PANCANCER subset with 8 initial disease types: Correlation between mRNA expression and clinical features
Maintained by TCGA GDAC Team (Broad Institute/Dana-Farber Cancer Institute/Harvard Medical School)
Overview
Introduction

This pipeline uses various statistical tests to identify mRNAs whose expression levels correlated to selected clinical features.

Summary

Testing the association between 17814 genes and 12 clinical features across 1569 samples, statistically thresholded by Q value < 0.05, 12 clinical features related to at least one genes.

  • 3027 genes correlated to 'Time to Death'.

    • KLHL14 ,  ASS1 ,  IGSF11 ,  UPK1B ,  C4ORF32 ,  ...

  • 3544 genes correlated to 'AGE'.

    • GREB1 ,  DUOX2 ,  COL18A1 ,  GALNTL1 ,  LDOC1 ,  ...

  • 17276 genes correlated to 'PRIMARY.SITE.OF.DISEASE'.

    • KLHL14 ,  ZG16 ,  SLC22A2 ,  FOXE1 ,  PDZD3 ,  ...

  • 9002 genes correlated to 'GENDER'.

    • ESR1 ,  DDX3Y ,  RSPO1 ,  RPS4Y1 ,  EIF1AY ,  ...

  • 274 genes correlated to 'KARNOFSKY.PERFORMANCE.SCORE'.

    • PPP1R9A ,  VASN ,  INSIG2 ,  PCYT1A ,  TLE1 ,  ...

  • 16527 genes correlated to 'HISTOLOGICAL.TYPE'.

    • KLHL14 ,  EMX2 ,  NPR1 ,  FUT4 ,  FXYD2 ,  ...

  • 5428 genes correlated to 'PATHOLOGY.T'.

    • NOXO1 ,  CDCA7 ,  C20ORF151 ,  B3GNT3 ,  C13ORF3 ,  ...

  • 12 genes correlated to 'PATHOLOGY.N'.

    • FAM73A ,  TNFRSF10C ,  MGC50559 ,  GABARAP ,  MMP11 ,  ...

  • 33 genes correlated to 'PATHOLOGICSPREAD(M)'.

    • KCNC2 ,  TCF7 ,  HCN1 ,  C8ORF30A ,  NPAS2 ,  ...

  • 7068 genes correlated to 'TUMOR.STAGE'.

    • FGF18 ,  WT1 ,  UGT1A6 ,  ZNF503 ,  KITLG ,  ...

  • 5808 genes correlated to 'RADIATIONS.RADIATION.REGIMENINDICATION'.

    • CYP3A5 ,  TCF21 ,  CYP3A4 ,  CYP3A7 ,  AADAC ,  ...

  • 7886 genes correlated to 'NEOADJUVANT.THERAPY'.

    • ATP6V1B1 ,  NPAS3 ,  UGT1A6 ,  SCARA3 ,  GPX2 ,  ...

Results
Overview of the results

Complete statistical result table is provided in Supplement Table 1

Table 1.  Get Full Table This table shows the clinical features, statistical methods used, and the number of genes that are significantly associated with each clinical feature at Q value < 0.05.

Clinical feature Statistical test Significant genes Associated with                 Associated with
Time to Death Cox regression test N=3027 shorter survival N=1360 longer survival N=1667
AGE Spearman correlation test N=3544 older N=1692 younger N=1852
PRIMARY SITE OF DISEASE ANOVA test N=17276        
GENDER t test N=9002 male N=4105 female N=4897
KARNOFSKY PERFORMANCE SCORE Spearman correlation test N=274 higher score N=164 lower score N=110
HISTOLOGICAL TYPE ANOVA test N=16527        
PATHOLOGY T Spearman correlation test N=5428 higher pT N=2925 lower pT N=2503
PATHOLOGY N Spearman correlation test N=12 higher pN N=4 lower pN N=8
PATHOLOGICSPREAD(M) ANOVA test N=33        
TUMOR STAGE Spearman correlation test N=7068 higher stage N=3772 lower stage N=3296
RADIATIONS RADIATION REGIMENINDICATION t test N=5808 yes N=2873 no N=2935
NEOADJUVANT THERAPY t test N=7886 yes N=4017 no N=3869
Clinical variable #1: 'Time to Death'

3027 genes related to 'Time to Death'.

Table S1.  Basic characteristics of clinical feature: 'Time to Death'

Time to Death Duration (Months) 0.1-223.4 (median=23.1)
  censored N = 978
  death N = 435
     
  Significant markers N = 3027
  associated with shorter survival 1360
  associated with longer survival 1667
List of top 10 genes significantly associated with 'Time to Death' by Cox regression test

Table S2.  Get Full Table List of top 10 genes significantly associated with 'Time to Death' by Cox regression test

HazardRatio Wald_P Q C_index
KLHL14 1.27 0 0 0.602
ASS1 1.32 0 0 0.622
IGSF11 1.36 0 0 0.641
UPK1B 1.22 0 0 0.637
C4ORF32 0.77 0 0 0.38
ZRSR1 0.54 0 0 0.348
MDS1 1.37 0 0 0.619
GALNT12 1.32 0 0 0.617
MUC4 1.22 0 0 0.637
NUBP1 0.47 0 0 0.353

Figure S1.  Get High-res Image As an example, this figure shows the association of KLHL14 to 'Time to Death'. four curves present the cumulative survival rates of 4 quartile subsets of patients. P value = 0 with univariate Cox regression analysis using continuous log-2 expression values.

Clinical variable #2: 'AGE'

3544 genes related to 'AGE'.

Table S3.  Basic characteristics of clinical feature: 'AGE'

AGE Mean (SD) 61.27 (13)
  Significant markers N = 3544
  pos. correlated 1692
  neg. correlated 1852
List of top 10 genes significantly correlated to 'AGE' by Spearman correlation test

Table S4.  Get Full Table List of top 10 genes significantly correlated to 'AGE' by Spearman correlation test

SpearmanCorr corrP Q
GREB1 -0.2885 3.417e-31 6.09e-27
DUOX2 0.2737 4.023e-28 7.17e-24
COL18A1 -0.2709 1.487e-27 2.65e-23
GALNTL1 -0.2655 1.69e-26 3.01e-22
LDOC1 -0.2641 3.213e-26 5.72e-22
CYP2C18 0.261 1.219e-25 2.17e-21
SYT12 -0.2608 1.374e-25 2.45e-21
OBSL1 -0.2581 4.382e-25 7.8e-21
SH3BP5 -0.258 4.545e-25 8.09e-21
BBOX1 -0.2564 9.031e-25 1.61e-20

Figure S2.  Get High-res Image As an example, this figure shows the association of GREB1 to 'AGE'. P value = 3.42e-31 with Spearman correlation analysis. The straight line presents the best linear regression.

Clinical variable #3: 'PRIMARY.SITE.OF.DISEASE'

17276 genes related to 'PRIMARY.SITE.OF.DISEASE'.

Table S5.  Basic characteristics of clinical feature: 'PRIMARY.SITE.OF.DISEASE'

PRIMARY.SITE.OF.DISEASE Labels N
  BREAST 529
  COLON 154
  ENDOMETRIAL 54
  KIDNEY 72
  LUNG 154
  OMENTUM 2
  OVARY 532
  PERITONEUM (OVARY) 2
  RECTUM 68
     
  Significant markers N = 17276
List of top 10 genes differentially expressed by 'PRIMARY.SITE.OF.DISEASE'

Table S6.  Get Full Table List of top 10 genes differentially expressed by 'PRIMARY.SITE.OF.DISEASE'

ANOVA_P Q
KLHL14 0 0
ZG16 0 0
SLC22A2 0 0
FOXE1 0 0
PDZD3 0 0
EMX2 0 0
NPR1 0 0
FUT4 0 0
ACF 0 0
FXYD2 0 0

Figure S3.  Get High-res Image As an example, this figure shows the association of KLHL14 to 'PRIMARY.SITE.OF.DISEASE'. P value = 0 with ANOVA analysis.

Clinical variable #4: 'GENDER'

9002 genes related to 'GENDER'.

Table S7.  Basic characteristics of clinical feature: 'GENDER'

GENDER Labels N
  FEMALE 1293
  MALE 276
     
  Significant markers N = 9002
  Higher in MALE 4105
  Higher in FEMALE 4897
List of top 10 genes differentially expressed by 'GENDER'

Table S8.  Get Full Table List of top 10 genes differentially expressed by 'GENDER'

T(pos if higher in 'MALE') ttestP Q AUC
ESR1 -37.57 4.305e-174 7.67e-170 0.8825
DDX3Y 54.97 1.99e-164 3.55e-160 0.988
RSPO1 -31.37 6.859e-164 1.22e-159 0.8489
RPS4Y1 53.88 2.037e-156 3.63e-152 0.9794
EIF1AY 50.16 1.226e-150 2.18e-146 0.9879
RPS4Y2 49.57 1.819e-146 3.24e-142 0.9867
ERBB4 -32.64 7.076e-145 1.26e-140 0.8607
JARID1D 44.65 2.215e-144 3.94e-140 0.9772
SCGB2A2 -29.3 2.86e-142 5.09e-138 0.8615
CYORF15A 43.63 1.659e-135 2.95e-131 0.9824

Figure S4.  Get High-res Image As an example, this figure shows the association of ESR1 to 'GENDER'. P value = 4.3e-174 with T-test analysis.

Clinical variable #5: 'KARNOFSKY.PERFORMANCE.SCORE'

274 genes related to 'KARNOFSKY.PERFORMANCE.SCORE'.

Table S9.  Basic characteristics of clinical feature: 'KARNOFSKY.PERFORMANCE.SCORE'

KARNOFSKY.PERFORMANCE.SCORE Mean (SD) 60.55 (34)
  Significant markers N = 274
  pos. correlated 164
  neg. correlated 110
List of top 10 genes significantly correlated to 'KARNOFSKY.PERFORMANCE.SCORE' by Spearman correlation test

Table S10.  Get Full Table List of top 10 genes significantly correlated to 'KARNOFSKY.PERFORMANCE.SCORE' by Spearman correlation test

SpearmanCorr corrP Q
PPP1R9A 0.6265 3.091e-11 5.51e-07
VASN 0.6117 1.189e-10 2.12e-06
INSIG2 -0.5926 6.1e-10 1.09e-05
PCYT1A -0.5809 1.57e-09 2.8e-05
TLE1 0.5742 2.663e-09 4.74e-05
MARCH7 -0.5725 3.049e-09 5.43e-05
CCDC135 0.5661 4.96e-09 8.83e-05
CXXC5 0.5647 5.543e-09 9.87e-05
ZNF503 0.5639 5.897e-09 0.000105
SOX5 0.5596 8.108e-09 0.000144

Figure S5.  Get High-res Image As an example, this figure shows the association of PPP1R9A to 'KARNOFSKY.PERFORMANCE.SCORE'. P value = 3.09e-11 with Spearman correlation analysis. The straight line presents the best linear regression.

Clinical variable #6: 'HISTOLOGICAL.TYPE'

16527 genes related to 'HISTOLOGICAL.TYPE'.

Table S11.  Basic characteristics of clinical feature: 'HISTOLOGICAL.TYPE'

HISTOLOGICAL.TYPE Labels N
  COLON ADENOCARCINOMA 128
  COLON MUCINOUS ADENOCARCINOMA 24
  ENDOMETRIOID ENDOMETRIAL ADENOCARCINOMA 41
  KIDNEY CLEAR CELL RENAL CARCINOMA 72
  LUNG BASALOID SQUAMOUS CELL CARCINOMA 5
  LUNG PAPILLARY SQUAMOUS CELL CARICNOMA 1
  LUNG SQUAMOUS CELL CARCINOMA- NOT OTHERWISE SPECIFIED (NOS) 148
  MIXED SEROUS AND ENDOMETRIOID 1
  RECTAL ADENOCARCINOMA 58
  RECTAL MUCINOUS ADENOCARCINOMA 7
  SEROUS CYSTADENOCARCINOMA 536
  SEROUS ENDOMETRIAL ADENOCARCINOMA 12
     
  Significant markers N = 16527
List of top 10 genes differentially expressed by 'HISTOLOGICAL.TYPE'

Table S12.  Get Full Table List of top 10 genes differentially expressed by 'HISTOLOGICAL.TYPE'

ANOVA_P Q
KLHL14 0 0
EMX2 0 0
NPR1 0 0
FUT4 0 0
FXYD2 0 0
SPINK4 0 0
SLC12A2 0 0
CDH17 0 0
SATB2 0 0
OVOL2 0 0

Figure S6.  Get High-res Image As an example, this figure shows the association of KLHL14 to 'HISTOLOGICAL.TYPE'. P value = 0 with ANOVA analysis.

Clinical variable #7: 'PATHOLOGY.T'

5428 genes related to 'PATHOLOGY.T'.

Table S13.  Basic characteristics of clinical feature: 'PATHOLOGY.T'

PATHOLOGY.T Mean (SD) 2.35 (0.85)
  N
  T1 80
  T2 160
  T3 179
  T4 29
     
  Significant markers N = 5428
  pos. correlated 2925
  neg. correlated 2503
List of top 10 genes significantly correlated to 'PATHOLOGY.T' by Spearman correlation test

Table S14.  Get Full Table List of top 10 genes significantly correlated to 'PATHOLOGY.T' by Spearman correlation test

SpearmanCorr corrP Q
NOXO1 0.5004 9.064e-30 1.61e-25
CDCA7 0.4999 1.052e-29 1.87e-25
C20ORF151 0.4993 1.281e-29 2.28e-25
B3GNT3 0.4967 2.749e-29 4.9e-25
C13ORF3 0.4934 7.329e-29 1.31e-24
C19ORF21 0.4912 1.388e-28 2.47e-24
LIFR -0.491 1.488e-28 2.65e-24
NR3C1 -0.4893 2.372e-28 4.22e-24
MUC13 0.4884 3.138e-28 5.59e-24
TACSTD1 0.4866 5.24e-28 9.33e-24

Figure S7.  Get High-res Image As an example, this figure shows the association of NOXO1 to 'PATHOLOGY.T'. P value = 9.06e-30 with Spearman correlation analysis.

Clinical variable #8: 'PATHOLOGY.N'

12 genes related to 'PATHOLOGY.N'.

Table S15.  Basic characteristics of clinical feature: 'PATHOLOGY.N'

PATHOLOGY.N Mean (SD) 0.52 (0.77)
  N
  N0 268
  N1 86
  N2 57
  N3 5
     
  Significant markers N = 12
  pos. correlated 4
  neg. correlated 8
List of top 10 genes significantly correlated to 'PATHOLOGY.N' by Spearman correlation test

Table S16.  Get Full Table List of top 10 genes significantly correlated to 'PATHOLOGY.N' by Spearman correlation test

SpearmanCorr corrP Q
FAM73A -0.2672 3.138e-08 0.000559
TNFRSF10C -0.2655 3.823e-08 0.000681
MGC50559 -0.264 4.611e-08 0.000821
GABARAP -0.2477 3.111e-07 0.00554
MMP11 0.2381 8.944e-07 0.0159
LUZP2 0.2377 9.368e-07 0.0167
ZNF273 0.233 1.551e-06 0.0276
CNDP2 -0.2328 1.588e-06 0.0283
IFIT3 -0.2294 2.261e-06 0.0403
BIRC3 -0.2294 2.268e-06 0.0404

Figure S8.  Get High-res Image As an example, this figure shows the association of FAM73A to 'PATHOLOGY.N'. P value = 3.14e-08 with Spearman correlation analysis.

Clinical variable #9: 'PATHOLOGICSPREAD(M)'

33 genes related to 'PATHOLOGICSPREAD(M)'.

Table S17.  Basic characteristics of clinical feature: 'PATHOLOGICSPREAD(M)'

PATHOLOGICSPREAD(M) Labels N
  M0 399
  M1 43
  M1A 1
     
  Significant markers N = 33
List of top 10 genes differentially expressed by 'PATHOLOGICSPREAD(M)'

Table S18.  Get Full Table List of top 10 genes differentially expressed by 'PATHOLOGICSPREAD(M)'

ANOVA_P Q
KCNC2 1.858e-11 3.31e-07
TCF7 9.2e-09 0.000164
HCN1 2.486e-08 0.000443
C8ORF30A 7.085e-08 0.00126
NPAS2 1.24e-07 0.00221
SERPINA4 1.444e-07 0.00257
PLOD3 4.299e-07 0.00765
SPACA3 4.451e-07 0.00793
NIBP 4.699e-07 0.00837
F10 8.055e-07 0.0143

Figure S9.  Get High-res Image As an example, this figure shows the association of KCNC2 to 'PATHOLOGICSPREAD(M)'. P value = 1.86e-11 with ANOVA analysis.

Clinical variable #10: 'TUMOR.STAGE'

7068 genes related to 'TUMOR.STAGE'.

Table S19.  Basic characteristics of clinical feature: 'TUMOR.STAGE'

TUMOR.STAGE Mean (SD) 2.58 (0.93)
  N
  Stage 1 183
  Stage 2 160
  Stage 3 512
  Stage 4 119
     
  Significant markers N = 7068
  pos. correlated 3772
  neg. correlated 3296
List of top 10 genes significantly correlated to 'TUMOR.STAGE' by Spearman correlation test

Table S20.  Get Full Table List of top 10 genes significantly correlated to 'TUMOR.STAGE' by Spearman correlation test

SpearmanCorr corrP Q
FGF18 0.5153 3.592e-67 6.4e-63
WT1 0.4972 6.068e-62 1.08e-57
UGT1A6 -0.4794 4.111e-57 7.32e-53
ZNF503 0.4768 1.932e-56 3.44e-52
KITLG -0.4722 3.069e-55 5.47e-51
DOK7 0.4718 3.919e-55 6.98e-51
SOCS2 -0.4711 5.78e-55 1.03e-50
RIMS4 0.471 6.333e-55 1.13e-50
DNASE1L3 -0.4708 7.054e-55 1.26e-50
RBM38 0.4703 9.45e-55 1.68e-50

Figure S10.  Get High-res Image As an example, this figure shows the association of FGF18 to 'TUMOR.STAGE'. P value = 3.59e-67 with Spearman correlation analysis.

Clinical variable #11: 'RADIATIONS.RADIATION.REGIMENINDICATION'

5808 genes related to 'RADIATIONS.RADIATION.REGIMENINDICATION'.

Table S21.  Basic characteristics of clinical feature: 'RADIATIONS.RADIATION.REGIMENINDICATION'

RADIATIONS.RADIATION.REGIMENINDICATION Labels N
  NO 178
  YES 1391
     
  Significant markers N = 5808
  Higher in YES 2873
  Higher in NO 2935
List of top 10 genes differentially expressed by 'RADIATIONS.RADIATION.REGIMENINDICATION'

Table S22.  Get Full Table List of top 10 genes differentially expressed by 'RADIATIONS.RADIATION.REGIMENINDICATION'

T(pos if higher in 'YES') ttestP Q AUC
CYP3A5 22.57 1.156e-79 2.06e-75 0.7893
TCF21 21.82 1.425e-70 2.54e-66 0.7935
CYP3A4 20.46 2.734e-69 4.87e-65 0.7745
CYP3A7 20.52 3.27e-69 5.82e-65 0.7648
AADAC 20.33 8.409e-65 1.5e-60 0.7731
MAP2K7 21.2 1.802e-64 3.21e-60 0.8071
CCDC71 21.42 6.771e-64 1.21e-59 0.8175
REG1A 19.5 4.283e-61 7.63e-57 0.7759
PDZD3 17.28 1.942e-58 3.46e-54 0.6873
CHMP4B 19.98 1.429e-57 2.54e-53 0.8079

Figure S11.  Get High-res Image As an example, this figure shows the association of CYP3A5 to 'RADIATIONS.RADIATION.REGIMENINDICATION'. P value = 1.16e-79 with T-test analysis.

Clinical variable #12: 'NEOADJUVANT.THERAPY'

7886 genes related to 'NEOADJUVANT.THERAPY'.

Table S23.  Basic characteristics of clinical feature: 'NEOADJUVANT.THERAPY'

NEOADJUVANT.THERAPY Labels N
  NO 690
  YES 879
     
  Significant markers N = 7886
  Higher in YES 4017
  Higher in NO 3869
List of top 10 genes differentially expressed by 'NEOADJUVANT.THERAPY'

Table S24.  Get Full Table List of top 10 genes differentially expressed by 'NEOADJUVANT.THERAPY'

T(pos if higher in 'YES') ttestP Q AUC
ATP6V1B1 -21.83 5.409e-92 9.63e-88 0.7813
NPAS3 -21.63 8.136e-90 1.45e-85 0.7879
UGT1A6 21.58 2.396e-88 4.27e-84 0.7335
SCARA3 -21.3 2.065e-87 3.68e-83 0.7803
GPX2 21.34 3.912e-86 6.97e-82 0.7297
PEA15 -20.89 2.634e-85 4.69e-81 0.777
SLC27A6 -20.73 9.426e-84 1.68e-79 0.7782
ZNF667 -20.61 1.379e-83 2.46e-79 0.7671
NPR1 -20.83 1.871e-83 3.33e-79 0.7744
KLHL14 -20.77 7.636e-83 1.36e-78 0.7503

Figure S12.  Get High-res Image As an example, this figure shows the association of ATP6V1B1 to 'NEOADJUVANT.THERAPY'. P value = 5.41e-92 with T-test analysis.

Methods & Data
Input

  • Expresson data file = PANCAN8.medianexp.txt

  • Clinical data file = PANCAN8.clin.merged.picked.txt

  • Number of patients = 1569

  • Number of genes = 17814

  • Number of clinical features = 12

Survival analysis

For survival clinical features, Wald's test in univariate Cox regression analysis with proportional hazards model (Andersen and Gill 1982) was used to estimate the P values using the 'coxph' function in R. Kaplan-Meier survival curves were plot using the four quartile subgroups of patients based on expression levels

Correlation analysis

For continuous numerical clinical features, Spearman's rank correlation coefficients (Spearman 1904) and two-tailed P values were estimated using 'cor.test' function in R

ANOVA analysis

For multi-class clinical features (ordinal or nominal), one-way analysis of variance (Howell 2002) was applied to compare the log2-expression levels between different clinical classes using 'anova' function in R

Student's t-test analysis

For two-class clinical features, two-tailed Student's t test with unequal variance (Lehmann and Romano 2005) was applied to compare the log2-expression levels between the two clinical classes using 't.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

This is an experimental feature. The full results of the analysis summarized in this report can be downloaded from the TCGA Data Coordination Center.

References
[1] Andersen and Gill, Cox's regression model for counting processes, a large sample study, Annals of Statistics 10(4):1100-1120 (1982)
[2] Spearman, C, The proof and measurement of association between two things, Amer. J. Psychol 15:72-101 (1904)
[3] Howell, D, Statistical Methods for Psychology. (5th ed.), Duxbury Press:324-5 (2002)
[4] Lehmann and Romano, Testing Statistical Hypotheses (3E ed.), New York: Springer. ISBN 0387988645 (2005)
[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)
Copyright © 2012 Broad Institute TCGA GDAC as part of the TCGA Research Network. All rights reserved.
Made with Nozzle