This pipeline computes the correlation between cancer subtypes identified by different molecular patterns and selected clinical features.
Testing the association between subtypes identified by 6 different clustering approaches and 5 clinical features across 155 patients, 3 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 'AGE' and 'NUMBER.OF.LYMPH.NODES'.
-
3 subtypes identified in current cancer cohort by 'METHLYATION CNMF'. These subtypes correlate to 'AGE'.
-
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.
-
4 subtypes identified in current cancer cohort by 'MIRSEQ CNMF'. These subtypes do not correlate to any clinical features.
-
3 subtypes identified in current cancer cohort by 'MIRSEQ CHIERARCHICAL'. These subtypes do not correlate to any clinical features.
Table 1. Get Full Table Overview of the association between subtypes identified by 6 different clustering approaches and 5 clinical features. Shown in the table are P values (Q values). Thresholded by P value < 0.05 and Q value < 0.25, 3 significant findings detected.
Clinical Features |
Time to Death |
AGE |
RADIATIONS RADIATION REGIMENINDICATION |
COMPLETENESS OF RESECTION |
NUMBER OF LYMPH NODES |
Statistical Tests | logrank test | ANOVA | Fisher's exact test | Chi-square test | ANOVA |
Copy Number Ratio CNMF subtypes |
100 (1.00) |
0.00523 (0.146) |
0.85 (1.00) |
0.647 (1.00) |
0.00288 (0.0837) |
METHLYATION CNMF |
100 (1.00) |
0.00237 (0.0711) |
0.862 (1.00) |
0.211 (1.00) |
0.123 (1.00) |
RNAseq CNMF subtypes |
100 (1.00) |
0.191 (1.00) |
0.868 (1.00) |
0.258 (1.00) |
0.164 (1.00) |
RNAseq cHierClus subtypes |
100 (1.00) |
0.86 (1.00) |
0.735 (1.00) |
0.166 (1.00) |
0.152 (1.00) |
MIRSEQ CNMF |
100 (1.00) |
0.0265 (0.715) |
0.466 (1.00) |
0.454 (1.00) |
0.0459 (1.00) |
MIRSEQ CHIERARCHICAL |
100 (1.00) |
0.399 (1.00) |
0.527 (1.00) |
0.697 (1.00) |
0.387 (1.00) |
Table S1. Get Full Table Description of clustering approach #1: 'Copy Number Ratio CNMF subtypes'
Cluster Labels | 1 | 2 | 3 | 4 |
---|---|---|---|---|
Number of samples | 28 | 72 | 52 | 2 |
P value = 100 (logrank test), Q value = 1
Table S2. Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #1: 'Time to Death'
nPatients | nDeath | Duration Range (Median), Month | |
---|---|---|---|
ALL | 152 | 1 | 0.3 - 66.0 (15.0) |
subtype1 | 28 | 0 | 0.3 - 63.3 (11.2) |
subtype2 | 72 | 0 | 1.0 - 65.9 (16.9) |
subtype3 | 52 | 1 | 0.9 - 66.0 (16.3) |
Figure S1. Get High-res Image Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #1: 'Time to Death'
![](D1V1.png)
P value = 0.00523 (ANOVA), Q value = 0.15
Table S3. Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #2: 'AGE'
nPatients | Mean (Std.Dev) | |
---|---|---|
ALL | 151 | 60.6 (6.7) |
subtype1 | 28 | 62.1 (5.7) |
subtype2 | 71 | 58.8 (7.1) |
subtype3 | 52 | 62.3 (6.1) |
Figure S2. Get High-res Image Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #2: 'AGE'
![](D1V2.png)
P value = 0.85 (Fisher's exact test), Q value = 1
Table S4. Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #3: 'RADIATIONS.RADIATION.REGIMENINDICATION'
nPatients | NO | YES |
---|---|---|
ALL | 5 | 147 |
subtype1 | 1 | 27 |
subtype2 | 3 | 69 |
subtype3 | 1 | 51 |
Figure S3. Get High-res Image Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #3: 'RADIATIONS.RADIATION.REGIMENINDICATION'
![](D1V3.png)
P value = 0.647 (Chi-square test), Q value = 1
Table S5. Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #4: 'COMPLETENESS.OF.RESECTION'
nPatients | R0 | R1 | RX |
---|---|---|---|
ALL | 116 | 29 | 2 |
subtype1 | 19 | 6 | 0 |
subtype2 | 56 | 13 | 2 |
subtype3 | 41 | 10 | 0 |
Figure S4. Get High-res Image Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #4: 'COMPLETENESS.OF.RESECTION'
![](D1V4.png)
P value = 0.00288 (ANOVA), Q value = 0.084
Table S6. Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #5: 'NUMBER.OF.LYMPH.NODES'
nPatients | Mean (Std.Dev) | |
---|---|---|
ALL | 133 | 0.2 (0.7) |
subtype1 | 26 | 0.0 (0.0) |
subtype2 | 59 | 0.1 (0.3) |
subtype3 | 48 | 0.5 (1.2) |
Figure S5. Get High-res Image Clustering Approach #1: 'Copy Number Ratio CNMF subtypes' versus Clinical Feature #5: 'NUMBER.OF.LYMPH.NODES'
![](D1V5.png)
Table S7. Get Full Table Description of clustering approach #2: 'METHLYATION CNMF'
Cluster Labels | 1 | 2 | 3 |
---|---|---|---|
Number of samples | 47 | 65 | 43 |
P value = 100 (logrank test), Q value = 1
Table S8. Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #1: 'Time to Death'
nPatients | nDeath | Duration Range (Median), Month | |
---|---|---|---|
ALL | 155 | 1 | 0.3 - 66.0 (15.0) |
subtype1 | 47 | 0 | 0.3 - 65.9 (15.1) |
subtype2 | 65 | 1 | 1.0 - 66.0 (16.6) |
subtype3 | 43 | 0 | 1.1 - 62.4 (13.9) |
Figure S6. Get High-res Image Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #1: 'Time to Death'
![](D2V1.png)
P value = 0.00237 (ANOVA), Q value = 0.071
Table S9. Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #2: 'AGE'
nPatients | Mean (Std.Dev) | |
---|---|---|
ALL | 154 | 60.4 (6.9) |
subtype1 | 46 | 61.9 (6.3) |
subtype2 | 65 | 61.4 (6.8) |
subtype3 | 43 | 57.4 (6.7) |
Figure S7. Get High-res Image Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #2: 'AGE'
![](D2V2.png)
P value = 0.862 (Fisher's exact test), Q value = 1
Table S10. Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #3: 'RADIATIONS.RADIATION.REGIMENINDICATION'
nPatients | NO | YES |
---|---|---|
ALL | 5 | 150 |
subtype1 | 1 | 46 |
subtype2 | 2 | 63 |
subtype3 | 2 | 41 |
Figure S8. Get High-res Image Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #3: 'RADIATIONS.RADIATION.REGIMENINDICATION'
![](D2V3.png)
P value = 0.211 (Chi-square test), Q value = 1
Table S11. Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #4: 'COMPLETENESS.OF.RESECTION'
nPatients | R0 | R1 | RX |
---|---|---|---|
ALL | 118 | 30 | 2 |
subtype1 | 35 | 11 | 0 |
subtype2 | 51 | 11 | 0 |
subtype3 | 32 | 8 | 2 |
Figure S9. Get High-res Image Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #4: 'COMPLETENESS.OF.RESECTION'
![](D2V4.png)
P value = 0.123 (ANOVA), Q value = 1
Table S12. Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #5: 'NUMBER.OF.LYMPH.NODES'
nPatients | Mean (Std.Dev) | |
---|---|---|
ALL | 136 | 0.2 (0.7) |
subtype1 | 43 | 0.2 (0.5) |
subtype2 | 58 | 0.3 (1.0) |
subtype3 | 35 | 0.0 (0.2) |
Figure S10. Get High-res Image Clustering Approach #2: 'METHLYATION CNMF' versus Clinical Feature #5: 'NUMBER.OF.LYMPH.NODES'
![](D2V5.png)
Table S13. Get Full Table Description of clustering approach #3: 'RNAseq CNMF subtypes'
Cluster Labels | 1 | 2 | 3 |
---|---|---|---|
Number of samples | 51 | 49 | 53 |
P value = 100 (logrank test), Q value = 1
Table S14. Clustering Approach #3: 'RNAseq CNMF subtypes' versus Clinical Feature #1: 'Time to Death'
nPatients | nDeath | Duration Range (Median), Month | |
---|---|---|---|
ALL | 153 | 1 | 0.3 - 66.0 (14.8) |
subtype1 | 51 | 0 | 0.3 - 65.9 (16.0) |
subtype2 | 49 | 0 | 1.0 - 62.4 (12.8) |
subtype3 | 53 | 1 | 1.0 - 66.0 (16.6) |
Figure S11. Get High-res Image Clustering Approach #3: 'RNAseq CNMF subtypes' versus Clinical Feature #1: 'Time to Death'
![](D3V1.png)
P value = 0.191 (ANOVA), Q value = 1
Table S15. Clustering Approach #3: 'RNAseq CNMF subtypes' versus Clinical Feature #2: 'AGE'
nPatients | Mean (Std.Dev) | |
---|---|---|
ALL | 152 | 60.5 (6.9) |
subtype1 | 50 | 60.8 (6.6) |
subtype2 | 49 | 59.1 (7.1) |
subtype3 | 53 | 61.5 (6.8) |
Figure S12. Get High-res Image Clustering Approach #3: 'RNAseq CNMF subtypes' versus Clinical Feature #2: 'AGE'
![](D3V2.png)
P value = 0.868 (Fisher's exact test), Q value = 1
Table S16. Clustering Approach #3: 'RNAseq CNMF subtypes' versus Clinical Feature #3: 'RADIATIONS.RADIATION.REGIMENINDICATION'
nPatients | NO | YES |
---|---|---|
ALL | 5 | 148 |
subtype1 | 1 | 50 |
subtype2 | 2 | 47 |
subtype3 | 2 | 51 |
Figure S13. Get High-res Image Clustering Approach #3: 'RNAseq CNMF subtypes' versus Clinical Feature #3: 'RADIATIONS.RADIATION.REGIMENINDICATION'
![](D3V3.png)
P value = 0.258 (Chi-square test), Q value = 1
Table S17. Clustering Approach #3: 'RNAseq CNMF subtypes' versus Clinical Feature #4: 'COMPLETENESS.OF.RESECTION'
nPatients | R0 | R1 | RX |
---|---|---|---|
ALL | 117 | 29 | 2 |
subtype1 | 39 | 12 | 0 |
subtype2 | 38 | 7 | 2 |
subtype3 | 40 | 10 | 0 |
Figure S14. Get High-res Image Clustering Approach #3: 'RNAseq CNMF subtypes' versus Clinical Feature #4: 'COMPLETENESS.OF.RESECTION'
![](D3V4.png)
P value = 0.164 (ANOVA), Q value = 1
Table S18. Clustering Approach #3: 'RNAseq CNMF subtypes' versus Clinical Feature #5: 'NUMBER.OF.LYMPH.NODES'
nPatients | Mean (Std.Dev) | |
---|---|---|
ALL | 135 | 0.2 (0.7) |
subtype1 | 44 | 0.2 (0.9) |
subtype2 | 44 | 0.0 (0.2) |
subtype3 | 47 | 0.3 (0.8) |
Figure S15. Get High-res Image Clustering Approach #3: 'RNAseq CNMF subtypes' versus Clinical Feature #5: 'NUMBER.OF.LYMPH.NODES'
![](D3V5.png)
Table S19. Get Full Table Description of clustering approach #4: 'RNAseq cHierClus subtypes'
Cluster Labels | 1 | 2 | 3 |
---|---|---|---|
Number of samples | 43 | 64 | 46 |
P value = 100 (logrank test), Q value = 1
Table S20. Clustering Approach #4: 'RNAseq cHierClus subtypes' versus Clinical Feature #1: 'Time to Death'
nPatients | nDeath | Duration Range (Median), Month | |
---|---|---|---|
ALL | 153 | 1 | 0.3 - 66.0 (14.8) |
subtype1 | 43 | 0 | 1.0 - 65.9 (16.0) |
subtype2 | 64 | 1 | 0.9 - 66.0 (14.8) |
subtype3 | 46 | 0 | 0.3 - 62.4 (12.4) |
Figure S16. Get High-res Image Clustering Approach #4: 'RNAseq cHierClus subtypes' versus Clinical Feature #1: 'Time to Death'
![](D4V1.png)
P value = 0.86 (ANOVA), Q value = 1
Table S21. Clustering Approach #4: 'RNAseq cHierClus subtypes' versus Clinical Feature #2: 'AGE'
nPatients | Mean (Std.Dev) | |
---|---|---|
ALL | 152 | 60.5 (6.9) |
subtype1 | 42 | 60.7 (6.9) |
subtype2 | 64 | 60.7 (6.9) |
subtype3 | 46 | 60.0 (6.9) |
Figure S17. Get High-res Image Clustering Approach #4: 'RNAseq cHierClus subtypes' versus Clinical Feature #2: 'AGE'
![](D4V2.png)
P value = 0.735 (Fisher's exact test), Q value = 1
Table S22. Clustering Approach #4: 'RNAseq cHierClus subtypes' versus Clinical Feature #3: 'RADIATIONS.RADIATION.REGIMENINDICATION'
nPatients | NO | YES |
---|---|---|
ALL | 5 | 148 |
subtype1 | 1 | 42 |
subtype2 | 3 | 61 |
subtype3 | 1 | 45 |
Figure S18. Get High-res Image Clustering Approach #4: 'RNAseq cHierClus subtypes' versus Clinical Feature #3: 'RADIATIONS.RADIATION.REGIMENINDICATION'
![](D4V3.png)
P value = 0.166 (Chi-square test), Q value = 1
Table S23. Clustering Approach #4: 'RNAseq cHierClus subtypes' versus Clinical Feature #4: 'COMPLETENESS.OF.RESECTION'
nPatients | R0 | R1 | RX |
---|---|---|---|
ALL | 117 | 29 | 2 |
subtype1 | 32 | 11 | 0 |
subtype2 | 49 | 12 | 0 |
subtype3 | 36 | 6 | 2 |
Figure S19. Get High-res Image Clustering Approach #4: 'RNAseq cHierClus subtypes' versus Clinical Feature #4: 'COMPLETENESS.OF.RESECTION'
![](D4V4.png)
P value = 0.152 (ANOVA), Q value = 1
Table S24. Clustering Approach #4: 'RNAseq cHierClus subtypes' versus Clinical Feature #5: 'NUMBER.OF.LYMPH.NODES'
nPatients | Mean (Std.Dev) | |
---|---|---|
ALL | 135 | 0.2 (0.7) |
subtype1 | 37 | 0.2 (1.0) |
subtype2 | 57 | 0.3 (0.8) |
subtype3 | 41 | 0.0 (0.2) |
Figure S20. Get High-res Image Clustering Approach #4: 'RNAseq cHierClus subtypes' versus Clinical Feature #5: 'NUMBER.OF.LYMPH.NODES'
![](D4V5.png)
Table S25. Get Full Table Description of clustering approach #5: 'MIRSEQ CNMF'
Cluster Labels | 1 | 2 | 3 | 4 |
---|---|---|---|---|
Number of samples | 43 | 40 | 26 | 45 |
P value = 100 (logrank test), Q value = 1
Table S26. Clustering Approach #5: 'MIRSEQ CNMF' versus Clinical Feature #1: 'Time to Death'
nPatients | nDeath | Duration Range (Median), Month | |
---|---|---|---|
ALL | 154 | 1 | 0.3 - 66.0 (14.9) |
subtype1 | 43 | 0 | 0.9 - 66.0 (16.6) |
subtype2 | 40 | 0 | 1.0 - 54.9 (19.1) |
subtype3 | 26 | 0 | 0.3 - 64.1 (16.0) |
subtype4 | 45 | 1 | 1.0 - 66.0 (5.6) |
Figure S21. Get High-res Image Clustering Approach #5: 'MIRSEQ CNMF' versus Clinical Feature #1: 'Time to Death'
![](D5V1.png)
P value = 0.0265 (ANOVA), Q value = 0.71
Table S27. Clustering Approach #5: 'MIRSEQ CNMF' versus Clinical Feature #2: 'AGE'
nPatients | Mean (Std.Dev) | |
---|---|---|
ALL | 153 | 60.4 (6.9) |
subtype1 | 43 | 63.0 (6.2) |
subtype2 | 40 | 58.7 (6.7) |
subtype3 | 26 | 59.4 (6.8) |
subtype4 | 44 | 60.0 (7.1) |
Figure S22. Get High-res Image Clustering Approach #5: 'MIRSEQ CNMF' versus Clinical Feature #2: 'AGE'
![](D5V2.png)
P value = 0.466 (Fisher's exact test), Q value = 1
Table S28. Clustering Approach #5: 'MIRSEQ CNMF' versus Clinical Feature #3: 'RADIATIONS.RADIATION.REGIMENINDICATION'
nPatients | NO | YES |
---|---|---|
ALL | 5 | 149 |
subtype1 | 2 | 41 |
subtype2 | 2 | 38 |
subtype3 | 1 | 25 |
subtype4 | 0 | 45 |
Figure S23. Get High-res Image Clustering Approach #5: 'MIRSEQ CNMF' versus Clinical Feature #3: 'RADIATIONS.RADIATION.REGIMENINDICATION'
![](D5V3.png)
P value = 0.454 (Chi-square test), Q value = 1
Table S29. Clustering Approach #5: 'MIRSEQ CNMF' versus Clinical Feature #4: 'COMPLETENESS.OF.RESECTION'
nPatients | R0 | R1 | RX |
---|---|---|---|
ALL | 118 | 29 | 2 |
subtype1 | 33 | 10 | 0 |
subtype2 | 30 | 8 | 1 |
subtype3 | 18 | 7 | 0 |
subtype4 | 37 | 4 | 1 |
Figure S24. Get High-res Image Clustering Approach #5: 'MIRSEQ CNMF' versus Clinical Feature #4: 'COMPLETENESS.OF.RESECTION'
![](D5V4.png)
P value = 0.0459 (ANOVA), Q value = 1
Table S30. Clustering Approach #5: 'MIRSEQ CNMF' versus Clinical Feature #5: 'NUMBER.OF.LYMPH.NODES'
nPatients | Mean (Std.Dev) | |
---|---|---|
ALL | 135 | 0.2 (0.7) |
subtype1 | 40 | 0.5 (1.2) |
subtype2 | 37 | 0.1 (0.3) |
subtype3 | 21 | 0.0 (0.2) |
subtype4 | 37 | 0.1 (0.4) |
Figure S25. Get High-res Image Clustering Approach #5: 'MIRSEQ CNMF' versus Clinical Feature #5: 'NUMBER.OF.LYMPH.NODES'
![](D5V5.png)
Table S31. Get Full Table Description of clustering approach #6: 'MIRSEQ CHIERARCHICAL'
Cluster Labels | 1 | 2 | 3 |
---|---|---|---|
Number of samples | 31 | 59 | 64 |
P value = 100 (logrank test), Q value = 1
Table S32. Clustering Approach #6: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #1: 'Time to Death'
nPatients | nDeath | Duration Range (Median), Month | |
---|---|---|---|
ALL | 154 | 1 | 0.3 - 66.0 (14.9) |
subtype1 | 31 | 0 | 0.3 - 64.1 (18.2) |
subtype2 | 59 | 0 | 1.0 - 54.9 (19.5) |
subtype3 | 64 | 1 | 0.9 - 66.0 (5.6) |
Figure S26. Get High-res Image Clustering Approach #6: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #1: 'Time to Death'
![](D6V1.png)
P value = 0.399 (ANOVA), Q value = 1
Table S33. Clustering Approach #6: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #2: 'AGE'
nPatients | Mean (Std.Dev) | |
---|---|---|
ALL | 153 | 60.4 (6.9) |
subtype1 | 30 | 59.1 (7.1) |
subtype2 | 59 | 60.3 (6.8) |
subtype3 | 64 | 61.1 (6.8) |
Figure S27. Get High-res Image Clustering Approach #6: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #2: 'AGE'
![](D6V2.png)
P value = 0.527 (Fisher's exact test), Q value = 1
Table S34. Clustering Approach #6: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #3: 'RADIATIONS.RADIATION.REGIMENINDICATION'
nPatients | NO | YES |
---|---|---|
ALL | 5 | 149 |
subtype1 | 1 | 30 |
subtype2 | 3 | 56 |
subtype3 | 1 | 63 |
Figure S28. Get High-res Image Clustering Approach #6: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #3: 'RADIATIONS.RADIATION.REGIMENINDICATION'
![](D6V3.png)
P value = 0.697 (Chi-square test), Q value = 1
Table S35. Clustering Approach #6: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #4: 'COMPLETENESS.OF.RESECTION'
nPatients | R0 | R1 | RX |
---|---|---|---|
ALL | 118 | 29 | 2 |
subtype1 | 22 | 8 | 0 |
subtype2 | 46 | 12 | 1 |
subtype3 | 50 | 9 | 1 |
Figure S29. Get High-res Image Clustering Approach #6: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #4: 'COMPLETENESS.OF.RESECTION'
![](D6V4.png)
P value = 0.387 (ANOVA), Q value = 1
Table S36. Clustering Approach #6: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #5: 'NUMBER.OF.LYMPH.NODES'
nPatients | Mean (Std.Dev) | |
---|---|---|
ALL | 135 | 0.2 (0.7) |
subtype1 | 23 | 0.0 (0.2) |
subtype2 | 55 | 0.3 (0.8) |
subtype3 | 57 | 0.2 (0.8) |
Figure S30. Get High-res Image Clustering Approach #6: 'MIRSEQ CHIERARCHICAL' versus Clinical Feature #5: 'NUMBER.OF.LYMPH.NODES'
![](D6V5.png)
-
Cluster data file = PRAD-TP.mergedcluster.txt
-
Clinical data file = PRAD-TP.clin.merged.picked.txt
-
Number of patients = 155
-
Number of clustering approaches = 6
-
Number of selected clinical features = 5
-
Exclude small clusters that include fewer than K patients, K = 3
consensus non-negative matrix factorization clustering approach (Brunet et al. 2004)
Resampling-based clustering method (Monti et al. 2003)
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
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
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
For multi-class clinical features (nominal or ordinal), Chi-square tests (Greenwood and Nikulin 1996) were used to estimate the P values using the 'chisq.test' function in R
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.
This is an experimental feature. The full results of the analysis summarized in this report can be downloaded from the TCGA Data Coordination Center.