Correlation between molecular cancer subtypes and selected clinical features

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 6 different clustering approaches and 6 clinical features across 32 patients, no significant finding detected with P value < 0.05 and Q value < 0.25.

3 subtypes identified in current cancer cohort by 'CN CNMF'. These subtypes do not correlate to any clinical features.
4 subtypes identified in current cancer cohort by 'METHLYATION CNMF'. These subtypes do not correlate to any clinical features.
CNMF clustering analysis on sequencing-based mRNA expression data identified 3 subtypes that do not correlate to any clinical features.
Consensus hierarchical clustering analysis on sequencing-based mRNA expression data identified 3 subtypes that do not correlate to any clinical features.
CNMF clustering analysis on sequencing-based miR expression data identified 3 subtypes that do not correlate to any clinical features.
Consensus hierarchical clustering analysis on sequencing-based miR expression data identified 3 subtypes that 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 6 different clustering approaches and 6 clinical features. Shown in the table are P values (Q values). Thresholded by P value < 0.05 and Q value < 0.25, no significant finding detected.


Clinical Features	Time to Death	AGE	RADIATIONS RADIATION REGIMENINDICATION	NUMBERPACKYEARSSMOKED	STOPPEDSMOKINGYEAR	TOBACCOSMOKINGHISTORYINDICATOR
Statistical Tests	logrank test	ANOVA	Fisher's exact test	ANOVA	ANOVA	ANOVA
CN CNMF	100 (1.00)	0.982 (1.00)	1 (1.00)			0.524 (1.00)
METHLYATION CNMF	0.0604 (1.00)	0.817 (1.00)	0.789 (1.00)			0.834 (1.00)
RNAseq CNMF subtypes	100 (1.00)	0.837 (1.00)	0.0228 (0.548)			0.966 (1.00)
RNAseq cHierClus subtypes	100 (1.00)	0.617 (1.00)	1 (1.00)			0.637 (1.00)
MIRseq CNMF subtypes	0.956 (1.00)	0.42 (1.00)	1 (1.00)			0.862 (1.00)
MIRseq cHierClus subtypes	0.183 (1.00)	0.259 (1.00)	0.856 (1.00)			0.975 (1.00)

Clustering Approach #1: 'CN CNMF'

Table S1. Get Full Table Description of clustering approach #1: 'CN CNMF'

Cluster Labels	1	2	3
Number of samples	11	3	12

'CN CNMF' versus 'Time to Death'

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

Table S2. Clustering Approach #1: 'CN CNMF' versus Clinical Feature #1: 'Time to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	25	1	0.1 - 95.1 (2.7)
subtype1	10	0	0.3 - 49.5 (2.2)
subtype2	3	0	1.4 - 70.8 (69.9)
subtype3	12	1	0.1 - 95.1 (3.8)

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

'CN CNMF' versus 'AGE'

P value = 0.982 (ANOVA), Q value = 1

Table S3. Clustering Approach #1: 'CN CNMF' versus Clinical Feature #2: 'AGE'

	nPatients	Mean (Std.Dev)
ALL	26	48.5 (11.8)
subtype1	11	48.5 (11.0)
subtype2	3	49.7 (4.2)
subtype3	12	48.2 (14.3)

Figure S2. Get High-res Image Clustering Approach #1: 'CN CNMF' versus Clinical Feature #2: 'AGE'

'CN CNMF' versus 'RADIATIONS.RADIATION.REGIMENINDICATION'

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

Table S4. Clustering Approach #1: 'CN CNMF' versus Clinical Feature #3: 'RADIATIONS.RADIATION.REGIMENINDICATION'

nPatients	NO	YES
ALL	4	22
subtype1	2	9
subtype2	0	3
subtype3	2	10

Figure S3. Get High-res Image Clustering Approach #1: 'CN CNMF' versus Clinical Feature #3: 'RADIATIONS.RADIATION.REGIMENINDICATION'

'CN CNMF' versus 'TOBACCOSMOKINGHISTORYINDICATOR'

P value = 0.524 (ANOVA), Q value = 1

Table S5. Clustering Approach #1: 'CN CNMF' versus Clinical Feature #6: 'TOBACCOSMOKINGHISTORYINDICATOR'

	nPatients	Mean (Std.Dev)
ALL	25	1.8 (1.1)
subtype1	10	1.7 (1.1)
subtype2	3	1.3 (0.6)
subtype3	12	2.1 (1.2)

Figure S4. Get High-res Image Clustering Approach #1: 'CN CNMF' versus Clinical Feature #6: 'TOBACCOSMOKINGHISTORYINDICATOR'

Clustering Approach #2: 'METHLYATION CNMF'

Table S6. Get Full Table Description of clustering approach #2: 'METHLYATION CNMF'

Cluster Labels	1	2	3	4	5
Number of samples	6	4	12	8	2

'METHLYATION CNMF' versus 'Time to Death'

P value = 0.0604 (logrank test), Q value = 1

Table S7. Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #1: 'Time to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	29	4	0.1 - 101.8 (6.0)
subtype1	6	0	2.7 - 101.8 (45.9)
subtype2	4	0	1.0 - 95.1 (21.4)
subtype3	11	1	0.6 - 36.8 (2.7)
subtype4	8	3	0.1 - 69.9 (5.2)

Figure S5. Get High-res Image Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #1: 'Time to Death'

'METHLYATION CNMF' versus 'AGE'

P value = 0.817 (ANOVA), Q value = 1

Table S8. Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #2: 'AGE'

	nPatients	Mean (Std.Dev)
ALL	30	50.4 (12.4)
subtype1	6	48.5 (10.0)
subtype2	4	51.5 (8.3)
subtype3	12	48.8 (12.8)
subtype4	8	53.9 (16.1)

Figure S6. Get High-res Image Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #2: 'AGE'

'METHLYATION CNMF' versus 'RADIATIONS.RADIATION.REGIMENINDICATION'

P value = 0.789 (Fisher's exact test), Q value = 1

Table S9. Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #3: 'RADIATIONS.RADIATION.REGIMENINDICATION'

nPatients	NO	YES
ALL	8	22
subtype1	1	5
subtype2	2	2
subtype3	3	9
subtype4	2	6

Figure S7. Get High-res Image Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #3: 'RADIATIONS.RADIATION.REGIMENINDICATION'

'METHLYATION CNMF' versus 'TOBACCOSMOKINGHISTORYINDICATOR'

P value = 0.834 (ANOVA), Q value = 1

Table S10. Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #6: 'TOBACCOSMOKINGHISTORYINDICATOR'

	nPatients	Mean (Std.Dev)
ALL	27	2.0 (1.2)
subtype1	5	2.4 (1.5)
subtype2	4	2.0 (1.4)
subtype3	11	2.0 (1.2)
subtype4	7	1.7 (1.1)

Figure S8. Get High-res Image Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #6: 'TOBACCOSMOKINGHISTORYINDICATOR'

Clustering Approach #3: 'RNAseq CNMF subtypes'

Table S11. Get Full Table Description of clustering approach #3: 'RNAseq CNMF subtypes'

Cluster Labels	1	2	3
Number of samples	16	5	7

'RNAseq CNMF subtypes' versus 'Time to Death'

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

Table S12. Clustering Approach #3: 'RNAseq CNMF subtypes' versus Clinical Feature #1: 'Time to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	27	2	0.1 - 101.8 (2.7)
subtype1	15	1	0.1 - 70.8 (1.7)
subtype2	5	1	12.4 - 95.1 (30.4)
subtype3	7	0	1.0 - 101.8 (2.2)

Figure S9. Get High-res Image Clustering Approach #3: 'RNAseq CNMF subtypes' versus Clinical Feature #1: 'Time to Death'

'RNAseq CNMF subtypes' versus 'AGE'

P value = 0.837 (ANOVA), Q value = 1

Table S13. Clustering Approach #3: 'RNAseq CNMF subtypes' versus Clinical Feature #2: 'AGE'

	nPatients	Mean (Std.Dev)
ALL	28	48.1 (10.6)
subtype1	16	47.4 (11.4)
subtype2	5	47.6 (2.2)
subtype3	7	50.3 (13.0)

Figure S10. Get High-res Image Clustering Approach #3: 'RNAseq CNMF subtypes' versus Clinical Feature #2: 'AGE'

'RNAseq CNMF subtypes' versus 'RADIATIONS.RADIATION.REGIMENINDICATION'

P value = 0.0228 (Fisher's exact test), Q value = 0.55

Table S14. Clustering Approach #3: 'RNAseq CNMF subtypes' versus Clinical Feature #3: 'RADIATIONS.RADIATION.REGIMENINDICATION'

nPatients	NO	YES
ALL	6	22
subtype1	1	15
subtype2	3	2
subtype3	2	5

Figure S11. Get High-res Image Clustering Approach #3: 'RNAseq CNMF subtypes' versus Clinical Feature #3: 'RADIATIONS.RADIATION.REGIMENINDICATION'

'RNAseq CNMF subtypes' versus 'TOBACCOSMOKINGHISTORYINDICATOR'

P value = 0.966 (ANOVA), Q value = 1

Table S15. Clustering Approach #3: 'RNAseq CNMF subtypes' versus Clinical Feature #6: 'TOBACCOSMOKINGHISTORYINDICATOR'

	nPatients	Mean (Std.Dev)
ALL	25	2.0 (1.2)
subtype1	14	2.0 (1.2)
subtype2	4	2.0 (1.4)
subtype3	7	1.9 (1.1)

Figure S12. Get High-res Image Clustering Approach #3: 'RNAseq CNMF subtypes' versus Clinical Feature #6: 'TOBACCOSMOKINGHISTORYINDICATOR'

Clustering Approach #4: 'RNAseq cHierClus subtypes'

Table S16. Get Full Table Description of clustering approach #4: 'RNAseq cHierClus subtypes'

Cluster Labels	1	2	3
Number of samples	4	5	19

'RNAseq cHierClus subtypes' versus 'Time to Death'

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

Table S17. Clustering Approach #4: 'RNAseq cHierClus subtypes' versus Clinical Feature #1: 'Time to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	27	2	0.1 - 101.8 (2.7)
subtype1	4	0	0.1 - 95.1 (35.5)
subtype2	5	1	1.2 - 101.8 (2.2)
subtype3	18	1	0.3 - 70.8 (4.3)

Figure S13. Get High-res Image Clustering Approach #4: 'RNAseq cHierClus subtypes' versus Clinical Feature #1: 'Time to Death'

'RNAseq cHierClus subtypes' versus 'AGE'

P value = 0.617 (ANOVA), Q value = 1

Table S18. Clustering Approach #4: 'RNAseq cHierClus subtypes' versus Clinical Feature #2: 'AGE'

	nPatients	Mean (Std.Dev)
ALL	28	48.1 (10.6)
subtype1	4	43.5 (9.8)
subtype2	5	47.4 (12.3)
subtype3	19	49.3 (10.6)

Figure S14. Get High-res Image Clustering Approach #4: 'RNAseq cHierClus subtypes' versus Clinical Feature #2: 'AGE'

'RNAseq cHierClus subtypes' versus 'RADIATIONS.RADIATION.REGIMENINDICATION'

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

Table S19. Clustering Approach #4: 'RNAseq cHierClus subtypes' versus Clinical Feature #3: 'RADIATIONS.RADIATION.REGIMENINDICATION'

nPatients	NO	YES
ALL	6	22
subtype1	1	3
subtype2	1	4
subtype3	4	15

Figure S15. Get High-res Image Clustering Approach #4: 'RNAseq cHierClus subtypes' versus Clinical Feature #3: 'RADIATIONS.RADIATION.REGIMENINDICATION'

'RNAseq cHierClus subtypes' versus 'TOBACCOSMOKINGHISTORYINDICATOR'

P value = 0.637 (ANOVA), Q value = 1

Table S20. Clustering Approach #4: 'RNAseq cHierClus subtypes' versus Clinical Feature #6: 'TOBACCOSMOKINGHISTORYINDICATOR'

	nPatients	Mean (Std.Dev)
ALL	25	2.0 (1.2)
subtype1	4	2.0 (1.4)
subtype2	5	2.4 (1.5)
subtype3	16	1.8 (1.0)

Figure S16. Get High-res Image Clustering Approach #4: 'RNAseq cHierClus subtypes' versus Clinical Feature #6: 'TOBACCOSMOKINGHISTORYINDICATOR'

Clustering Approach #5: 'MIRseq CNMF subtypes'

Table S21. Get Full Table Description of clustering approach #5: 'MIRseq CNMF subtypes'

Cluster Labels	1	2	3
Number of samples	11	14	7

'MIRseq CNMF subtypes' versus 'Time to Death'

P value = 0.956 (logrank test), Q value = 1

Table S22. Clustering Approach #5: 'MIRseq CNMF subtypes' versus Clinical Feature #1: 'Time to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	31	4	0.1 - 101.8 (5.8)
subtype1	11	2	0.1 - 95.1 (12.4)
subtype2	13	2	0.6 - 101.8 (5.5)
subtype3	7	0	1.0 - 6.0 (1.2)

Figure S17. Get High-res Image Clustering Approach #5: 'MIRseq CNMF subtypes' versus Clinical Feature #1: 'Time to Death'

'MIRseq CNMF subtypes' versus 'AGE'

P value = 0.42 (ANOVA), Q value = 1

Table S23. Clustering Approach #5: 'MIRseq CNMF subtypes' versus Clinical Feature #2: 'AGE'

	nPatients	Mean (Std.Dev)
ALL	32	49.9 (12.7)
subtype1	11	47.5 (12.6)
subtype2	14	53.3 (11.9)
subtype3	7	46.9 (14.6)

Figure S18. Get High-res Image Clustering Approach #5: 'MIRseq CNMF subtypes' versus Clinical Feature #2: 'AGE'

'MIRseq CNMF subtypes' versus 'RADIATIONS.RADIATION.REGIMENINDICATION'

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

Table S24. Clustering Approach #5: 'MIRseq CNMF subtypes' versus Clinical Feature #3: 'RADIATIONS.RADIATION.REGIMENINDICATION'

nPatients	NO	YES
ALL	8	24
subtype1	3	8
subtype2	3	11
subtype3	2	5

Figure S19. Get High-res Image Clustering Approach #5: 'MIRseq CNMF subtypes' versus Clinical Feature #3: 'RADIATIONS.RADIATION.REGIMENINDICATION'

'MIRseq CNMF subtypes' versus 'TOBACCOSMOKINGHISTORYINDICATOR'

P value = 0.862 (ANOVA), Q value = 1

Table S25. Clustering Approach #5: 'MIRseq CNMF subtypes' versus Clinical Feature #6: 'TOBACCOSMOKINGHISTORYINDICATOR'

	nPatients	Mean (Std.Dev)
ALL	29	2.0 (1.2)
subtype1	10	1.8 (1.2)
subtype2	13	2.1 (1.2)
subtype3	6	2.0 (1.3)

Figure S20. Get High-res Image Clustering Approach #5: 'MIRseq CNMF subtypes' versus Clinical Feature #6: 'TOBACCOSMOKINGHISTORYINDICATOR'

Clustering Approach #6: 'MIRseq cHierClus subtypes'

Table S26. Get Full Table Description of clustering approach #6: 'MIRseq cHierClus subtypes'

Cluster Labels	1	2	3
Number of samples	3	18	11

'MIRseq cHierClus subtypes' versus 'Time to Death'

P value = 0.183 (logrank test), Q value = 1

Table S27. Clustering Approach #6: 'MIRseq cHierClus subtypes' versus Clinical Feature #1: 'Time to Death'

	nPatients	nDeath	Duration Range (Median), Month
ALL	31	4	0.1 - 101.8 (5.8)
subtype1	3	0	1.2 - 6.0 (2.2)
subtype2	18	4	0.3 - 101.8 (7.3)
subtype3	10	0	0.1 - 95.1 (4.8)

Figure S21. Get High-res Image Clustering Approach #6: 'MIRseq cHierClus subtypes' versus Clinical Feature #1: 'Time to Death'

'MIRseq cHierClus subtypes' versus 'AGE'

P value = 0.259 (ANOVA), Q value = 1

Table S28. Clustering Approach #6: 'MIRseq cHierClus subtypes' versus Clinical Feature #2: 'AGE'

	nPatients	Mean (Std.Dev)
ALL	32	49.9 (12.7)
subtype1	3	43.0 (13.7)
subtype2	18	53.1 (13.6)
subtype3	11	46.5 (10.2)

Figure S22. Get High-res Image Clustering Approach #6: 'MIRseq cHierClus subtypes' versus Clinical Feature #2: 'AGE'

'MIRseq cHierClus subtypes' versus 'RADIATIONS.RADIATION.REGIMENINDICATION'

P value = 0.856 (Fisher's exact test), Q value = 1

Table S29. Clustering Approach #6: 'MIRseq cHierClus subtypes' versus Clinical Feature #3: 'RADIATIONS.RADIATION.REGIMENINDICATION'

nPatients	NO	YES
ALL	8	24
subtype1	1	2
subtype2	5	13
subtype3	2	9

Figure S23. Get High-res Image Clustering Approach #6: 'MIRseq cHierClus subtypes' versus Clinical Feature #3: 'RADIATIONS.RADIATION.REGIMENINDICATION'

'MIRseq cHierClus subtypes' versus 'TOBACCOSMOKINGHISTORYINDICATOR'

P value = 0.975 (ANOVA), Q value = 1

Table S30. Clustering Approach #6: 'MIRseq cHierClus subtypes' versus Clinical Feature #6: 'TOBACCOSMOKINGHISTORYINDICATOR'

	nPatients	Mean (Std.Dev)
ALL	29	2.0 (1.2)
subtype1	3	2.0 (1.0)
subtype2	17	2.0 (1.2)
subtype3	9	1.9 (1.3)

Figure S24. Get High-res Image Clustering Approach #6: 'MIRseq cHierClus subtypes' versus Clinical Feature #6: 'TOBACCOSMOKINGHISTORYINDICATOR'

Methods & Data

Input

Cluster data file = CESC-TP.mergedcluster.txt
Clinical data file = CESC-TP.clin.merged.picked.txt
Number of patients = 32
Number of clustering approaches = 6
Number of selected clinical features = 6
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

ANOVA analysis

For continuous numerical clinical features, one-way analysis of variance (Howell 2002) was applied to compare the clinical values between tumor subtypes using 'anova' 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.

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] Howell, D, Statistical Methods for Psychology. (5th ed.), Duxbury Press:324-5 (2002)

[5] 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)

[6] 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