Correlation between aggregated molecular cancer subtypes and selected clinical features

Skin Cutaneous Melanoma (Metastatic)

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/C1PZ57XT

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 14 clinical features across 359 patients, 26 significant findings 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 correlate to 'Time to Death', 'YEARS_TO_BIRTH', and 'PRIMARY_SITE_OF_DISEASE'.
3 subtypes identified in current cancer cohort by 'METHLYATION CNMF'. These subtypes correlate to 'Time from Specimen Diagnosis to Death', 'Time to Death', and 'PRIMARY_SITE_OF_DISEASE'.
CNMF clustering analysis on RPPA data identified 4 subtypes that do not correlate to any clinical features.
Consensus hierarchical clustering analysis on RPPA data identified 4 subtypes that do not correlate to any clinical features.
CNMF clustering analysis on sequencing-based mRNA expression data identified 3 subtypes that correlate to 'Time from Specimen Diagnosis to Death', 'Time to Death', 'YEARS_TO_BIRTH', 'PRIMARY_SITE_OF_DISEASE', and 'BRESLOW_THICKNESS'.
Consensus hierarchical clustering analysis on sequencing-based mRNA expression data identified 4 subtypes that correlate to 'Time from Specimen Diagnosis to Death', 'Time to Death', 'YEARS_TO_BIRTH', 'PRIMARY_SITE_OF_DISEASE', 'PATHOLOGY_T_STAGE', and 'BRESLOW_THICKNESS'.
5 subtypes identified in current cancer cohort by 'MIRSEQ CNMF'. These subtypes correlate to 'Time from Specimen Diagnosis to Death', 'PRIMARY_SITE_OF_DISEASE', 'PATHOLOGY_T_STAGE', and 'MELANOMA_PRIMARY_KNOWN'.
4 subtypes identified in current cancer cohort by 'MIRSEQ CHIERARCHICAL'. These subtypes do not correlate to any clinical features.
3 subtypes identified in current cancer cohort by 'MIRseq Mature CNMF subtypes'. These subtypes correlate to 'PRIMARY_SITE_OF_DISEASE', 'NEOPLASM_DISEASESTAGE', 'PATHOLOGY_T_STAGE', and 'BRESLOW_THICKNESS'.
3 subtypes identified in current cancer cohort by 'MIRseq Mature cHierClus subtypes'. These subtypes correlate to 'PRIMARY_SITE_OF_DISEASE'.

Results

Overview of the results

Table 1. Get Full Table Overview of the association between subtypes identified by 10 different clustering approaches and 14 clinical features. Shown in the table are P values (Q values). Thresholded by P value < 0.05 and Q value < 0.25, 26 significant findings detected.


Clinical Features	Statistical Tests	Copy Number Ratio CNMF subtypes	METHLYATION CNMF	RPPA CNMF subtypes	RPPA cHierClus subtypes	RNAseq CNMF subtypes	RNAseq cHierClus subtypes	MIRSEQ CNMF	MIRSEQ CHIERARCHICAL	MIRseq Mature CNMF subtypes	MIRseq Mature cHierClus subtypes
Time from Specimen Diagnosis to Death	logrank test	0.129 (0.392)	0.00726 (0.0619)	0.063 (0.27)	0.688 (0.854)	0.00011 (0.00192)	2.39e-06 (0.000111)	0.0108 (0.0797)	0.198 (0.501)	0.897 (0.977)	0.494 (0.728)
Time to Death	logrank test	0.00586 (0.0547)	0.0104 (0.0797)	0.117 (0.381)	0.689 (0.854)	2.2e-06 (0.000111)	2.09e-07 (2.93e-05)	0.164 (0.468)	0.31 (0.574)	0.555 (0.79)	0.64 (0.815)
YEARS TO BIRTH	Kruskal-Wallis (anova)	0.0265 (0.159)	0.0552 (0.27)	0.305 (0.574)	0.548 (0.79)	0.00751 (0.0619)	0.00151 (0.0192)	0.232 (0.538)	0.95 (1.00)	0.0566 (0.27)	0.254 (0.565)
PRIMARY SITE OF DISEASE	Fisher's exact test	0.02 (0.128)	0.0125 (0.0876)	0.104 (0.381)	0.287 (0.574)	4e-05 (8e-04)	3e-05 (7e-04)	3e-05 (7e-04)	0.84 (0.963)	0.00223 (0.026)	0.00026 (0.00404)
NEOPLASM DISEASESTAGE	Fisher's exact test	0.821 (0.963)	0.312 (0.574)	0.559 (0.79)	0.299 (0.574)	0.412 (0.671)	0.0801 (0.33)	0.122 (0.381)	0.321 (0.574)	0.0051 (0.051)	0.847 (0.963)
PATHOLOGY T STAGE	Fisher's exact test	0.85 (0.963)	0.0559 (0.27)	0.46 (0.698)	0.302 (0.574)	0.0598 (0.27)	0.00436 (0.047)	0.0182 (0.122)	0.354 (0.612)	0.0289 (0.16)	0.324 (0.574)
PATHOLOGY N STAGE	Fisher's exact test	0.118 (0.381)	0.319 (0.574)	0.305 (0.574)	0.463 (0.698)	0.64 (0.815)	0.621 (0.815)	0.0614 (0.27)	0.179 (0.487)	0.324 (0.574)	0.626 (0.815)
PATHOLOGY M STAGE	Fisher's exact test	0.473 (0.705)	0.715 (0.879)	0.311 (0.574)	0.462 (0.698)	0.43 (0.692)	0.853 (0.963)	0.0904 (0.352)	0.183 (0.487)	0.864 (0.967)	0.387 (0.647)
MELANOMA ULCERATION	Fisher's exact test	0.939 (1.00)	0.088 (0.352)	0.329 (0.575)	0.841 (0.963)	0.402 (0.663)	0.769 (0.921)	0.2 (0.501)	0.283 (0.574)	0.109 (0.381)	0.37 (0.631)
MELANOMA PRIMARY KNOWN	Fisher's exact test	0.958 (1.00)	0.211 (0.515)	0.68 (0.854)	0.132 (0.393)	0.184 (0.487)	0.149 (0.436)	0.0273 (0.159)	0.238 (0.538)	0.594 (0.815)	0.618 (0.815)
BRESLOW THICKNESS	Kruskal-Wallis (anova)	0.638 (0.815)	0.114 (0.381)	0.224 (0.532)	0.882 (0.977)	0.000468 (0.00656)	1.48e-05 (0.000516)	0.168 (0.471)	0.0636 (0.27)	0.0298 (0.16)	0.122 (0.381)
GENDER	Fisher's exact test	0.304 (0.574)	0.449 (0.698)	0.839 (0.963)	0.274 (0.574)	0.32 (0.574)	0.759 (0.917)	0.388 (0.647)	0.612 (0.815)	0.463 (0.698)	0.9 (0.977)
RACE	Fisher's exact test	0.623 (0.815)	0.76 (0.917)			0.505 (0.736)	0.122 (0.381)	0.199 (0.501)	0.237 (0.538)	0.614 (0.815)	0.439 (0.698)
ETHNICITY	Fisher's exact test	0.895 (0.977)	1 (1.00)			0.633 (0.815)	0.213 (0.515)	0.0949 (0.359)	0.95 (1.00)	1 (1.00)	1 (1.00)

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	99	135	124

'Copy Number Ratio CNMF subtypes' versus 'Time from Specimen Diagnosis to Death'

P value = 0.129 (logrank test), Q value = 0.39

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

	nPatients	nDeath	Duration Range (Median), Month
ALL	239	118	0.0 - 346.5 (46.0)
subtype1	62	35	0.0 - 196.0 (45.8)
subtype2	92	40	0.2 - 239.1 (44.2)
subtype3	85	43	0.3 - 346.5 (55.8)

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

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

P value = 0.00586 (logrank test), Q value = 0.055

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

	nPatients	nDeath	Duration Range (Median), Month
ALL	349	180	0.2 - 369.9 (50.9)
subtype1	97	59	0.2 - 368.8 (47.0)
subtype2	132	57	2.0 - 340.1 (54.5)
subtype3	120	64	0.2 - 369.9 (56.0)

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

'Copy Number Ratio CNMF subtypes' versus 'YEARS_TO_BIRTH'

P value = 0.0265 (Kruskal-Wallis (anova)), Q value = 0.16

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

	nPatients	Mean (Std.Dev)
ALL	350	56.2 (15.6)
subtype1	97	59.4 (17.4)
subtype2	133	56.1 (14.6)
subtype3	120	53.8 (15.0)

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

'Copy Number Ratio CNMF subtypes' versus 'PRIMARY_SITE_OF_DISEASE'

P value = 0.02 (Fisher's exact test), Q value = 0.13

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

nPatients	DISTANT METASTASIS	PRIMARY TUMOR	REGIONAL CUTANEOUS OR SUBCUTANEOUS TISSUE	REGIONAL LYMPH NODE
ALL	65	5	72	215
subtype1	26	1	13	59
subtype2	15	1	29	89
subtype3	24	3	30	67

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

'Copy Number Ratio CNMF subtypes' versus 'NEOPLASM_DISEASESTAGE'

P value = 0.821 (Fisher's exact test), Q value = 0.96

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

nPatients	I OR II NOS	STAGE 0	STAGE I	STAGE IA	STAGE IB	STAGE II	STAGE IIA	STAGE IIB	STAGE IIC	STAGE III	STAGE IIIA	STAGE IIIB	STAGE IIIC	STAGE IV
ALL	12	7	28	18	27	24	14	19	12	37	15	32	60	20
subtype1	1	2	6	6	9	4	5	4	4	11	4	11	20	4
subtype2	5	3	15	6	9	10	6	10	2	12	4	9	23	6
subtype3	6	2	7	6	9	10	3	5	6	14	7	12	17	10

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

'Copy Number Ratio CNMF subtypes' versus 'PATHOLOGY_T_STAGE'

P value = 0.85 (Fisher's exact test), Q value = 0.96

Table S7. Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #6: 'PATHOLOGY_T_STAGE'

nPatients	T0+T1	T2	T3	T4
ALL	64	72	77	68
subtype1	20	22	20	21
subtype2	19	25	32	26
subtype3	25	25	25	21

Figure S6. Get High-res Image Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #6: 'PATHOLOGY_T_STAGE'

'Copy Number Ratio CNMF subtypes' versus 'PATHOLOGY_N_STAGE'

P value = 0.118 (Fisher's exact test), Q value = 0.38

Table S8. Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #7: 'PATHOLOGY_N_STAGE'

nPatients	N0	N1	N2	N3
ALL	169	63	38	49
subtype1	43	16	17	12
subtype2	67	18	11	22
subtype3	59	29	10	15

Figure S7. Get High-res Image Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #7: 'PATHOLOGY_N_STAGE'

'Copy Number Ratio CNMF subtypes' versus 'PATHOLOGY_M_STAGE'

P value = 0.473 (Fisher's exact test), Q value = 0.7

Table S9. Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #8: 'PATHOLOGY_M_STAGE'

nPatients	0	1
ALL	313	21
subtype1	89	4
subtype2	116	7
subtype3	108	10

Figure S8. Get High-res Image Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #8: 'PATHOLOGY_M_STAGE'

'Copy Number Ratio CNMF subtypes' versus 'MELANOMA_ULCERATION'

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

Table S10. Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #9: 'MELANOMA_ULCERATION'

nPatients	NO	YES
ALL	132	88
subtype1	40	28
subtype2	50	31
subtype3	42	29

Figure S9. Get High-res Image Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #9: 'MELANOMA_ULCERATION'

'Copy Number Ratio CNMF subtypes' versus 'MELANOMA_PRIMARY_KNOWN'

P value = 0.958 (Fisher's exact test), Q value = 1

Table S11. Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #10: 'MELANOMA_PRIMARY_KNOWN'

nPatients	NO	YES
ALL	44	313
subtype1	13	86
subtype2	16	119
subtype3	15	108

Figure S10. Get High-res Image Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #10: 'MELANOMA_PRIMARY_KNOWN'

'Copy Number Ratio CNMF subtypes' versus 'BRESLOW_THICKNESS'

P value = 0.638 (Kruskal-Wallis (anova)), Q value = 0.82

Table S12. Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #11: 'BRESLOW_THICKNESS'

	nPatients	Mean (Std.Dev)
ALL	262	3.5 (4.7)
subtype1	78	3.3 (3.4)
subtype2	96	3.9 (6.1)
subtype3	88	3.1 (4.0)

Figure S11. Get High-res Image Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #11: 'BRESLOW_THICKNESS'

'Copy Number Ratio CNMF subtypes' versus 'GENDER'

P value = 0.304 (Fisher's exact test), Q value = 0.57

Table S13. Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #12: 'GENDER'

nPatients	FEMALE	MALE
ALL	135	223
subtype1	40	59
subtype2	44	91
subtype3	51	73

Figure S12. Get High-res Image Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #12: 'GENDER'

'Copy Number Ratio CNMF subtypes' versus 'RACE'

P value = 0.623 (Fisher's exact test), Q value = 0.82

Table S14. Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #13: 'RACE'

nPatients	ASIAN	BLACK OR AFRICAN AMERICAN	WHITE
ALL	5	1	344
subtype1	1	0	95
subtype2	1	1	132
subtype3	3	0	117

Figure S13. Get High-res Image Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #13: 'RACE'

'Copy Number Ratio CNMF subtypes' versus 'ETHNICITY'

P value = 0.895 (Fisher's exact test), Q value = 0.98

Table S15. Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #14: 'ETHNICITY'

nPatients	HISPANIC OR LATINO	NOT HISPANIC OR LATINO
ALL	7	344
subtype1	2	95
subtype2	2	132
subtype3	3	117

Figure S14. Get High-res Image Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #14: 'ETHNICITY'

Clustering Approach #2: 'METHLYATION CNMF'

Table S16. Description of clustering approach #2: 'METHLYATION CNMF'

Cluster Labels	1	2	3
Number of samples	124	143	92

'METHLYATION CNMF' versus 'Time from Specimen Diagnosis to Death'

P value = 0.00726 (logrank test), Q value = 0.062

Table S17. Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #1: 'Time from Specimen Diagnosis to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	240	119	0.0 - 346.5 (45.9)
subtype1	85	47	3.1 - 312.1 (57.9)
subtype2	92	32	0.2 - 346.5 (43.0)
subtype3	63	40	0.0 - 209.8 (35.9)

Figure S15. Get High-res Image Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #1: 'Time from Specimen Diagnosis to Death'

'METHLYATION CNMF' versus 'Time to Death'

P value = 0.0104 (logrank test), Q value = 0.08

Table S18. Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #2: 'Time to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	350	181	0.2 - 369.9 (50.8)
subtype1	119	72	0.7 - 357.4 (61.2)
subtype2	140	53	0.2 - 369.9 (49.3)
subtype3	91	56	0.2 - 368.8 (43.4)

Figure S16. Get High-res Image Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #2: 'Time to Death'

'METHLYATION CNMF' versus 'YEARS_TO_BIRTH'

P value = 0.0552 (Kruskal-Wallis (anova)), Q value = 0.27

Table S19. Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #3: 'YEARS_TO_BIRTH'

	nPatients	Mean (Std.Dev)
ALL	351	56.2 (15.6)
subtype1	119	58.6 (15.6)
subtype2	141	54.9 (15.4)
subtype3	91	55.1 (15.7)

Figure S17. Get High-res Image Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #3: 'YEARS_TO_BIRTH'

'METHLYATION CNMF' versus 'PRIMARY_SITE_OF_DISEASE'

P value = 0.0125 (Fisher's exact test), Q value = 0.088

Table S20. Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #4: 'PRIMARY_SITE_OF_DISEASE'

nPatients	DISTANT METASTASIS	PRIMARY TUMOR	REGIONAL CUTANEOUS OR SUBCUTANEOUS TISSUE	REGIONAL LYMPH NODE
ALL	65	5	72	216
subtype1	28	2	25	69
subtype2	15	1	25	101
subtype3	22	2	22	46

Figure S18. Get High-res Image Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #4: 'PRIMARY_SITE_OF_DISEASE'

'METHLYATION CNMF' versus 'NEOPLASM_DISEASESTAGE'

P value = 0.312 (Fisher's exact test), Q value = 0.57

Table S21. Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #5: 'NEOPLASM_DISEASESTAGE'

nPatients	I OR II NOS	STAGE 0	STAGE I	STAGE IA	STAGE IB	STAGE II	STAGE IIA	STAGE IIB	STAGE IIC	STAGE III	STAGE IIIA	STAGE IIIB	STAGE IIIC	STAGE IV
ALL	12	7	28	18	27	24	14	19	12	37	15	32	61	20
subtype1	6	3	9	9	10	6	6	9	6	11	8	8	15	7
subtype2	5	2	13	7	9	8	4	3	1	18	5	14	28	9
subtype3	1	2	6	2	8	10	4	7	5	8	2	10	18	4

Figure S19. Get High-res Image Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #5: 'NEOPLASM_DISEASESTAGE'

'METHLYATION CNMF' versus 'PATHOLOGY_T_STAGE'

P value = 0.0559 (Fisher's exact test), Q value = 0.27

Table S22. Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #6: 'PATHOLOGY_T_STAGE'

nPatients	T0+T1	T2	T3	T4
ALL	64	72	77	69
subtype1	23	23	26	25
subtype2	33	25	31	21
subtype3	8	24	20	23

Figure S20. Get High-res Image Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #6: 'PATHOLOGY_T_STAGE'

'METHLYATION CNMF' versus 'PATHOLOGY_N_STAGE'

P value = 0.319 (Fisher's exact test), Q value = 0.57

Table S23. Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #7: 'PATHOLOGY_N_STAGE'

nPatients	N0	N1	N2	N3
ALL	169	64	38	49
subtype1	65	17	14	14
subtype2	57	29	15	25
subtype3	47	18	9	10

Figure S21. Get High-res Image Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #7: 'PATHOLOGY_N_STAGE'

'METHLYATION CNMF' versus 'PATHOLOGY_M_STAGE'

P value = 0.715 (Fisher's exact test), Q value = 0.88

Table S24. Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #8: 'PATHOLOGY_M_STAGE'

nPatients	0	1
ALL	314	21
subtype1	110	7
subtype2	122	10
subtype3	82	4

Figure S22. Get High-res Image Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #8: 'PATHOLOGY_M_STAGE'

'METHLYATION CNMF' versus 'MELANOMA_ULCERATION'

P value = 0.088 (Fisher's exact test), Q value = 0.35

Table S25. Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #9: 'MELANOMA_ULCERATION'

nPatients	NO	YES
ALL	132	89
subtype1	53	34
subtype2	55	28
subtype3	24	27

Figure S23. Get High-res Image Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #9: 'MELANOMA_ULCERATION'

'METHLYATION CNMF' versus 'MELANOMA_PRIMARY_KNOWN'

P value = 0.211 (Fisher's exact test), Q value = 0.52

Table S26. Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #10: 'MELANOMA_PRIMARY_KNOWN'

nPatients	NO	YES
ALL	44	314
subtype1	13	111
subtype2	23	120
subtype3	8	83

Figure S24. Get High-res Image Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #10: 'MELANOMA_PRIMARY_KNOWN'

'METHLYATION CNMF' versus 'BRESLOW_THICKNESS'

P value = 0.114 (Kruskal-Wallis (anova)), Q value = 0.38

Table S27. Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #11: 'BRESLOW_THICKNESS'

	nPatients	Mean (Std.Dev)
ALL	263	3.5 (4.7)
subtype1	93	3.3 (2.9)
subtype2	99	3.4 (6.0)
subtype3	71	3.8 (4.6)

Figure S25. Get High-res Image Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #11: 'BRESLOW_THICKNESS'

'METHLYATION CNMF' versus 'GENDER'

P value = 0.449 (Fisher's exact test), Q value = 0.7

Table S28. Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #12: 'GENDER'

nPatients	FEMALE	MALE
ALL	135	224
subtype1	42	82
subtype2	54	89
subtype3	39	53

Figure S26. Get High-res Image Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #12: 'GENDER'

'METHLYATION CNMF' versus 'RACE'

P value = 0.76 (Fisher's exact test), Q value = 0.92

Table S29. Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #13: 'RACE'

nPatients	ASIAN	BLACK OR AFRICAN AMERICAN	WHITE
ALL	5	1	345
subtype1	2	0	120
subtype2	1	1	138
subtype3	2	0	87

Figure S27. Get High-res Image Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #13: 'RACE'

'METHLYATION CNMF' versus 'ETHNICITY'

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

Table S30. Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #14: 'ETHNICITY'

nPatients	HISPANIC OR LATINO	NOT HISPANIC OR LATINO
ALL	7	345
subtype1	2	118
subtype2	3	138
subtype3	2	89

Figure S28. Get High-res Image Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #14: 'ETHNICITY'

Clustering Approach #3: 'RPPA CNMF subtypes'

Table S31. Description of clustering approach #3: 'RPPA CNMF subtypes'

Cluster Labels	1	2	3	4
Number of samples	47	58	26	38

'RPPA CNMF subtypes' versus 'Time from Specimen Diagnosis to Death'

P value = 0.063 (logrank test), Q value = 0.27

Table S32. Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #1: 'Time from Specimen Diagnosis to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	121	68	0.0 - 346.5 (49.7)
subtype1	30	13	0.3 - 346.5 (64.5)
subtype2	44	26	4.5 - 312.1 (46.9)
subtype3	20	13	4.4 - 150.1 (44.8)
subtype4	27	16	0.0 - 150.6 (47.7)

Figure S29. Get High-res Image Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #1: 'Time from Specimen Diagnosis to Death'

'RPPA CNMF subtypes' versus 'Time to Death'

P value = 0.117 (logrank test), Q value = 0.38

Table S33. Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #2: 'Time to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	165	95	0.2 - 369.9 (54.5)
subtype1	45	24	9.9 - 369.9 (63.8)
subtype2	58	33	0.2 - 357.4 (58.3)
subtype3	26	15	3.6 - 162.1 (47.0)
subtype4	36	23	2.6 - 176.6 (46.1)

Figure S30. Get High-res Image Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #2: 'Time to Death'

'RPPA CNMF subtypes' versus 'YEARS_TO_BIRTH'

P value = 0.305 (Kruskal-Wallis (anova)), Q value = 0.57

Table S34. Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #3: 'YEARS_TO_BIRTH'

	nPatients	Mean (Std.Dev)
ALL	166	55.3 (15.8)
subtype1	46	51.8 (13.4)
subtype2	58	55.3 (15.7)
subtype3	26	56.6 (17.1)
subtype4	36	58.8 (17.6)

Figure S31. Get High-res Image Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #3: 'YEARS_TO_BIRTH'

'RPPA CNMF subtypes' versus 'PRIMARY_SITE_OF_DISEASE'

P value = 0.104 (Fisher's exact test), Q value = 0.38

Table S35. Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #4: 'PRIMARY_SITE_OF_DISEASE'

nPatients	DISTANT METASTASIS	PRIMARY TUMOR	REGIONAL CUTANEOUS OR SUBCUTANEOUS TISSUE	REGIONAL LYMPH NODE
ALL	22	1	32	113
subtype1	2	0	14	31
subtype2	11	1	10	36
subtype3	4	0	5	17
subtype4	5	0	3	29

Figure S32. Get High-res Image Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #4: 'PRIMARY_SITE_OF_DISEASE'

'RPPA CNMF subtypes' versus 'NEOPLASM_DISEASESTAGE'

P value = 0.559 (Fisher's exact test), Q value = 0.79

Table S36. Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #5: 'NEOPLASM_DISEASESTAGE'

nPatients	I OR II NOS	STAGE 0	STAGE I	STAGE IA	STAGE IB	STAGE II	STAGE IIA	STAGE IIB	STAGE IIC	STAGE III	STAGE IIIA	STAGE IIIB	STAGE IIIC	STAGE IV
ALL	8	3	14	8	17	7	5	6	3	20	5	15	27	9
subtype1	2	1	5	2	7	1	2	1	0	3	3	5	9	2
subtype2	2	2	7	1	7	2	2	2	3	8	1	2	6	5
subtype3	1	0	1	2	1	0	0	1	0	5	1	4	5	0
subtype4	3	0	1	3	2	4	1	2	0	4	0	4	7	2

Figure S33. Get High-res Image Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #5: 'NEOPLASM_DISEASESTAGE'

'RPPA CNMF subtypes' versus 'PATHOLOGY_T_STAGE'

P value = 0.46 (Fisher's exact test), Q value = 0.7

Table S37. Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #6: 'PATHOLOGY_T_STAGE'

nPatients	T0+T1	T2	T3	T4
ALL	36	37	28	31
subtype1	8	12	10	9
subtype2	9	13	7	13
subtype3	7	7	5	2
subtype4	12	5	6	7

Figure S34. Get High-res Image Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #6: 'PATHOLOGY_T_STAGE'

'RPPA CNMF subtypes' versus 'PATHOLOGY_N_STAGE'

P value = 0.305 (Fisher's exact test), Q value = 0.57

Table S38. Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #7: 'PATHOLOGY_N_STAGE'

nPatients	N0	N1	N2	N3
ALL	85	24	26	21
subtype1	22	7	6	8
subtype2	36	6	6	6
subtype3	8	6	7	3
subtype4	19	5	7	4

Figure S35. Get High-res Image Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #7: 'PATHOLOGY_N_STAGE'

'RPPA CNMF subtypes' versus 'PATHOLOGY_M_STAGE'

P value = 0.311 (Fisher's exact test), Q value = 0.57

Table S39. Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #8: 'PATHOLOGY_M_STAGE'

nPatients	0	1
ALL	149	10
subtype1	42	2
subtype2	48	6
subtype3	25	0
subtype4	34	2

Figure S36. Get High-res Image Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #8: 'PATHOLOGY_M_STAGE'

'RPPA CNMF subtypes' versus 'MELANOMA_ULCERATION'

P value = 0.329 (Fisher's exact test), Q value = 0.57

Table S40. Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #9: 'MELANOMA_ULCERATION'

nPatients	NO	YES
ALL	65	37
subtype1	21	7
subtype2	23	15
subtype3	11	5
subtype4	10	10

Figure S37. Get High-res Image Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #9: 'MELANOMA_ULCERATION'

'RPPA CNMF subtypes' versus 'MELANOMA_PRIMARY_KNOWN'

P value = 0.68 (Fisher's exact test), Q value = 0.85

Table S41. Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #10: 'MELANOMA_PRIMARY_KNOWN'

nPatients	NO	YES
ALL	25	144
subtype1	6	41
subtype2	8	50
subtype3	3	23
subtype4	8	30

Figure S38. Get High-res Image Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #10: 'MELANOMA_PRIMARY_KNOWN'

'RPPA CNMF subtypes' versus 'BRESLOW_THICKNESS'

P value = 0.224 (Kruskal-Wallis (anova)), Q value = 0.53

Table S42. Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #11: 'BRESLOW_THICKNESS'

	nPatients	Mean (Std.Dev)
ALL	119	3.4 (5.3)
subtype1	36	3.6 (8.2)
subtype2	40	3.3 (3.0)
subtype3	20	2.6 (3.9)
subtype4	23	3.8 (3.5)

Figure S39. Get High-res Image Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #11: 'BRESLOW_THICKNESS'

'RPPA CNMF subtypes' versus 'GENDER'

P value = 0.839 (Fisher's exact test), Q value = 0.96

Table S43. Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #12: 'GENDER'

nPatients	FEMALE	MALE
ALL	64	105
subtype1	17	30
subtype2	20	38
subtype3	11	15
subtype4	16	22

Figure S40. Get High-res Image Clustering Approach #3: 'RPPA CNMF subtypes' versus Clinical Feature #12: 'GENDER'

Clustering Approach #4: 'RPPA cHierClus subtypes'

Table S44. Description of clustering approach #4: 'RPPA cHierClus subtypes'

Cluster Labels	1	2	3	4
Number of samples	27	58	35	49

'RPPA cHierClus subtypes' versus 'Time from Specimen Diagnosis to Death'

P value = 0.688 (logrank test), Q value = 0.85

Table S45. Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #1: 'Time from Specimen Diagnosis to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	121	68	0.0 - 346.5 (49.7)
subtype1	17	9	0.0 - 214.6 (25.1)
subtype2	44	21	4.5 - 312.1 (48.3)
subtype3	26	17	0.2 - 346.5 (57.8)
subtype4	34	21	4.4 - 227.0 (60.6)

Figure S41. Get High-res Image Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #1: 'Time from Specimen Diagnosis to Death'

'RPPA cHierClus subtypes' versus 'Time to Death'

P value = 0.689 (logrank test), Q value = 0.85

Table S46. Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #2: 'Time to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	165	95	0.2 - 369.9 (54.5)
subtype1	25	15	9.9 - 215.4 (53.3)
subtype2	57	31	6.4 - 357.4 (53.9)
subtype3	34	22	13.9 - 369.9 (61.2)
subtype4	49	27	0.2 - 248.7 (50.9)

Figure S42. Get High-res Image Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #2: 'Time to Death'

'RPPA cHierClus subtypes' versus 'YEARS_TO_BIRTH'

P value = 0.548 (Kruskal-Wallis (anova)), Q value = 0.79

Table S47. Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #3: 'YEARS_TO_BIRTH'

	nPatients	Mean (Std.Dev)
ALL	166	55.3 (15.8)
subtype1	26	58.5 (15.9)
subtype2	57	53.5 (15.2)
subtype3	34	56.4 (15.9)
subtype4	49	54.9 (16.6)

Figure S43. Get High-res Image Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #3: 'YEARS_TO_BIRTH'

'RPPA cHierClus subtypes' versus 'PRIMARY_SITE_OF_DISEASE'

P value = 0.287 (Fisher's exact test), Q value = 0.57

Table S48. Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #4: 'PRIMARY_SITE_OF_DISEASE'

nPatients	DISTANT METASTASIS	PRIMARY TUMOR	REGIONAL CUTANEOUS OR SUBCUTANEOUS TISSUE	REGIONAL LYMPH NODE
ALL	22	1	32	113
subtype1	3	0	8	16
subtype2	9	1	8	40
subtype3	2	0	4	28
subtype4	8	0	12	29

Figure S44. Get High-res Image Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #4: 'PRIMARY_SITE_OF_DISEASE'

'RPPA cHierClus subtypes' versus 'NEOPLASM_DISEASESTAGE'

P value = 0.299 (Fisher's exact test), Q value = 0.57

Table S49. Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #5: 'NEOPLASM_DISEASESTAGE'

nPatients	I OR II NOS	STAGE 0	STAGE I	STAGE IA	STAGE IB	STAGE II	STAGE IIA	STAGE IIB	STAGE IIC	STAGE III	STAGE IIIA	STAGE IIIB	STAGE IIIC	STAGE IV
ALL	8	3	14	8	17	7	5	6	3	20	5	15	27	9
subtype1	1	1	2	1	3	0	1	0	0	1	2	5	7	2
subtype2	2	1	7	3	5	0	3	4	3	10	0	2	9	2
subtype3	4	1	2	1	4	3	1	0	0	3	2	3	5	1
subtype4	1	0	3	3	5	4	0	2	0	6	1	5	6	4

Figure S45. Get High-res Image Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #5: 'NEOPLASM_DISEASESTAGE'

'RPPA cHierClus subtypes' versus 'PATHOLOGY_T_STAGE'

P value = 0.302 (Fisher's exact test), Q value = 0.57

Table S50. Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #6: 'PATHOLOGY_T_STAGE'

nPatients	T0+T1	T2	T3	T4
ALL	36	37	28	31
subtype1	4	6	7	7
subtype2	16	12	4	11
subtype3	6	9	8	3
subtype4	10	10	9	10

Figure S46. Get High-res Image Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #6: 'PATHOLOGY_T_STAGE'

'RPPA cHierClus subtypes' versus 'PATHOLOGY_N_STAGE'

P value = 0.463 (Fisher's exact test), Q value = 0.7

Table S51. Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #7: 'PATHOLOGY_N_STAGE'

nPatients	N0	N1	N2	N3
ALL	85	24	26	21
subtype1	10	8	4	4
subtype2	32	4	9	9
subtype3	18	5	6	3
subtype4	25	7	7	5

Figure S47. Get High-res Image Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #7: 'PATHOLOGY_N_STAGE'

'RPPA cHierClus subtypes' versus 'PATHOLOGY_M_STAGE'

P value = 0.462 (Fisher's exact test), Q value = 0.7

Table S52. Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #8: 'PATHOLOGY_M_STAGE'

nPatients	0	1
ALL	149	10
subtype1	24	2
subtype2	51	2
subtype3	32	1
subtype4	42	5

Figure S48. Get High-res Image Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #8: 'PATHOLOGY_M_STAGE'

'RPPA cHierClus subtypes' versus 'MELANOMA_ULCERATION'

P value = 0.841 (Fisher's exact test), Q value = 0.96

Table S53. Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #9: 'MELANOMA_ULCERATION'

nPatients	NO	YES
ALL	65	37
subtype1	11	6
subtype2	19	14
subtype3	12	5
subtype4	23	12

Figure S49. Get High-res Image Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #9: 'MELANOMA_ULCERATION'

'RPPA cHierClus subtypes' versus 'MELANOMA_PRIMARY_KNOWN'

P value = 0.132 (Fisher's exact test), Q value = 0.39

Table S54. Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #10: 'MELANOMA_PRIMARY_KNOWN'

nPatients	NO	YES
ALL	25	144
subtype1	3	24
subtype2	14	44
subtype3	3	32
subtype4	5	44

Figure S50. Get High-res Image Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #10: 'MELANOMA_PRIMARY_KNOWN'

'RPPA cHierClus subtypes' versus 'BRESLOW_THICKNESS'

P value = 0.882 (Kruskal-Wallis (anova)), Q value = 0.98

Table S55. Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #11: 'BRESLOW_THICKNESS'

	nPatients	Mean (Std.Dev)
ALL	119	3.4 (5.3)
subtype1	21	5.4 (10.7)
subtype2	36	3.1 (3.0)
subtype3	24	2.2 (1.8)
subtype4	38	3.2 (3.5)

Figure S51. Get High-res Image Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #11: 'BRESLOW_THICKNESS'

'RPPA cHierClus subtypes' versus 'GENDER'

P value = 0.274 (Fisher's exact test), Q value = 0.57

Table S56. Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #12: 'GENDER'

nPatients	FEMALE	MALE
ALL	64	105
subtype1	6	21
subtype2	25	33
subtype3	15	20
subtype4	18	31

Figure S52. Get High-res Image Clustering Approach #4: 'RPPA cHierClus subtypes' versus Clinical Feature #12: 'GENDER'

Clustering Approach #5: 'RNAseq CNMF subtypes'

Table S57. Description of clustering approach #5: 'RNAseq CNMF subtypes'

Cluster Labels	1	2	3
Number of samples	133	103	123

'RNAseq CNMF subtypes' versus 'Time from Specimen Diagnosis to Death'

P value = 0.00011 (logrank test), Q value = 0.0019

Table S58. Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #1: 'Time from Specimen Diagnosis to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	240	119	0.0 - 346.5 (45.9)
subtype1	91	54	0.0 - 174.1 (36.5)
subtype2	67	20	0.3 - 209.8 (46.9)
subtype3	82	45	0.2 - 346.5 (58.5)

Figure S53. Get High-res Image Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #1: 'Time from Specimen Diagnosis to Death'

'RNAseq CNMF subtypes' versus 'Time to Death'

P value = 2.2e-06 (logrank test), Q value = 0.00011

Table S59. Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #2: 'Time to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	350	181	0.2 - 369.9 (50.8)
subtype1	130	81	0.7 - 247.0 (43.0)
subtype2	102	33	0.3 - 268.9 (55.2)
subtype3	118	67	0.2 - 369.9 (59.7)

Figure S54. Get High-res Image Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #2: 'Time to Death'

'RNAseq CNMF subtypes' versus 'YEARS_TO_BIRTH'

P value = 0.00751 (Kruskal-Wallis (anova)), Q value = 0.062

Table S60. Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #3: 'YEARS_TO_BIRTH'

	nPatients	Mean (Std.Dev)
ALL	351	56.2 (15.6)
subtype1	130	58.9 (14.9)
subtype2	103	57.0 (15.7)
subtype3	118	52.5 (15.8)

Figure S55. Get High-res Image Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #3: 'YEARS_TO_BIRTH'

'RNAseq CNMF subtypes' versus 'PRIMARY_SITE_OF_DISEASE'

P value = 4e-05 (Fisher's exact test), Q value = 8e-04

Table S61. Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #4: 'PRIMARY_SITE_OF_DISEASE'

nPatients	DISTANT METASTASIS	PRIMARY TUMOR	REGIONAL CUTANEOUS OR SUBCUTANEOUS TISSUE	REGIONAL LYMPH NODE
ALL	65	5	72	216
subtype1	32	5	19	76
subtype2	11	0	14	78
subtype3	22	0	39	62

Figure S56. Get High-res Image Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #4: 'PRIMARY_SITE_OF_DISEASE'

'RNAseq CNMF subtypes' versus 'NEOPLASM_DISEASESTAGE'

P value = 0.412 (Fisher's exact test), Q value = 0.67

Table S62. Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #5: 'NEOPLASM_DISEASESTAGE'

nPatients	I OR II NOS	STAGE 0	STAGE I	STAGE IA	STAGE IB	STAGE II	STAGE IIA	STAGE IIB	STAGE IIC	STAGE III	STAGE IIIA	STAGE IIIB	STAGE IIIC	STAGE IV
ALL	12	7	28	18	27	24	14	19	12	37	15	32	61	20
subtype1	3	3	6	7	10	7	4	9	8	13	8	12	27	6
subtype2	4	1	11	9	7	6	5	4	2	10	1	11	18	4
subtype3	5	3	11	2	10	11	5	6	2	14	6	9	16	10

Figure S57. Get High-res Image Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #5: 'NEOPLASM_DISEASESTAGE'

'RNAseq CNMF subtypes' versus 'PATHOLOGY_T_STAGE'

P value = 0.0598 (Fisher's exact test), Q value = 0.27

Table S63. Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #6: 'PATHOLOGY_T_STAGE'

nPatients	T0+T1	T2	T3	T4
ALL	64	72	77	69
subtype1	19	23	29	35
subtype2	25	17	21	17
subtype3	20	32	27	17

Figure S58. Get High-res Image Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #6: 'PATHOLOGY_T_STAGE'

'RNAseq CNMF subtypes' versus 'PATHOLOGY_N_STAGE'

P value = 0.64 (Fisher's exact test), Q value = 0.82

Table S64. Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #7: 'PATHOLOGY_N_STAGE'

nPatients	N0	N1	N2	N3
ALL	169	64	38	49
subtype1	55	24	17	21
subtype2	53	18	7	13
subtype3	61	22	14	15

Figure S59. Get High-res Image Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #7: 'PATHOLOGY_N_STAGE'

'RNAseq CNMF subtypes' versus 'PATHOLOGY_M_STAGE'

P value = 0.43 (Fisher's exact test), Q value = 0.69

Table S65. Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #8: 'PATHOLOGY_M_STAGE'

nPatients	0	1
ALL	314	21
subtype1	119	7
subtype2	91	4
subtype3	104	10

Figure S60. Get High-res Image Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #8: 'PATHOLOGY_M_STAGE'

'RNAseq CNMF subtypes' versus 'MELANOMA_ULCERATION'

P value = 0.402 (Fisher's exact test), Q value = 0.66

Table S66. Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #9: 'MELANOMA_ULCERATION'

nPatients	NO	YES
ALL	132	89
subtype1	56	40
subtype2	43	22
subtype3	33	27

Figure S61. Get High-res Image Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #9: 'MELANOMA_ULCERATION'

'RNAseq CNMF subtypes' versus 'MELANOMA_PRIMARY_KNOWN'

P value = 0.184 (Fisher's exact test), Q value = 0.49

Table S67. Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #10: 'MELANOMA_PRIMARY_KNOWN'

nPatients	NO	YES
ALL	44	314
subtype1	11	122
subtype2	16	87
subtype3	17	105

Figure S62. Get High-res Image Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #10: 'MELANOMA_PRIMARY_KNOWN'

'RNAseq CNMF subtypes' versus 'BRESLOW_THICKNESS'

P value = 0.000468 (Kruskal-Wallis (anova)), Q value = 0.0066

Table S68. Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #11: 'BRESLOW_THICKNESS'

	nPatients	Mean (Std.Dev)
ALL	263	3.5 (4.7)
subtype1	103	4.3 (4.3)
subtype2	72	2.9 (3.6)
subtype3	88	3.1 (5.7)

Figure S63. Get High-res Image Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #11: 'BRESLOW_THICKNESS'

'RNAseq CNMF subtypes' versus 'GENDER'

P value = 0.32 (Fisher's exact test), Q value = 0.57

Table S69. Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #12: 'GENDER'

nPatients	FEMALE	MALE
ALL	135	224
subtype1	44	89
subtype2	39	64
subtype3	52	71

Figure S64. Get High-res Image Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #12: 'GENDER'

'RNAseq CNMF subtypes' versus 'RACE'

P value = 0.505 (Fisher's exact test), Q value = 0.74

Table S70. Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #13: 'RACE'

nPatients	ASIAN	BLACK OR AFRICAN AMERICAN	WHITE
ALL	5	1	345
subtype1	3	0	126
subtype2	1	1	100
subtype3	1	0	119

Figure S65. Get High-res Image Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #13: 'RACE'

'RNAseq CNMF subtypes' versus 'ETHNICITY'

P value = 0.633 (Fisher's exact test), Q value = 0.82

Table S71. Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #14: 'ETHNICITY'

nPatients	HISPANIC OR LATINO	NOT HISPANIC OR LATINO
ALL	7	345
subtype1	4	127
subtype2	1	101
subtype3	2	117

Figure S66. Get High-res Image Clustering Approach #5: 'RNAseq CNMF subtypes' versus Clinical Feature #14: 'ETHNICITY'

Clustering Approach #6: 'RNAseq cHierClus subtypes'

Table S72. Description of clustering approach #6: 'RNAseq cHierClus subtypes'

Cluster Labels	1	2	3	4
Number of samples	104	104	94	57

'RNAseq cHierClus subtypes' versus 'Time from Specimen Diagnosis to Death'

P value = 2.39e-06 (logrank test), Q value = 0.00011

Table S73. Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #1: 'Time from Specimen Diagnosis to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	240	119	0.0 - 346.5 (45.9)
subtype1	70	49	0.0 - 297.0 (36.4)
subtype2	72	23	0.8 - 239.1 (50.1)
subtype3	60	33	0.2 - 346.5 (59.3)
subtype4	38	14	0.3 - 174.1 (37.0)

Figure S67. Get High-res Image Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #1: 'Time from Specimen Diagnosis to Death'

'RNAseq cHierClus subtypes' versus 'Time to Death'

P value = 2.09e-07 (logrank test), Q value = 2.9e-05

Table S74. Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #2: 'Time to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	350	181	0.2 - 369.9 (50.8)
subtype1	101	72	0.2 - 297.9 (42.8)
subtype2	103	36	0.7 - 340.1 (59.4)
subtype3	90	49	0.2 - 369.9 (59.7)
subtype4	56	24	0.3 - 228.2 (46.1)

Figure S68. Get High-res Image Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #2: 'Time to Death'

'RNAseq cHierClus subtypes' versus 'YEARS_TO_BIRTH'

P value = 0.00151 (Kruskal-Wallis (anova)), Q value = 0.019

Table S75. Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #3: 'YEARS_TO_BIRTH'

	nPatients	Mean (Std.Dev)
ALL	351	56.2 (15.6)
subtype1	101	59.1 (15.0)
subtype2	103	56.1 (16.3)
subtype3	90	50.9 (15.5)
subtype4	57	59.6 (13.6)

Figure S69. Get High-res Image Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #3: 'YEARS_TO_BIRTH'

'RNAseq cHierClus subtypes' versus 'PRIMARY_SITE_OF_DISEASE'

P value = 3e-05 (Fisher's exact test), Q value = 7e-04

Table S76. Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #4: 'PRIMARY_SITE_OF_DISEASE'

nPatients	DISTANT METASTASIS	PRIMARY TUMOR	REGIONAL CUTANEOUS OR SUBCUTANEOUS TISSUE	REGIONAL LYMPH NODE
ALL	65	5	72	216
subtype1	35	4	18	46
subtype2	12	1	17	74
subtype3	15	0	25	54
subtype4	3	0	12	42

Figure S70. Get High-res Image Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #4: 'PRIMARY_SITE_OF_DISEASE'

'RNAseq cHierClus subtypes' versus 'NEOPLASM_DISEASESTAGE'

P value = 0.0801 (Fisher's exact test), Q value = 0.33

Table S77. Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #5: 'NEOPLASM_DISEASESTAGE'

nPatients	I OR II NOS	STAGE 0	STAGE I	STAGE IA	STAGE IB	STAGE II	STAGE IIA	STAGE IIB	STAGE IIC	STAGE III	STAGE IIIA	STAGE IIIB	STAGE IIIC	STAGE IV
ALL	12	7	28	18	27	24	14	19	12	37	15	32	61	20
subtype1	2	2	3	4	7	5	5	9	9	9	7	7	22	5
subtype2	2	0	14	6	9	5	5	4	2	11	3	14	14	6
subtype3	5	3	9	3	7	10	3	2	1	12	5	6	13	5
subtype4	3	2	2	5	4	4	1	4	0	5	0	5	12	4

Figure S71. Get High-res Image Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #5: 'NEOPLASM_DISEASESTAGE'

'RNAseq cHierClus subtypes' versus 'PATHOLOGY_T_STAGE'

P value = 0.00436 (Fisher's exact test), Q value = 0.047

Table S78. Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #6: 'PATHOLOGY_T_STAGE'

nPatients	T0+T1	T2	T3	T4
ALL	64	72	77	69
subtype1	11	16	22	35
subtype2	25	25	21	15
subtype3	17	23	22	10
subtype4	11	8	12	9

Figure S72. Get High-res Image Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #6: 'PATHOLOGY_T_STAGE'

'RNAseq cHierClus subtypes' versus 'PATHOLOGY_N_STAGE'

P value = 0.621 (Fisher's exact test), Q value = 0.82

Table S79. Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #7: 'PATHOLOGY_N_STAGE'

nPatients	N0	N1	N2	N3
ALL	169	64	38	49
subtype1	45	17	13	15
subtype2	49	18	14	10
subtype3	46	20	7	12
subtype4	29	9	4	12

Figure S73. Get High-res Image Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #7: 'PATHOLOGY_N_STAGE'

'RNAseq cHierClus subtypes' versus 'PATHOLOGY_M_STAGE'

P value = 0.853 (Fisher's exact test), Q value = 0.96

Table S80. Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #8: 'PATHOLOGY_M_STAGE'

nPatients	0	1
ALL	314	21
subtype1	90	5
subtype2	90	6
subtype3	82	5
subtype4	52	5

Figure S74. Get High-res Image Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #8: 'PATHOLOGY_M_STAGE'

'RNAseq cHierClus subtypes' versus 'MELANOMA_ULCERATION'

P value = 0.769 (Fisher's exact test), Q value = 0.92

Table S81. Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #9: 'MELANOMA_ULCERATION'

nPatients	NO	YES
ALL	132	89
subtype1	40	32
subtype2	41	25
subtype3	29	16
subtype4	22	16

Figure S75. Get High-res Image Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #9: 'MELANOMA_ULCERATION'

'RNAseq cHierClus subtypes' versus 'MELANOMA_PRIMARY_KNOWN'

P value = 0.149 (Fisher's exact test), Q value = 0.44

Table S82. Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #10: 'MELANOMA_PRIMARY_KNOWN'

nPatients	NO	YES
ALL	44	314
subtype1	9	95
subtype2	18	86
subtype3	13	80
subtype4	4	53

Figure S76. Get High-res Image Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #10: 'MELANOMA_PRIMARY_KNOWN'

'RNAseq cHierClus subtypes' versus 'BRESLOW_THICKNESS'

P value = 1.48e-05 (Kruskal-Wallis (anova)), Q value = 0.00052

Table S83. Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #11: 'BRESLOW_THICKNESS'

	nPatients	Mean (Std.Dev)
ALL	263	3.5 (4.7)
subtype1	81	4.7 (4.5)
subtype2	78	2.7 (3.5)
subtype3	66	3.0 (6.3)
subtype4	38	3.6 (3.6)

Figure S77. Get High-res Image Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #11: 'BRESLOW_THICKNESS'

'RNAseq cHierClus subtypes' versus 'GENDER'

P value = 0.759 (Fisher's exact test), Q value = 0.92

Table S84. Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #12: 'GENDER'

nPatients	FEMALE	MALE
ALL	135	224
subtype1	36	68
subtype2	40	64
subtype3	39	55
subtype4	20	37

Figure S78. Get High-res Image Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #12: 'GENDER'

'RNAseq cHierClus subtypes' versus 'RACE'

P value = 0.122 (Fisher's exact test), Q value = 0.38

Table S85. Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #13: 'RACE'

nPatients	ASIAN	BLACK OR AFRICAN AMERICAN	WHITE
ALL	5	1	345
subtype1	2	0	100
subtype2	0	0	101
subtype3	1	0	91
subtype4	2	1	53

Figure S79. Get High-res Image Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #13: 'RACE'

'RNAseq cHierClus subtypes' versus 'ETHNICITY'

P value = 0.213 (Fisher's exact test), Q value = 0.52

Table S86. Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #14: 'ETHNICITY'

nPatients	HISPANIC OR LATINO	NOT HISPANIC OR LATINO
ALL	7	345
subtype1	4	99
subtype2	0	102
subtype3	2	89
subtype4	1	55

Figure S80. Get High-res Image Clustering Approach #6: 'RNAseq cHierClus subtypes' versus Clinical Feature #14: 'ETHNICITY'

Clustering Approach #7: 'MIRSEQ CNMF'

Table S87. Description of clustering approach #7: 'MIRSEQ CNMF'

Cluster Labels	1	2	3	4	5
Number of samples	46	101	77	111	8

'MIRSEQ CNMF' versus 'Time from Specimen Diagnosis to Death'

P value = 0.0108 (logrank test), Q value = 0.08

Table S88. Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #1: 'Time from Specimen Diagnosis to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	232	118	0.0 - 346.5 (45.9)
subtype1	31	9	0.0 - 312.1 (45.7)
subtype2	63	34	0.2 - 239.1 (46.0)
subtype3	52	23	0.3 - 346.5 (52.5)
subtype4	82	52	0.2 - 209.8 (42.9)
subtype5	4	0	43.5 - 144.5 (55.0)

Figure S81. Get High-res Image Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #1: 'Time from Specimen Diagnosis to Death'

'MIRSEQ CNMF' versus 'Time to Death'

P value = 0.164 (logrank test), Q value = 0.47

Table S89. Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #2: 'Time to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	334	175	0.2 - 369.9 (51.1)
subtype1	45	19	0.2 - 357.4 (42.1)
subtype2	98	52	0.2 - 368.8 (54.6)
subtype3	74	35	7.0 - 369.9 (53.1)
subtype4	110	68	0.3 - 268.7 (50.8)
subtype5	7	1	6.8 - 145.1 (53.9)

Figure S82. Get High-res Image Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #2: 'Time to Death'

'MIRSEQ CNMF' versus 'YEARS_TO_BIRTH'

P value = 0.232 (Kruskal-Wallis (anova)), Q value = 0.54

Table S90. Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #3: 'YEARS_TO_BIRTH'

	nPatients	Mean (Std.Dev)
ALL	335	56.3 (15.6)
subtype1	46	57.9 (16.5)
subtype2	98	54.0 (15.3)
subtype3	74	56.6 (13.7)
subtype4	110	57.9 (16.4)
subtype5	7	48.9 (17.7)

Figure S83. Get High-res Image Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #3: 'YEARS_TO_BIRTH'

'MIRSEQ CNMF' versus 'PRIMARY_SITE_OF_DISEASE'

P value = 3e-05 (Fisher's exact test), Q value = 7e-04

Table S91. Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #4: 'PRIMARY_SITE_OF_DISEASE'

nPatients	DISTANT METASTASIS	PRIMARY TUMOR	REGIONAL CUTANEOUS OR SUBCUTANEOUS TISSUE	REGIONAL LYMPH NODE
ALL	61	4	69	208
subtype1	5	0	9	32
subtype2	18	0	35	48
subtype3	16	4	5	52
subtype4	22	0	17	71
subtype5	0	0	3	5

Figure S84. Get High-res Image Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #4: 'PRIMARY_SITE_OF_DISEASE'

'MIRSEQ CNMF' versus 'NEOPLASM_DISEASESTAGE'

P value = 0.122 (Fisher's exact test), Q value = 0.38

Table S92. Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #5: 'NEOPLASM_DISEASESTAGE'

nPatients	I OR II NOS	STAGE 0	STAGE I	STAGE IA	STAGE IB	STAGE II	STAGE IIA	STAGE IIB	STAGE IIC	STAGE III	STAGE IIIA	STAGE IIIB	STAGE IIIC	STAGE IV
ALL	12	7	27	16	27	22	12	18	11	35	14	30	60	19
subtype1	1	0	2	1	2	6	4	1	2	4	2	7	5	4
subtype2	3	1	9	2	11	3	5	5	2	12	6	8	21	4
subtype3	5	5	9	8	3	5	1	2	3	6	2	4	15	1
subtype4	2	1	7	5	11	8	2	10	4	11	4	10	16	9
subtype5	1	0	0	0	0	0	0	0	0	2	0	1	3	1

Figure S85. Get High-res Image Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #5: 'NEOPLASM_DISEASESTAGE'

'MIRSEQ CNMF' versus 'PATHOLOGY_T_STAGE'

P value = 0.0182 (Fisher's exact test), Q value = 0.12

Table S93. Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #6: 'PATHOLOGY_T_STAGE'

nPatients	T0+T1	T2	T3	T4
ALL	60	72	72	64
subtype1	9	7	12	10
subtype2	8	31	23	17
subtype3	20	11	12	13
subtype4	19	22	25	22
subtype5	4	1	0	2

Figure S86. Get High-res Image Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #6: 'PATHOLOGY_T_STAGE'

'MIRSEQ CNMF' versus 'PATHOLOGY_N_STAGE'

P value = 0.0614 (Fisher's exact test), Q value = 0.27

Table S94. Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #7: 'PATHOLOGY_N_STAGE'

nPatients	N0	N1	N2	N3
ALL	162	60	38	47
subtype1	20	8	9	4
subtype2	44	24	10	13
subtype3	41	9	6	12
subtype4	56	16	13	14
subtype5	1	3	0	4

Figure S87. Get High-res Image Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #7: 'PATHOLOGY_N_STAGE'

'MIRSEQ CNMF' versus 'PATHOLOGY_M_STAGE'

P value = 0.0904 (Fisher's exact test), Q value = 0.35

Table S95. Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #8: 'PATHOLOGY_M_STAGE'

nPatients	0	1
ALL	301	20
subtype1	39	4
subtype2	90	4
subtype3	70	1
subtype4	95	10
subtype5	7	1

Figure S88. Get High-res Image Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #8: 'PATHOLOGY_M_STAGE'

'MIRSEQ CNMF' versus 'MELANOMA_ULCERATION'

P value = 0.2 (Fisher's exact test), Q value = 0.5

Table S96. Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #9: 'MELANOMA_ULCERATION'

nPatients	NO	YES
ALL	125	85
subtype1	12	15
subtype2	38	22
subtype3	29	17
subtype4	46	29
subtype5	0	2

Figure S89. Get High-res Image Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #9: 'MELANOMA_ULCERATION'

'MIRSEQ CNMF' versus 'MELANOMA_PRIMARY_KNOWN'

P value = 0.0273 (Fisher's exact test), Q value = 0.16

Table S97. Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #10: 'MELANOMA_PRIMARY_KNOWN'

nPatients	NO	YES
ALL	42	300
subtype1	7	39
subtype2	10	91
subtype3	7	70
subtype4	14	97
subtype5	4	3

Figure S90. Get High-res Image Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #10: 'MELANOMA_PRIMARY_KNOWN'

'MIRSEQ CNMF' versus 'BRESLOW_THICKNESS'

P value = 0.168 (Kruskal-Wallis (anova)), Q value = 0.47

Table S98. Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #11: 'BRESLOW_THICKNESS'

	nPatients	Mean (Std.Dev)
ALL	250	3.4 (4.8)
subtype1	32	3.4 (3.1)
subtype2	78	3.2 (6.0)
subtype3	52	3.3 (4.5)
subtype4	85	3.7 (4.2)
subtype5	3	4.9 (3.4)

Figure S91. Get High-res Image Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #11: 'BRESLOW_THICKNESS'

'MIRSEQ CNMF' versus 'GENDER'

P value = 0.388 (Fisher's exact test), Q value = 0.65

Table S99. Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #12: 'GENDER'

nPatients	FEMALE	MALE
ALL	128	215
subtype1	17	29
subtype2	32	69
subtype3	32	45
subtype4	42	69
subtype5	5	3

Figure S92. Get High-res Image Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #12: 'GENDER'

'MIRSEQ CNMF' versus 'RACE'

P value = 0.199 (Fisher's exact test), Q value = 0.5

Table S100. Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #13: 'RACE'

nPatients	ASIAN	BLACK OR AFRICAN AMERICAN	WHITE
ALL	5	1	329
subtype1	1	0	45
subtype2	1	0	96
subtype3	3	1	73
subtype4	0	0	107
subtype5	0	0	8

Figure S93. Get High-res Image Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #13: 'RACE'

'MIRSEQ CNMF' versus 'ETHNICITY'

P value = 0.0949 (Fisher's exact test), Q value = 0.36

Table S101. Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #14: 'ETHNICITY'

nPatients	HISPANIC OR LATINO	NOT HISPANIC OR LATINO
ALL	7	329
subtype1	0	46
subtype2	2	95
subtype3	0	76
subtype4	4	105
subtype5	1	7

Figure S94. Get High-res Image Clustering Approach #7: 'MIRSEQ CNMF' versus Clinical Feature #14: 'ETHNICITY'

Clustering Approach #8: 'MIRSEQ CHIERARCHICAL'

Table S102. Description of clustering approach #8: 'MIRSEQ CHIERARCHICAL'

Cluster Labels	1	2	3	4
Number of samples	90	143	72	38

'MIRSEQ CHIERARCHICAL' versus 'Time from Specimen Diagnosis to Death'

P value = 0.198 (logrank test), Q value = 0.5

Table S103. Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #1: 'Time from Specimen Diagnosis to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	232	118	0.0 - 346.5 (45.9)
subtype1	63	38	0.2 - 239.1 (45.5)
subtype2	96	47	0.2 - 297.0 (46.2)
subtype3	48	25	2.0 - 346.5 (51.3)
subtype4	25	8	0.0 - 312.1 (36.3)

Figure S95. Get High-res Image Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #1: 'Time from Specimen Diagnosis to Death'

'MIRSEQ CHIERARCHICAL' versus 'Time to Death'

P value = 0.31 (logrank test), Q value = 0.57

Table S104. Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #2: 'Time to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	334	175	0.2 - 369.9 (51.1)
subtype1	90	53	0.3 - 340.1 (50.4)
subtype2	137	71	0.7 - 297.9 (50.9)
subtype3	70	35	0.2 - 369.9 (60.9)
subtype4	37	16	0.2 - 357.4 (42.1)

Figure S96. Get High-res Image Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #2: 'Time to Death'

'MIRSEQ CHIERARCHICAL' versus 'YEARS_TO_BIRTH'

P value = 0.95 (Kruskal-Wallis (anova)), Q value = 1

Table S105. Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #3: 'YEARS_TO_BIRTH'

	nPatients	Mean (Std.Dev)
ALL	335	56.3 (15.6)
subtype1	90	56.8 (15.2)
subtype2	137	55.8 (16.7)
subtype3	70	56.2 (13.5)
subtype4	38	56.9 (16.5)

Figure S97. Get High-res Image Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #3: 'YEARS_TO_BIRTH'

'MIRSEQ CHIERARCHICAL' versus 'PRIMARY_SITE_OF_DISEASE'

P value = 0.84 (Fisher's exact test), Q value = 0.96

Table S106. Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #4: 'PRIMARY_SITE_OF_DISEASE'

nPatients	DISTANT METASTASIS	PRIMARY TUMOR	REGIONAL CUTANEOUS OR SUBCUTANEOUS TISSUE	REGIONAL LYMPH NODE
ALL	61	4	69	208
subtype1	19	2	16	53
subtype2	23	2	27	90
subtype3	14	0	15	43
subtype4	5	0	11	22

Figure S98. Get High-res Image Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #4: 'PRIMARY_SITE_OF_DISEASE'

'MIRSEQ CHIERARCHICAL' versus 'NEOPLASM_DISEASESTAGE'

P value = 0.321 (Fisher's exact test), Q value = 0.57

Table S107. Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #5: 'NEOPLASM_DISEASESTAGE'

nPatients	I OR II NOS	STAGE 0	STAGE I	STAGE IA	STAGE IB	STAGE II	STAGE IIA	STAGE IIB	STAGE IIC	STAGE III	STAGE IIIA	STAGE IIIB	STAGE IIIC	STAGE IV
ALL	12	7	27	16	27	22	12	18	11	35	14	30	60	19
subtype1	1	4	5	4	8	6	3	5	4	10	5	7	16	4
subtype2	8	0	11	8	14	7	5	10	4	14	5	9	27	10
subtype3	2	3	10	3	4	5	1	2	2	8	3	5	13	1
subtype4	1	0	1	1	1	4	3	1	1	3	1	9	4	4

Figure S99. Get High-res Image Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #5: 'NEOPLASM_DISEASESTAGE'

'MIRSEQ CHIERARCHICAL' versus 'PATHOLOGY_T_STAGE'

P value = 0.354 (Fisher's exact test), Q value = 0.61

Table S108. Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #6: 'PATHOLOGY_T_STAGE'

nPatients	T0+T1	T2	T3	T4
ALL	60	72	72	64
subtype1	15	17	24	18
subtype2	21	37	25	27
subtype3	16	14	12	11
subtype4	8	4	11	8

Figure S100. Get High-res Image Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #6: 'PATHOLOGY_T_STAGE'

'MIRSEQ CHIERARCHICAL' versus 'PATHOLOGY_N_STAGE'

P value = 0.179 (Fisher's exact test), Q value = 0.49

Table S109. Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #7: 'PATHOLOGY_N_STAGE'

nPatients	N0	N1	N2	N3
ALL	162	60	38	47
subtype1	39	18	12	10
subtype2	73	19	15	22
subtype3	36	15	3	11
subtype4	14	8	8	4

Figure S101. Get High-res Image Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #7: 'PATHOLOGY_N_STAGE'

'MIRSEQ CHIERARCHICAL' versus 'PATHOLOGY_M_STAGE'

P value = 0.183 (Fisher's exact test), Q value = 0.49

Table S110. Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #8: 'PATHOLOGY_M_STAGE'

nPatients	0	1
ALL	301	20
subtype1	79	5
subtype2	125	10
subtype3	65	1
subtype4	32	4

Figure S102. Get High-res Image Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #8: 'PATHOLOGY_M_STAGE'

'MIRSEQ CHIERARCHICAL' versus 'MELANOMA_ULCERATION'

P value = 0.283 (Fisher's exact test), Q value = 0.57

Table S111. Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #9: 'MELANOMA_ULCERATION'

nPatients	NO	YES
ALL	125	85
subtype1	40	25
subtype2	56	31
subtype3	19	16
subtype4	10	13

Figure S103. Get High-res Image Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #9: 'MELANOMA_ULCERATION'

'MIRSEQ CHIERARCHICAL' versus 'MELANOMA_PRIMARY_KNOWN'

P value = 0.238 (Fisher's exact test), Q value = 0.54

Table S112. Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #10: 'MELANOMA_PRIMARY_KNOWN'

nPatients	NO	YES
ALL	42	300
subtype1	6	84
subtype2	19	123
subtype3	11	61
subtype4	6	32

Figure S104. Get High-res Image Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #10: 'MELANOMA_PRIMARY_KNOWN'

'MIRSEQ CHIERARCHICAL' versus 'BRESLOW_THICKNESS'

P value = 0.0636 (Kruskal-Wallis (anova)), Q value = 0.27

Table S113. Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #11: 'BRESLOW_THICKNESS'

	nPatients	Mean (Std.Dev)
ALL	250	3.4 (4.8)
subtype1	69	3.3 (3.0)
subtype2	106	3.4 (4.1)
subtype3	49	3.8 (8.0)
subtype4	26	3.2 (2.8)

Figure S105. Get High-res Image Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #11: 'BRESLOW_THICKNESS'

'MIRSEQ CHIERARCHICAL' versus 'GENDER'

P value = 0.612 (Fisher's exact test), Q value = 0.82

Table S114. Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #12: 'GENDER'

nPatients	FEMALE	MALE
ALL	128	215
subtype1	30	60
subtype2	58	85
subtype3	28	44
subtype4	12	26

Figure S106. Get High-res Image Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #12: 'GENDER'

'MIRSEQ CHIERARCHICAL' versus 'RACE'

P value = 0.237 (Fisher's exact test), Q value = 0.54

Table S115. Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #13: 'RACE'

nPatients	ASIAN	BLACK OR AFRICAN AMERICAN	WHITE
ALL	5	1	329
subtype1	1	1	85
subtype2	1	0	139
subtype3	1	0	69
subtype4	2	0	36

Figure S107. Get High-res Image Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #13: 'RACE'

'MIRSEQ CHIERARCHICAL' versus 'ETHNICITY'

P value = 0.95 (Fisher's exact test), Q value = 1

Table S116. Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #14: 'ETHNICITY'

nPatients	HISPANIC OR LATINO	NOT HISPANIC OR LATINO
ALL	7	329
subtype1	2	86
subtype2	4	139
subtype3	1	67
subtype4	0	37

Figure S108. Get High-res Image Clustering Approach #8: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #14: 'ETHNICITY'

Clustering Approach #9: 'MIRseq Mature CNMF subtypes'

Table S117. Description of clustering approach #9: 'MIRseq Mature CNMF subtypes'

Cluster Labels	1	2	3
Number of samples	92	104	133

'MIRseq Mature CNMF subtypes' versus 'Time from Specimen Diagnosis to Death'

P value = 0.897 (logrank test), Q value = 0.98

Table S118. Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #1: 'Time from Specimen Diagnosis to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	222	116	0.0 - 346.5 (47.5)
subtype1	57	24	1.1 - 185.1 (35.3)
subtype2	70	38	0.0 - 312.1 (53.3)
subtype3	95	54	0.2 - 346.5 (47.3)

Figure S109. Get High-res Image Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #1: 'Time from Specimen Diagnosis to Death'

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

P value = 0.555 (logrank test), Q value = 0.79

Table S119. Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #2: 'Time to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	320	170	0.2 - 369.9 (52.6)
subtype1	86	38	3.6 - 268.9 (46.6)
subtype2	102	55	0.2 - 368.8 (60.4)
subtype3	132	77	0.2 - 369.9 (49.8)

Figure S110. Get High-res Image Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #2: 'Time to Death'

'MIRseq Mature CNMF subtypes' versus 'YEARS_TO_BIRTH'

P value = 0.0566 (Kruskal-Wallis (anova)), Q value = 0.27

Table S120. Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #3: 'YEARS_TO_BIRTH'

	nPatients	Mean (Std.Dev)
ALL	321	56.4 (15.7)
subtype1	87	58.1 (14.3)
subtype2	102	53.2 (16.5)
subtype3	132	57.8 (15.7)

Figure S111. Get High-res Image Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #3: 'YEARS_TO_BIRTH'

'MIRseq Mature CNMF subtypes' versus 'PRIMARY_SITE_OF_DISEASE'

P value = 0.00223 (Fisher's exact test), Q value = 0.026

Table S121. Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #4: 'PRIMARY_SITE_OF_DISEASE'

nPatients	DISTANT METASTASIS	PRIMARY TUMOR	REGIONAL CUTANEOUS OR SUBCUTANEOUS TISSUE	REGIONAL LYMPH NODE
ALL	60	4	66	198
subtype1	11	0	12	68
subtype2	21	0	31	52
subtype3	28	4	23	78

Figure S112. Get High-res Image Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #4: 'PRIMARY_SITE_OF_DISEASE'

'MIRseq Mature CNMF subtypes' versus 'NEOPLASM_DISEASESTAGE'

P value = 0.0051 (Fisher's exact test), Q value = 0.051

Table S122. Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #5: 'NEOPLASM_DISEASESTAGE'

nPatients	I OR II NOS	STAGE 0	STAGE I	STAGE IA	STAGE IB	STAGE II	STAGE IIA	STAGE IIB	STAGE IIC	STAGE III	STAGE IIIA	STAGE IIIB	STAGE IIIC	STAGE IV
ALL	12	7	27	16	26	22	12	17	11	35	13	28	55	17
subtype1	7	2	5	10	4	4	3	0	0	13	1	9	21	3
subtype2	2	2	11	2	12	6	6	5	3	9	6	7	16	6
subtype3	3	3	11	4	10	12	3	12	8	13	6	12	18	8

Figure S113. Get High-res Image Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #5: 'NEOPLASM_DISEASESTAGE'

'MIRseq Mature CNMF subtypes' versus 'PATHOLOGY_T_STAGE'

P value = 0.0289 (Fisher's exact test), Q value = 0.16

Table S123. Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #6: 'PATHOLOGY_T_STAGE'

nPatients	T0+T1	T2	T3	T4
ALL	58	71	69	59
subtype1	26	13	17	13
subtype2	14	29	21	17
subtype3	18	29	31	29

Figure S114. Get High-res Image Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #6: 'PATHOLOGY_T_STAGE'

'MIRseq Mature CNMF subtypes' versus 'PATHOLOGY_N_STAGE'

P value = 0.324 (Fisher's exact test), Q value = 0.57

Table S124. Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #7: 'PATHOLOGY_N_STAGE'

nPatients	N0	N1	N2	N3
ALL	160	59	34	42
subtype1	40	14	12	18
subtype2	53	18	11	11
subtype3	67	27	11	13

Figure S115. Get High-res Image Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #7: 'PATHOLOGY_N_STAGE'

'MIRseq Mature CNMF subtypes' versus 'PATHOLOGY_M_STAGE'

P value = 0.864 (Fisher's exact test), Q value = 0.97

Table S125. Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #8: 'PATHOLOGY_M_STAGE'

nPatients	0	1
ALL	291	18
subtype1	83	4
subtype2	89	6
subtype3	119	8

Figure S116. Get High-res Image Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #8: 'PATHOLOGY_M_STAGE'

'MIRseq Mature CNMF subtypes' versus 'MELANOMA_ULCERATION'

P value = 0.109 (Fisher's exact test), Q value = 0.38

Table S126. Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #9: 'MELANOMA_ULCERATION'

nPatients	NO	YES
ALL	120	80
subtype1	38	16
subtype2	37	23
subtype3	45	41

Figure S117. Get High-res Image Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #9: 'MELANOMA_ULCERATION'

'MIRseq Mature CNMF subtypes' versus 'MELANOMA_PRIMARY_KNOWN'

P value = 0.594 (Fisher's exact test), Q value = 0.82

Table S127. Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #10: 'MELANOMA_PRIMARY_KNOWN'

nPatients	NO	YES
ALL	38	291
subtype1	13	79
subtype2	12	92
subtype3	13	120

Figure S118. Get High-res Image Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #10: 'MELANOMA_PRIMARY_KNOWN'

'MIRseq Mature CNMF subtypes' versus 'BRESLOW_THICKNESS'

P value = 0.0298 (Kruskal-Wallis (anova)), Q value = 0.16

Table S128. Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #11: 'BRESLOW_THICKNESS'

	nPatients	Mean (Std.Dev)
ALL	241	3.4 (4.8)
subtype1	57	3.3 (4.4)
subtype2	82	3.3 (6.1)
subtype3	102	3.5 (3.7)

Figure S119. Get High-res Image Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #11: 'BRESLOW_THICKNESS'

'MIRseq Mature CNMF subtypes' versus 'GENDER'

P value = 0.463 (Fisher's exact test), Q value = 0.7

Table S129. Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #12: 'GENDER'

nPatients	FEMALE	MALE
ALL	120	209
subtype1	38	54
subtype2	34	70
subtype3	48	85

Figure S120. Get High-res Image Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #12: 'GENDER'

'MIRseq Mature CNMF subtypes' versus 'RACE'

P value = 0.614 (Fisher's exact test), Q value = 0.82

Table S130. Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #13: 'RACE'

nPatients	ASIAN	BLACK OR AFRICAN AMERICAN	WHITE
ALL	5	1	316
subtype1	1	1	89
subtype2	1	0	99
subtype3	3	0	128

Figure S121. Get High-res Image Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #13: 'RACE'

'MIRseq Mature CNMF subtypes' versus 'ETHNICITY'

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

Table S131. Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #14: 'ETHNICITY'

nPatients	HISPANIC OR LATINO	NOT HISPANIC OR LATINO
ALL	7	316
subtype1	2	89
subtype2	2	98
subtype3	3	129

Figure S122. Get High-res Image Clustering Approach #9: 'MIRseq Mature CNMF subtypes' versus Clinical Feature #14: 'ETHNICITY'

Clustering Approach #10: 'MIRseq Mature cHierClus subtypes'

Table S132. Description of clustering approach #10: 'MIRseq Mature cHierClus subtypes'

Cluster Labels	1	2	3
Number of samples	163	136	30

'MIRseq Mature cHierClus subtypes' versus 'Time from Specimen Diagnosis to Death'

P value = 0.494 (logrank test), Q value = 0.73

Table S133. Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #1: 'Time from Specimen Diagnosis to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	222	116	0.0 - 346.5 (47.5)
subtype1	115	60	0.2 - 297.0 (42.5)
subtype2	87	49	0.2 - 346.5 (58.1)
subtype3	20	7	0.0 - 312.1 (41.0)

Figure S123. Get High-res Image Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #1: 'Time from Specimen Diagnosis to Death'

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

P value = 0.64 (logrank test), Q value = 0.82

Table S134. Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #2: 'Time to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	320	170	0.2 - 369.9 (52.6)
subtype1	160	85	0.3 - 297.9 (48.8)
subtype2	130	71	0.2 - 369.9 (60.6)
subtype3	30	14	0.2 - 357.4 (43.7)

Figure S124. Get High-res Image Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #2: 'Time to Death'

'MIRseq Mature cHierClus subtypes' versus 'YEARS_TO_BIRTH'

P value = 0.254 (Kruskal-Wallis (anova)), Q value = 0.56

Table S135. Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #3: 'YEARS_TO_BIRTH'

	nPatients	Mean (Std.Dev)
ALL	321	56.4 (15.7)
subtype1	161	57.6 (15.6)
subtype2	130	54.7 (15.4)
subtype3	30	57.3 (17.5)

Figure S125. Get High-res Image Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #3: 'YEARS_TO_BIRTH'

'MIRseq Mature cHierClus subtypes' versus 'PRIMARY_SITE_OF_DISEASE'

P value = 0.00026 (Fisher's exact test), Q value = 0.004

Table S136. Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #4: 'PRIMARY_SITE_OF_DISEASE'

nPatients	DISTANT METASTASIS	PRIMARY TUMOR	REGIONAL CUTANEOUS OR SUBCUTANEOUS TISSUE	REGIONAL LYMPH NODE
ALL	60	4	66	198
subtype1	32	4	17	109
subtype2	23	0	43	70
subtype3	5	0	6	19

Figure S126. Get High-res Image Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #4: 'PRIMARY_SITE_OF_DISEASE'

'MIRseq Mature cHierClus subtypes' versus 'NEOPLASM_DISEASESTAGE'

P value = 0.847 (Fisher's exact test), Q value = 0.96

Table S137. Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #5: 'NEOPLASM_DISEASESTAGE'

nPatients	I OR II NOS	STAGE 0	STAGE I	STAGE IA	STAGE IB	STAGE II	STAGE IIA	STAGE IIB	STAGE IIC	STAGE III	STAGE IIIA	STAGE IIIB	STAGE IIIC	STAGE IV
ALL	12	7	27	16	26	22	12	17	11	35	13	28	55	17
subtype1	5	4	11	11	13	12	5	10	7	18	6	13	27	8
subtype2	6	3	15	4	12	6	4	6	3	15	6	11	25	6
subtype3	1	0	1	1	1	4	3	1	1	2	1	4	3	3

Figure S127. Get High-res Image Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #5: 'NEOPLASM_DISEASESTAGE'

'MIRseq Mature cHierClus subtypes' versus 'PATHOLOGY_T_STAGE'

P value = 0.324 (Fisher's exact test), Q value = 0.57

Table S138. Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #6: 'PATHOLOGY_T_STAGE'

nPatients	T0+T1	T2	T3	T4
ALL	58	71	69	59
subtype1	31	32	34	34
subtype2	21	36	27	19
subtype3	6	3	8	6

Figure S128. Get High-res Image Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #6: 'PATHOLOGY_T_STAGE'

'MIRseq Mature cHierClus subtypes' versus 'PATHOLOGY_N_STAGE'

P value = 0.626 (Fisher's exact test), Q value = 0.82

Table S139. Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #7: 'PATHOLOGY_N_STAGE'

nPatients	N0	N1	N2	N3
ALL	160	59	34	42
subtype1	81	28	16	21
subtype2	65	27	12	19
subtype3	14	4	6	2

Figure S129. Get High-res Image Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #7: 'PATHOLOGY_N_STAGE'

'MIRseq Mature cHierClus subtypes' versus 'PATHOLOGY_M_STAGE'

P value = 0.387 (Fisher's exact test), Q value = 0.65

Table S140. Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #8: 'PATHOLOGY_M_STAGE'

nPatients	0	1
ALL	291	18
subtype1	145	9
subtype2	121	6
subtype3	25	3

Figure S130. Get High-res Image Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #8: 'PATHOLOGY_M_STAGE'

'MIRseq Mature cHierClus subtypes' versus 'MELANOMA_ULCERATION'

P value = 0.37 (Fisher's exact test), Q value = 0.63

Table S141. Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #9: 'MELANOMA_ULCERATION'

nPatients	NO	YES
ALL	120	80
subtype1	65	41
subtype2	47	29
subtype3	8	10

Figure S131. Get High-res Image Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #9: 'MELANOMA_ULCERATION'

'MIRseq Mature cHierClus subtypes' versus 'MELANOMA_PRIMARY_KNOWN'

P value = 0.618 (Fisher's exact test), Q value = 0.82

Table S142. Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #10: 'MELANOMA_PRIMARY_KNOWN'

nPatients	NO	YES
ALL	38	291
subtype1	18	145
subtype2	15	121
subtype3	5	25

Figure S132. Get High-res Image Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #10: 'MELANOMA_PRIMARY_KNOWN'

'MIRseq Mature cHierClus subtypes' versus 'BRESLOW_THICKNESS'

P value = 0.122 (Kruskal-Wallis (anova)), Q value = 0.38

Table S143. Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #11: 'BRESLOW_THICKNESS'

	nPatients	Mean (Std.Dev)
ALL	241	3.4 (4.8)
subtype1	121	3.6 (4.2)
subtype2	100	3.1 (5.7)
subtype3	20	3.3 (3.1)

Figure S133. Get High-res Image Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #11: 'BRESLOW_THICKNESS'

'MIRseq Mature cHierClus subtypes' versus 'GENDER'

P value = 0.9 (Fisher's exact test), Q value = 0.98

Table S144. Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #12: 'GENDER'

nPatients	FEMALE	MALE
ALL	120	209
subtype1	59	104
subtype2	49	87
subtype3	12	18

Figure S134. Get High-res Image Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #12: 'GENDER'

'MIRseq Mature cHierClus subtypes' versus 'RACE'

P value = 0.439 (Fisher's exact test), Q value = 0.7

Table S145. Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #13: 'RACE'

nPatients	ASIAN	BLACK OR AFRICAN AMERICAN	WHITE
ALL	5	1	316
subtype1	3	1	156
subtype2	1	0	131
subtype3	1	0	29

Figure S135. Get High-res Image Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #13: 'RACE'

'MIRseq Mature cHierClus subtypes' versus 'ETHNICITY'

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

Table S146. Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #14: 'ETHNICITY'

nPatients	HISPANIC OR LATINO	NOT HISPANIC OR LATINO
ALL	7	316
subtype1	4	157
subtype2	3	129
subtype3	0	30

Figure S136. Get High-res Image Clustering Approach #10: 'MIRseq Mature cHierClus subtypes' versus Clinical Feature #14: 'ETHNICITY'

Methods & Data

Input

Cluster data file = /xchip/cga/gdac-prod/tcga-gdac/jobResults/GDAC_mergedClustering/SKCM-TM/15111349/SKCM-TM.mergedcluster.txt
Clinical data file = /xchip/cga/gdac-prod/tcga-gdac/jobResults/Append_Data/SKCM-TM/15087681/SKCM-TM.merged_data.txt
Number of patients = 359
Number of clustering approaches = 10
Number of selected clinical features = 14
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