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	162	6	0.1 - 316.7 (21.4)
subtype1	60	3	0.7 - 316.7 (22.4)
subtype2	63	1	0.8 - 127.6 (25.3)
subtype3	39	2	0.1 - 137.6 (14.2)

Figure S1. Get High-res Image Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #1: 'Time to Death'

'Copy Number Ratio CNMF subtypes' versus 'YEARS_TO_BIRTH'

P value = 0.813 (Kruskal-Wallis (anova)), Q value = 1

Table S3. Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #2: 'YEARS_TO_BIRTH'

	nPatients	Mean (Std.Dev)
ALL	162	47.5 (15.2)
subtype1	60	48.4 (15.3)
subtype2	63	46.5 (15.6)
subtype3	39	47.7 (15.0)

Figure S2. Get High-res Image Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #2: 'YEARS_TO_BIRTH'

'Copy Number Ratio CNMF subtypes' versus 'GENDER'

P value = 0.939 (Fisher's exact test), Q value = 1

Table S4. Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #3: 'GENDER'

nPatients	FEMALE	MALE
ALL	88	74
subtype1	33	27
subtype2	33	30
subtype3	22	17

Figure S3. Get High-res Image Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #3: 'GENDER'

'Copy Number Ratio CNMF subtypes' versus 'RACE'

P value = 0.418 (Fisher's exact test), Q value = 0.93

Table S5. Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #4: 'RACE'

nPatients	AMERICAN INDIAN OR ALASKA NATIVE	ASIAN	BLACK OR AFRICAN AMERICAN	WHITE
ALL	1	5	19	134
subtype1	1	3	4	51
subtype2	0	2	10	50
subtype3	0	0	5	33

Figure S4. Get High-res Image Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #4: 'RACE'

'Copy Number Ratio CNMF subtypes' versus 'ETHNICITY'

P value = 0.678 (Fisher's exact test), Q value = 1

Table S6. Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #5: 'ETHNICITY'

nPatients	HISPANIC OR LATINO	NOT HISPANIC OR LATINO
ALL	2	128
subtype1	0	51
subtype2	2	51
subtype3	0	26

Figure S5. Get High-res Image Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #5: 'ETHNICITY'

Clustering Approach #2: 'METHLYATION CNMF'

Table S7. Description of clustering approach #2: 'METHLYATION CNMF'

Cluster Labels	1	2	3
Number of samples	45	85	49

'METHLYATION CNMF' versus 'Time to Death'

P value = 0.298 (logrank test), Q value = 0.93

Table S8. Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #1: 'Time to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	179	6	0.1 - 316.7 (20.6)
subtype1	45	3	0.2 - 316.7 (24.8)
subtype2	85	3	0.1 - 127.6 (18.3)
subtype3	49	0	0.7 - 121.0 (25.3)

Figure S6. Get High-res Image Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #1: 'Time to Death'

'METHLYATION CNMF' versus 'YEARS_TO_BIRTH'

P value = 0.107 (Kruskal-Wallis (anova)), Q value = 0.87

Table S9. Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #2: 'YEARS_TO_BIRTH'

	nPatients	Mean (Std.Dev)
ALL	179	47.3 (15.1)
subtype1	45	45.0 (15.0)
subtype2	85	50.0 (14.7)
subtype3	49	44.9 (15.5)

Figure S7. Get High-res Image Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #2: 'YEARS_TO_BIRTH'

'METHLYATION CNMF' versus 'GENDER'

P value = 0.206 (Fisher's exact test), Q value = 0.87

Table S10. Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #3: 'GENDER'

nPatients	FEMALE	MALE
ALL	101	78
subtype1	21	24
subtype2	48	37
subtype3	32	17

Figure S8. Get High-res Image Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #3: 'GENDER'

'METHLYATION CNMF' versus 'RACE'

P value = 0.21 (Fisher's exact test), Q value = 0.87

Table S11. Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #4: 'RACE'

nPatients	AMERICAN INDIAN OR ALASKA NATIVE	ASIAN	BLACK OR AFRICAN AMERICAN	WHITE
ALL	1	6	20	148
subtype1	0	1	7	36
subtype2	1	2	5	74
subtype3	0	3	8	38

Figure S9. Get High-res Image Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #4: 'RACE'

'METHLYATION CNMF' versus 'ETHNICITY'

P value = 0.719 (Fisher's exact test), Q value = 1

Table S12. Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #5: 'ETHNICITY'

nPatients	HISPANIC OR LATINO	NOT HISPANIC OR LATINO
ALL	5	138
subtype1	2	34
subtype2	2	63
subtype3	1	41

Figure S10. Get High-res Image Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #5: 'ETHNICITY'

Clustering Approach #3: 'RPPA CNMF subtypes'

Table S13. Description of clustering approach #3: 'RPPA CNMF subtypes'

Cluster Labels	1	2	3	4	5	6
Number of samples	14	9	20	6	17	13

'RPPA CNMF subtypes' versus 'Time to Death'

P value = 100 (logrank test), Q value = 1

Table S14. Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #1: 'Time to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	79	2	0.7 - 137.6 (23.8)
subtype1	14	0	0.8 - 78.2 (23.4)
subtype2	9	1	0.9 - 137.6 (25.3)
subtype3	20	0	0.7 - 81.7 (21.6)
subtype4	6	0	2.9 - 72.4 (15.4)
subtype5	17	1	2.9 - 108.3 (28.1)
subtype6	13	0	1.8 - 127.6 (20.1)

Figure S11. Get High-res Image Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #1: 'Time to Death'

'RPPA CNMF subtypes' versus 'YEARS_TO_BIRTH'

P value = 0.591 (Kruskal-Wallis (anova)), Q value = 1

Table S15. Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #2: 'YEARS_TO_BIRTH'

	nPatients	Mean (Std.Dev)
ALL	79	47.8 (14.7)
subtype1	14	51.9 (16.2)
subtype2	9	50.2 (8.2)
subtype3	20	50.5 (14.5)
subtype4	6	43.0 (16.7)
subtype5	17	44.5 (11.8)
subtype6	13	43.8 (18.9)

Figure S12. Get High-res Image Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #2: 'YEARS_TO_BIRTH'

'RPPA CNMF subtypes' versus 'GENDER'

P value = 0.887 (Fisher's exact test), Q value = 1

Table S16. Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #3: 'GENDER'

nPatients	FEMALE	MALE
ALL	40	39
subtype1	8	6
subtype2	4	5
subtype3	11	9
subtype4	4	2
subtype5	7	10
subtype6	6	7

Figure S13. Get High-res Image Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #3: 'GENDER'

'RPPA CNMF subtypes' versus 'RACE'

P value = 0.257 (Fisher's exact test), Q value = 0.92

Table S17. Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #4: 'RACE'

nPatients	ASIAN	BLACK OR AFRICAN AMERICAN	WHITE
ALL	4	9	65
subtype1	2	0	11
subtype2	0	0	9
subtype3	0	4	16
subtype4	1	1	4
subtype5	1	3	13
subtype6	0	1	12

Figure S14. Get High-res Image Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #4: 'RACE'

'RPPA CNMF subtypes' versus 'ETHNICITY'

P value = 0.357 (Fisher's exact test), Q value = 0.93

Table S18. Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #5: 'ETHNICITY'

nPatients	HISPANIC OR LATINO	NOT HISPANIC OR LATINO
ALL	1	58
subtype1	1	9
subtype2	0	7
subtype3	0	16
subtype4	0	4
subtype5	0	11
subtype6	0	11

Figure S15. Get High-res Image Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #5: 'ETHNICITY'

Clustering Approach #4: 'RPPA cHierClus subtypes'

Table S19. Description of clustering approach #4: 'RPPA cHierClus subtypes'

Cluster Labels	1	2	3
Number of samples	33	18	28

'RPPA cHierClus subtypes' versus 'Time to Death'

P value = 100 (logrank test), Q value = 1

Table S20. Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #1: 'Time to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	79	2	0.7 - 137.6 (23.8)
subtype1	33	2	0.9 - 137.6 (28.1)
subtype2	18	0	0.8 - 81.7 (20.0)
subtype3	28	0	0.7 - 72.8 (17.7)

Figure S16. Get High-res Image Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #1: 'Time to Death'

'RPPA cHierClus subtypes' versus 'YEARS_TO_BIRTH'

P value = 0.188 (Kruskal-Wallis (anova)), Q value = 0.87

Table S21. Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #2: 'YEARS_TO_BIRTH'

	nPatients	Mean (Std.Dev)
ALL	79	47.8 (14.7)
subtype1	33	50.9 (13.6)
subtype2	18	48.2 (12.9)
subtype3	28	43.8 (16.4)

Figure S17. Get High-res Image Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #2: 'YEARS_TO_BIRTH'

'RPPA cHierClus subtypes' versus 'GENDER'

P value = 0.134 (Fisher's exact test), Q value = 0.87

Table S22. Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #3: 'GENDER'

nPatients	FEMALE	MALE
ALL	40	39
subtype1	20	13
subtype2	10	8
subtype3	10	18

Figure S18. Get High-res Image Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #3: 'GENDER'

'RPPA cHierClus subtypes' versus 'RACE'

P value = 0.835 (Fisher's exact test), Q value = 1

Table S23. Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #4: 'RACE'

nPatients	ASIAN	BLACK OR AFRICAN AMERICAN	WHITE
ALL	4	9	65
subtype1	1	4	28
subtype2	1	1	16
subtype3	2	4	21

Figure S19. Get High-res Image Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #4: 'RACE'

'RPPA cHierClus subtypes' versus 'ETHNICITY'

P value = 1 (Fisher's exact test), Q value = 1

Table S24. Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #5: 'ETHNICITY'

nPatients	HISPANIC OR LATINO	NOT HISPANIC OR LATINO
ALL	1	58
subtype1	1	24
subtype2	0	15
subtype3	0	19

Figure S20. Get High-res Image Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #5: 'ETHNICITY'

Clustering Approach #5: 'RNAseq CNMF subtypes'

Table S25. Description of clustering approach #5: 'RNAseq CNMF subtypes'

Cluster Labels	1	2	3	4
Number of samples	39	69	50	21

'RNAseq CNMF subtypes' versus 'Time to Death'

P value = 0.826 (logrank test), Q value = 1

Table S26. Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #1: 'Time to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	179	6	0.1 - 316.7 (20.6)
subtype1	39	2	0.1 - 316.7 (17.9)
subtype2	69	2	0.1 - 114.8 (20.1)
subtype3	50	2	0.8 - 137.6 (21.0)
subtype4	21	0	0.8 - 111.2 (27.9)

Figure S21. Get High-res Image Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #1: 'Time to Death'

'RNAseq CNMF subtypes' versus 'YEARS_TO_BIRTH'

P value = 0.459 (Kruskal-Wallis (anova)), Q value = 0.93

Table S27. Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #2: 'YEARS_TO_BIRTH'

	nPatients	Mean (Std.Dev)
ALL	179	47.3 (15.1)
subtype1	39	45.1 (15.2)
subtype2	69	49.4 (14.9)
subtype3	50	45.7 (15.4)
subtype4	21	48.5 (15.1)

Figure S22. Get High-res Image Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #2: 'YEARS_TO_BIRTH'

'RNAseq CNMF subtypes' versus 'GENDER'

P value = 0.42 (Fisher's exact test), Q value = 0.93

Table S28. Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #3: 'GENDER'

nPatients	FEMALE	MALE
ALL	101	78
subtype1	25	14
subtype2	35	34
subtype3	27	23
subtype4	14	7

Figure S23. Get High-res Image Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #3: 'GENDER'

'RNAseq CNMF subtypes' versus 'RACE'

P value = 0.238 (Fisher's exact test), Q value = 0.91

Table S29. Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #4: 'RACE'

nPatients	AMERICAN INDIAN OR ALASKA NATIVE	ASIAN	BLACK OR AFRICAN AMERICAN	WHITE
ALL	1	6	20	148
subtype1	0	1	7	28
subtype2	1	2	3	62
subtype3	0	2	6	42
subtype4	0	1	4	16

Figure S24. Get High-res Image Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #4: 'RACE'

'RNAseq CNMF subtypes' versus 'ETHNICITY'

P value = 0.0568 (Fisher's exact test), Q value = 0.71

Table S30. Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #5: 'ETHNICITY'

nPatients	HISPANIC OR LATINO	NOT HISPANIC OR LATINO
ALL	5	138
subtype1	2	26
subtype2	1	56
subtype3	0	40
subtype4	2	16

Figure S25. Get High-res Image Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #5: 'ETHNICITY'

Clustering Approach #6: 'RNAseq cHierClus subtypes'

Table S31. Description of clustering approach #6: 'RNAseq cHierClus subtypes'

Cluster Labels	1	2	3	4
Number of samples	39	71	48	21

'RNAseq cHierClus subtypes' versus 'Time to Death'

P value = 0.814 (logrank test), Q value = 1

Table S32. Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #1: 'Time to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	179	6	0.1 - 316.7 (20.6)
subtype1	39	2	0.8 - 127.6 (26.8)
subtype2	71	2	0.1 - 114.8 (18.3)
subtype3	48	2	0.2 - 316.7 (18.4)
subtype4	21	0	0.8 - 111.2 (27.9)

Figure S26. Get High-res Image Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #1: 'Time to Death'

'RNAseq cHierClus subtypes' versus 'YEARS_TO_BIRTH'

P value = 0.00266 (Kruskal-Wallis (anova)), Q value = 0.13

Table S33. Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #2: 'YEARS_TO_BIRTH'

	nPatients	Mean (Std.Dev)
ALL	179	47.3 (15.1)
subtype1	39	51.9 (12.6)
subtype2	71	49.3 (14.8)
subtype3	48	40.8 (15.6)
subtype4	21	47.1 (15.3)

Figure S27. Get High-res Image Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #2: 'YEARS_TO_BIRTH'

'RNAseq cHierClus subtypes' versus 'GENDER'

P value = 0.496 (Fisher's exact test), Q value = 0.95

Table S34. Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #3: 'GENDER'

nPatients	FEMALE	MALE
ALL	101	78
subtype1	23	16
subtype2	41	30
subtype3	23	25
subtype4	14	7

Figure S28. Get High-res Image Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #3: 'GENDER'

'RNAseq cHierClus subtypes' versus 'RACE'

P value = 0.773 (Fisher's exact test), Q value = 1

Table S35. Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #4: 'RACE'

nPatients	AMERICAN INDIAN OR ALASKA NATIVE	ASIAN	BLACK OR AFRICAN AMERICAN	WHITE
ALL	1	6	20	148
subtype1	0	2	4	33
subtype2	1	2	5	60
subtype3	0	1	8	38
subtype4	0	1	3	17

Figure S29. Get High-res Image Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #4: 'RACE'

'RNAseq cHierClus subtypes' versus 'ETHNICITY'

P value = 0.193 (Fisher's exact test), Q value = 0.87

Table S36. Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #5: 'ETHNICITY'

nPatients	HISPANIC OR LATINO	NOT HISPANIC OR LATINO
ALL	5	138
subtype1	1	29
subtype2	2	55
subtype3	0	38
subtype4	2	16

Figure S30. Get High-res Image Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #5: 'ETHNICITY'

Clustering Approach #7: 'MIRSEQ CNMF'

Table S37. Description of clustering approach #7: 'MIRSEQ CNMF'

Cluster Labels	1	2	3
Number of samples	75	65	39

'MIRSEQ CNMF' versus 'Time to Death'

P value = 0.17 (logrank test), Q value = 0.87

Table S38. Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #1: 'Time to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	179	6	0.1 - 316.7 (20.6)
subtype1	75	5	0.2 - 316.7 (24.2)
subtype2	65	0	0.1 - 114.8 (19.9)
subtype3	39	1	0.7 - 137.6 (23.1)

Figure S31. Get High-res Image Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #1: 'Time to Death'

'MIRSEQ CNMF' versus 'YEARS_TO_BIRTH'

P value = 0.44 (Kruskal-Wallis (anova)), Q value = 0.93

Table S39. Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #2: 'YEARS_TO_BIRTH'

	nPatients	Mean (Std.Dev)
ALL	179	47.3 (15.1)
subtype1	75	45.5 (16.0)
subtype2	65	48.2 (14.7)
subtype3	39	49.4 (14.1)

Figure S32. Get High-res Image Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #2: 'YEARS_TO_BIRTH'

'MIRSEQ CNMF' versus 'GENDER'

P value = 1 (Fisher's exact test), Q value = 1

Table S40. Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #3: 'GENDER'

nPatients	FEMALE	MALE
ALL	101	78
subtype1	42	33
subtype2	37	28
subtype3	22	17

Figure S33. Get High-res Image Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #3: 'GENDER'

'MIRSEQ CNMF' versus 'RACE'

P value = 0.888 (Fisher's exact test), Q value = 1

Table S41. Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #4: 'RACE'

nPatients	AMERICAN INDIAN OR ALASKA NATIVE	ASIAN	BLACK OR AFRICAN AMERICAN	WHITE
ALL	1	6	20	148
subtype1	1	3	9	61
subtype2	0	2	5	55
subtype3	0	1	6	32

Figure S34. Get High-res Image Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #4: 'RACE'

'MIRSEQ CNMF' versus 'ETHNICITY'

P value = 0.358 (Fisher's exact test), Q value = 0.93

Table S42. Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #5: 'ETHNICITY'

nPatients	HISPANIC OR LATINO	NOT HISPANIC OR LATINO
ALL	5	138
subtype1	1	60
subtype2	2	50
subtype3	2	28

Figure S35. Get High-res Image Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #5: 'ETHNICITY'

Clustering Approach #8: 'MIRSEQ CHIERARCHICAL'

Table S43. Description of clustering approach #8: 'MIRSEQ CHIERARCHICAL'

Cluster Labels	1	2	3	4
Number of samples	77	59	18	25

'MIRSEQ CHIERARCHICAL' versus 'Time to Death'

P value = 0.65 (logrank test), Q value = 1

Table S44. Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #1: 'Time to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	179	6	0.1 - 316.7 (20.6)
subtype1	77	4	0.2 - 316.7 (19.6)
subtype2	59	2	0.1 - 114.8 (24.4)
subtype3	18	0	0.7 - 86.9 (27.5)
subtype4	25	0	0.8 - 111.2 (19.9)

Figure S36. Get High-res Image Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #1: 'Time to Death'

'MIRSEQ CHIERARCHICAL' versus 'YEARS_TO_BIRTH'

P value = 0.918 (Kruskal-Wallis (anova)), Q value = 1

Table S45. Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #2: 'YEARS_TO_BIRTH'

	nPatients	Mean (Std.Dev)
ALL	179	47.3 (15.1)
subtype1	77	46.4 (15.7)
subtype2	59	48.6 (15.3)
subtype3	18	47.2 (13.0)
subtype4	25	47.5 (14.9)

Figure S37. Get High-res Image Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #2: 'YEARS_TO_BIRTH'

'MIRSEQ CHIERARCHICAL' versus 'GENDER'

P value = 0.858 (Fisher's exact test), Q value = 1

Table S46. Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #3: 'GENDER'

nPatients	FEMALE	MALE
ALL	101	78
subtype1	42	35
subtype2	35	24
subtype3	9	9
subtype4	15	10

Figure S38. Get High-res Image Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #3: 'GENDER'

'MIRSEQ CHIERARCHICAL' versus 'RACE'

P value = 0.667 (Fisher's exact test), Q value = 1

Table S47. Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #4: 'RACE'

nPatients	AMERICAN INDIAN OR ALASKA NATIVE	ASIAN	BLACK OR AFRICAN AMERICAN	WHITE
ALL	1	6	20	148
subtype1	0	3	10	63
subtype2	1	2	4	49
subtype3	0	0	1	17
subtype4	0	1	5	19

Figure S39. Get High-res Image Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #4: 'RACE'

'MIRSEQ CHIERARCHICAL' versus 'ETHNICITY'

P value = 0.329 (Fisher's exact test), Q value = 0.93

Table S48. Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #5: 'ETHNICITY'

nPatients	HISPANIC OR LATINO	NOT HISPANIC OR LATINO
ALL	5	138
subtype1	1	61
subtype2	2	45
subtype3	0	12
subtype4	2	20

Figure S40. Get High-res Image Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #5: 'ETHNICITY'

Clustering Approach #9: 'MIRseq Mature CNMF subtypes'

Table S49. Description of clustering approach #9: 'MIRseq Mature CNMF subtypes'

Cluster Labels	1	2	3	4	5
Number of samples	36	64	12	42	24

'MIRseq Mature CNMF subtypes' versus 'Time to Death'

P value = 0.443 (logrank test), Q value = 0.93

Table S50. Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #1: 'Time to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	178	6	0.1 - 316.7 (20.4)
subtype1	36	0	0.7 - 127.6 (25.5)
subtype2	64	2	0.1 - 114.8 (19.8)
subtype3	12	1	1.2 - 116.2 (10.5)
subtype4	42	2	0.2 - 121.0 (22.5)
subtype5	24	1	0.9 - 316.7 (26.5)

Figure S41. Get High-res Image Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #1: 'Time to Death'

'MIRseq Mature CNMF subtypes' versus 'YEARS_TO_BIRTH'

P value = 0.0109 (Kruskal-Wallis (anova)), Q value = 0.27

Table S51. Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #2: 'YEARS_TO_BIRTH'

	nPatients	Mean (Std.Dev)
ALL	178	47.2 (15.1)
subtype1	36	51.1 (13.3)
subtype2	64	48.9 (14.6)
subtype3	12	53.8 (13.0)
subtype4	42	42.1 (16.2)
subtype5	24	42.8 (15.3)

Figure S42. Get High-res Image Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #2: 'YEARS_TO_BIRTH'

'MIRseq Mature CNMF subtypes' versus 'GENDER'

P value = 0.61 (Fisher's exact test), Q value = 1

Table S52. Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #3: 'GENDER'

nPatients	FEMALE	MALE
ALL	101	77
subtype1	24	12
subtype2	35	29
subtype3	7	5
subtype4	24	18
subtype5	11	13

Figure S43. Get High-res Image Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #3: 'GENDER'

'MIRseq Mature CNMF subtypes' versus 'RACE'

P value = 0.577 (Fisher's exact test), Q value = 1

Table S53. Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #4: 'RACE'

nPatients	AMERICAN INDIAN OR ALASKA NATIVE	ASIAN	BLACK OR AFRICAN AMERICAN	WHITE
ALL	1	6	20	147
subtype1	0	0	5	31
subtype2	0	3	5	54
subtype3	1	0	1	9
subtype4	0	2	6	33
subtype5	0	1	3	20

Figure S44. Get High-res Image Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #4: 'RACE'

'MIRseq Mature CNMF subtypes' versus 'ETHNICITY'

P value = 0.437 (Fisher's exact test), Q value = 0.93

Table S54. Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #5: 'ETHNICITY'

nPatients	HISPANIC OR LATINO	NOT HISPANIC OR LATINO
ALL	5	137
subtype1	1	25
subtype2	2	52
subtype3	0	8
subtype4	0	33
subtype5	2	19

Figure S45. Get High-res Image Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #5: 'ETHNICITY'

Clustering Approach #10: 'MIRseq Mature cHierClus subtypes'

Table S55. Description of clustering approach #10: 'MIRseq Mature cHierClus subtypes'

Cluster Labels	1	2	3	4
Number of samples	80	61	18	19

'MIRseq Mature cHierClus subtypes' versus 'Time to Death'

P value = 0.687 (logrank test), Q value = 1

Table S56. Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #1: 'Time to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	178	6	0.1 - 316.7 (20.4)
subtype1	80	3	0.1 - 114.8 (19.7)
subtype2	61	3	0.2 - 316.7 (25.3)
subtype3	18	0	0.7 - 86.9 (27.5)
subtype4	19	0	0.8 - 111.2 (20.6)

Figure S46. Get High-res Image Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #1: 'Time to Death'

'MIRseq Mature cHierClus subtypes' versus 'YEARS_TO_BIRTH'

P value = 0.0453 (Kruskal-Wallis (anova)), Q value = 0.71

Table S57. Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #2: 'YEARS_TO_BIRTH'

	nPatients	Mean (Std.Dev)
ALL	178	47.2 (15.1)
subtype1	80	50.3 (14.6)
subtype2	61	43.0 (15.5)
subtype3	18	47.2 (13.0)
subtype4	19	48.2 (15.7)

Figure S47. Get High-res Image Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #2: 'YEARS_TO_BIRTH'

'MIRseq Mature cHierClus subtypes' versus 'GENDER'

P value = 0.463 (Fisher's exact test), Q value = 0.93

Table S58. Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #3: 'GENDER'

nPatients	FEMALE	MALE
ALL	101	77
subtype1	48	32
subtype2	31	30
subtype3	9	9
subtype4	13	6

Figure S48. Get High-res Image Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #3: 'GENDER'

'MIRseq Mature cHierClus subtypes' versus 'RACE'

P value = 0.857 (Fisher's exact test), Q value = 1

Table S59. Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #4: 'RACE'

nPatients	AMERICAN INDIAN OR ALASKA NATIVE	ASIAN	BLACK OR AFRICAN AMERICAN	WHITE
ALL	1	6	20	147
subtype1	1	3	7	66
subtype2	0	2	9	49
subtype3	0	0	1	17
subtype4	0	1	3	15

Figure S49. Get High-res Image Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #4: 'RACE'

'MIRseq Mature cHierClus subtypes' versus 'ETHNICITY'

P value = 0.114 (Fisher's exact test), Q value = 0.87

Table S60. Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #5: 'ETHNICITY'

nPatients	HISPANIC OR LATINO	NOT HISPANIC OR LATINO
ALL	5	137
subtype1	3	62
subtype2	0	49
subtype3	0	12
subtype4	2	14

Figure S50. Get High-res Image Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #5: 'ETHNICITY'

Methods & Data

Input

Cluster data file = /xchip/cga/gdac-prod/tcga-gdac/jobResults/GDAC_mergedClustering/PCPG-TP/15107813/PCPG-TP.mergedcluster.txt
Clinical data file = /xchip/cga/gdac-prod/tcga-gdac/jobResults/Append_Data/PCPG-TP/15087180/PCPG-TP.merged_data.txt
Number of patients = 179
Number of clustering approaches = 10
Number of selected clinical features = 5
Exclude small clusters that include fewer than K patients, K = 3

Clustering approaches

CNMF clustering

consensus non-negative matrix factorization clustering approach (Brunet et al. 2004)

Consensus hierarchical clustering

Resampling-based clustering method (Monti et al. 2003)

Survival analysis

For survival clinical features, the Kaplan-Meier survival curves of tumors with and without gene mutations were plotted and the statistical significance P values were estimated by logrank test (Bland and Altman 2004) using the 'survdiff' function in R

Fisher's exact test

For binary clinical features, 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] Brunet et al., Metagenes and molecular pattern discovery using matrix factorization, PNAS 101(12):4164-9 (2004)

[2] Monti et al., Consensus Clustering: A Resampling-Based Method for Class Discovery and Visualization of Gene Expression Microarray Data, Machine Learning 52(1):91-118 (2003)

[3] Bland and Altman, Statistics notes: The logrank test, BMJ 328(7447):1073 (2004)

[4] 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)

[5] 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