Correlation between mRNAseq expression and clinical features
Prostate Adenocarcinoma (Primary solid tumor)
15 January 2014  |  analyses__2014_01_15
Maintainer Information
Citation Information
Maintained by Juok Cho (Broad Institute)
Cite as Broad Institute TCGA Genome Data Analysis Center (2014): Correlation between mRNAseq expression and clinical features. Broad Institute of MIT and Harvard. doi:10.7908/C1N29VDS
Overview
Introduction

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

Summary

Testing the association between 18226 genes and 13 clinical features across 185 samples, statistically thresholded by Q value < 0.05, 10 clinical features related to at least one genes.

  • 1 gene correlated to 'AGE'.

    • ADAP2|55803

  • 8 genes correlated to 'PATHOLOGY.T.STAGE'.

    • FA2H|79152 ,  FAM189A2|9413 ,  GMDS|2762 ,  ABAT|18 ,  CHRNA2|1135 ,  ...

  • 3 genes correlated to 'PATHOLOGY.N.STAGE'.

    • PRTN3|5657 ,  NKX6-2|84504 ,  CRHR2|1395

  • 2 genes correlated to 'COMPLETENESS.OF.RESECTION'.

    • GLP2R|9340 ,  CNGB3|54714

  • 19 genes correlated to 'GLEASON_SCORE_COMBINED'.

    • ZNF250|58500 ,  HERPUD1|9709 ,  FA2H|79152 ,  PARP2|10038 ,  B3GNT4|79369 ,  ...

  • 64 genes correlated to 'GLEASON_SCORE_PRIMARY'.

    • METTL6|131965 ,  KANK4|163782 ,  TMEM183A|92703 ,  FERMT1|55612 ,  SH3RF2|153769 ,  ...

  • 47 genes correlated to 'GLEASON_SCORE'.

    • FA2H|79152 ,  KIAA0319L|79932 ,  FMOD|2331 ,  HMGB2|3148 ,  HERPUD1|9709 ,  ...

  • 15 genes correlated to 'PSA_RESULT_PREOP'.

    • DDX19B|11269 ,  KANK4|163782 ,  PFKP|5214 ,  SH2D3A|10045 ,  ATCAY|85300 ,  ...

  • 378 genes correlated to 'DAYS_TO_PREOP_PSA'.

    • CCDC124|115098 ,  HSPBP1|23640 ,  LYSMD3|116068 ,  POLR2E|5434 ,  ALKBH7|84266 ,  ...

  • 12 genes correlated to 'DAYS_TO_PSA'.

    • HOOK2|29911 ,  PAN2|9924 ,  WDR73|84942 ,  RGL2|5863 ,  GTF2IRD2B|389524 ,  ...

  • No genes correlated to 'NUMBER.OF.LYMPH.NODES', 'GLEASON_SCORE_SECONDARY', and 'PSA_VALUE'.

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
AGE Spearman correlation test N=1 older N=1 younger N=0
PATHOLOGY T STAGE Spearman correlation test N=8 higher stage N=0 lower stage N=8
PATHOLOGY N STAGE t test N=3 class1 N=0 class0 N=3
COMPLETENESS OF RESECTION ANOVA test N=2        
NUMBER OF LYMPH NODES Spearman correlation test   N=0        
GLEASON_SCORE_COMBINED Spearman correlation test N=19 higher score N=13 lower score N=6
GLEASON_SCORE_PRIMARY Spearman correlation test N=64 higher score N=26 lower score N=38
GLEASON_SCORE_SECONDARY Spearman correlation test   N=0        
GLEASON_SCORE Spearman correlation test N=47 higher score N=33 lower score N=14
PSA_RESULT_PREOP Spearman correlation test N=15 higher psa_result_preop N=1 lower psa_result_preop N=14
DAYS_TO_PREOP_PSA Spearman correlation test N=378 higher days_to_preop_psa N=140 lower days_to_preop_psa N=238
PSA_VALUE Spearman correlation test   N=0        
DAYS_TO_PSA Spearman correlation test N=12 higher days_to_psa N=0 lower days_to_psa N=12
Clinical variable #1: 'AGE'

One gene related to 'AGE'.

Table S1.  Basic characteristics of clinical feature: 'AGE'

AGE Mean (SD) 60.28 (7)
  Significant markers N = 1
  pos. correlated 1
  neg. correlated 0
List of one gene significantly correlated to 'AGE' by Spearman correlation test

Table S2.  Get Full Table List of one gene significantly correlated to 'AGE' by Spearman correlation test

SpearmanCorr corrP Q
ADAP2|55803 0.3603 5.461e-07 0.00995

Figure S1.  Get High-res Image As an example, this figure shows the association of ADAP2|55803 to 'AGE'. P value = 5.46e-07 with Spearman correlation analysis. The straight line presents the best linear regression.

Clinical variable #2: 'PATHOLOGY.T.STAGE'

8 genes related to 'PATHOLOGY.T.STAGE'.

Table S3.  Basic characteristics of clinical feature: 'PATHOLOGY.T.STAGE'

PATHOLOGY.T.STAGE Mean (SD) 2.59 (0.55)
  N
  2 80
  3 98
  4 5
     
  Significant markers N = 8
  pos. correlated 0
  neg. correlated 8
List of 8 genes significantly correlated to 'PATHOLOGY.T.STAGE' by Spearman correlation test

Table S4.  Get Full Table List of 8 genes significantly correlated to 'PATHOLOGY.T.STAGE' by Spearman correlation test

SpearmanCorr corrP Q
FA2H|79152 -0.3657 3.58e-07 0.00652
FAM189A2|9413 -0.3573 6.877e-07 0.0125
GMDS|2762 -0.3509 1.115e-06 0.0203
ABAT|18 -0.3507 1.131e-06 0.0206
CHRNA2|1135 -0.3476 1.429e-06 0.026
EPHX2|2053 -0.3446 1.775e-06 0.0323
PLA2G4F|255189 -0.3435 1.932e-06 0.0352
KIAA0319L|79932 -0.3434 1.944e-06 0.0354

Figure S2.  Get High-res Image As an example, this figure shows the association of FA2H|79152 to 'PATHOLOGY.T.STAGE'. P value = 3.58e-07 with Spearman correlation analysis.

Clinical variable #3: 'PATHOLOGY.N.STAGE'

3 genes related to 'PATHOLOGY.N.STAGE'.

Table S5.  Basic characteristics of clinical feature: 'PATHOLOGY.N.STAGE'

PATHOLOGY.N.STAGE Labels N
  class0 145
  class1 17
     
  Significant markers N = 3
  Higher in class1 0
  Higher in class0 3
List of 3 genes differentially expressed by 'PATHOLOGY.N.STAGE'

Table S6.  Get Full Table List of 3 genes differentially expressed by 'PATHOLOGY.N.STAGE'

T(pos if higher in 'class1') ttestP Q AUC
PRTN3|5657 -7.13 1.625e-08 0.000296 0.7916
NKX6-2|84504 -6.67 3.838e-08 0.000699 0.8794
CRHR2|1395 -9.19 1.315e-07 0.0024 0.9476

Figure S3.  Get High-res Image As an example, this figure shows the association of PRTN3|5657 to 'PATHOLOGY.N.STAGE'. P value = 1.62e-08 with T-test analysis.

Clinical variable #4: 'COMPLETENESS.OF.RESECTION'

2 genes related to 'COMPLETENESS.OF.RESECTION'.

Table S7.  Basic characteristics of clinical feature: 'COMPLETENESS.OF.RESECTION'

COMPLETENESS.OF.RESECTION Labels N
  R0 135
  R1 37
  RX 3
     
  Significant markers N = 2
List of 2 genes differentially expressed by 'COMPLETENESS.OF.RESECTION'

Table S8.  Get Full Table List of 2 genes differentially expressed by 'COMPLETENESS.OF.RESECTION'

ANOVA_P Q
GLP2R|9340 2.964e-10 5.4e-06
CNGB3|54714 5.504e-07 0.01

Figure S4.  Get High-res Image As an example, this figure shows the association of GLP2R|9340 to 'COMPLETENESS.OF.RESECTION'. P value = 2.96e-10 with ANOVA analysis.

Clinical variable #5: 'NUMBER.OF.LYMPH.NODES'

No gene related to 'NUMBER.OF.LYMPH.NODES'.

Table S9.  Basic characteristics of clinical feature: 'NUMBER.OF.LYMPH.NODES'

NUMBER.OF.LYMPH.NODES Mean (SD) 0.19 (0.69)
  Significant markers N = 0
Clinical variable #6: 'GLEASON_SCORE_COMBINED'

19 genes related to 'GLEASON_SCORE_COMBINED'.

Table S10.  Basic characteristics of clinical feature: 'GLEASON_SCORE_COMBINED'

GLEASON_SCORE_COMBINED Mean (SD) 7.26 (0.77)
  Score N
  6 12
  7 135
  8 17
  9 19
  10 2
     
  Significant markers N = 19
  pos. correlated 13
  neg. correlated 6
List of top 10 genes significantly correlated to 'GLEASON_SCORE_COMBINED' by Spearman correlation test

Table S11.  Get Full Table List of top 10 genes significantly correlated to 'GLEASON_SCORE_COMBINED' by Spearman correlation test

SpearmanCorr corrP Q
ZNF250|58500 0.4021 1.399e-08 0.000255
HERPUD1|9709 -0.3893 4.315e-08 0.000786
FA2H|79152 -0.379 1.039e-07 0.00189
PARP2|10038 0.3734 1.651e-07 0.00301
B3GNT4|79369 0.3758 1.716e-07 0.00313
DPP4|1803 -0.3685 2.462e-07 0.00449
CNIH2|254263 0.3662 2.95e-07 0.00537
HMGB2|3148 0.3654 3.138e-07 0.00572
STMN1|3925 0.3639 3.551e-07 0.00647
FMOD|2331 -0.3566 6.273e-07 0.0114

Figure S5.  Get High-res Image As an example, this figure shows the association of ZNF250|58500 to 'GLEASON_SCORE_COMBINED'. P value = 1.4e-08 with Spearman correlation analysis.

Clinical variable #7: 'GLEASON_SCORE_PRIMARY'

64 genes related to 'GLEASON_SCORE_PRIMARY'.

Table S12.  Basic characteristics of clinical feature: 'GLEASON_SCORE_PRIMARY'

GLEASON_SCORE_PRIMARY Mean (SD) 3.48 (0.57)
  Score N
  2 1
  3 101
  4 77
  5 6
     
  Significant markers N = 64
  pos. correlated 26
  neg. correlated 38
List of top 10 genes significantly correlated to 'GLEASON_SCORE_PRIMARY' by Spearman correlation test

Table S13.  Get Full Table List of top 10 genes significantly correlated to 'GLEASON_SCORE_PRIMARY' by Spearman correlation test

SpearmanCorr corrP Q
METTL6|131965 0.4432 2.659e-10 4.85e-06
KANK4|163782 -0.4133 5.023e-09 9.15e-05
TMEM183A|92703 0.4039 1.192e-08 0.000217
FERMT1|55612 -0.4008 1.58e-08 0.000288
SH3RF2|153769 -0.3965 2.307e-08 0.00042
ZNRF3|84133 -0.3925 3.283e-08 0.000598
SH3BGRL2|83699 -0.3866 5.479e-08 0.000998
C20ORF20|55257 0.3788 1.059e-07 0.00193
ZDHHC2|51201 -0.3775 1.179e-07 0.00215
MTERFD1|51001 0.3765 1.276e-07 0.00232

Figure S6.  Get High-res Image As an example, this figure shows the association of METTL6|131965 to 'GLEASON_SCORE_PRIMARY'. P value = 2.66e-10 with Spearman correlation analysis.

Clinical variable #8: 'GLEASON_SCORE_SECONDARY'

No gene related to 'GLEASON_SCORE_SECONDARY'.

Table S14.  Basic characteristics of clinical feature: 'GLEASON_SCORE_SECONDARY'

GLEASON_SCORE_SECONDARY Mean (SD) 3.79 (0.61)
  Score N
  3 58
  4 108
  5 19
     
  Significant markers N = 0
Clinical variable #9: 'GLEASON_SCORE'

47 genes related to 'GLEASON_SCORE'.

Table S15.  Basic characteristics of clinical feature: 'GLEASON_SCORE'

GLEASON_SCORE Mean (SD) 7.31 (0.81)
  Score N
  6 12
  7 130
  8 19
  9 22
  10 2
     
  Significant markers N = 47
  pos. correlated 33
  neg. correlated 14
List of top 10 genes significantly correlated to 'GLEASON_SCORE' by Spearman correlation test

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

SpearmanCorr corrP Q
FA2H|79152 -0.4238 1.838e-09 3.35e-05
KIAA0319L|79932 -0.3997 1.741e-08 0.000317
FMOD|2331 -0.3902 4.01e-08 0.000731
HMGB2|3148 0.3844 6.584e-08 0.0012
HERPUD1|9709 -0.3831 7.371e-08 0.00134
ZNF250|58500 0.3826 7.655e-08 0.00139
DSN1|79980 0.382 8.052e-08 0.00147
CNIH2|254263 0.3793 1.012e-07 0.00184
CBX1|10951 0.3786 1.075e-07 0.00196
PARP2|10038 0.3756 1.377e-07 0.00251

Figure S7.  Get High-res Image As an example, this figure shows the association of FA2H|79152 to 'GLEASON_SCORE'. P value = 1.84e-09 with Spearman correlation analysis.

Clinical variable #10: 'PSA_RESULT_PREOP'

15 genes related to 'PSA_RESULT_PREOP'.

Table S17.  Basic characteristics of clinical feature: 'PSA_RESULT_PREOP'

PSA_RESULT_PREOP Mean (SD) 10.32 (10)
  Significant markers N = 15
  pos. correlated 1
  neg. correlated 14
List of top 10 genes significantly correlated to 'PSA_RESULT_PREOP' by Spearman correlation test

Table S18.  Get Full Table List of top 10 genes significantly correlated to 'PSA_RESULT_PREOP' by Spearman correlation test

SpearmanCorr corrP Q
DDX19B|11269 -0.412 6.837e-09 0.000125
KANK4|163782 -0.3972 2.59e-08 0.000472
PFKP|5214 -0.3771 1.429e-07 0.0026
SH2D3A|10045 -0.3711 2.33e-07 0.00425
ATCAY|85300 -0.369 2.754e-07 0.00502
PTPLA|9200 -0.3492 1.268e-06 0.0231
ALDH1A2|8854 -0.3488 1.301e-06 0.0237
CPNE6|9362 -0.3551 1.424e-06 0.026
CA11|770 -0.3471 1.484e-06 0.027
R3HCC1|203069 -0.3461 1.596e-06 0.0291

Figure S8.  Get High-res Image As an example, this figure shows the association of DDX19B|11269 to 'PSA_RESULT_PREOP'. P value = 6.84e-09 with Spearman correlation analysis. The straight line presents the best linear regression.

Clinical variable #11: 'DAYS_TO_PREOP_PSA'

378 genes related to 'DAYS_TO_PREOP_PSA'.

Table S19.  Basic characteristics of clinical feature: 'DAYS_TO_PREOP_PSA'

DAYS_TO_PREOP_PSA Mean (SD) -1.53 (100)
  Significant markers N = 378
  pos. correlated 140
  neg. correlated 238
List of top 10 genes significantly correlated to 'DAYS_TO_PREOP_PSA' by Spearman correlation test

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

SpearmanCorr corrP Q
CCDC124|115098 -0.4309 1.553e-09 2.83e-05
HSPBP1|23640 -0.4206 4.148e-09 7.56e-05
LYSMD3|116068 0.4205 4.163e-09 7.59e-05
POLR2E|5434 -0.4162 6.241e-09 0.000114
ALKBH7|84266 -0.4148 7.047e-09 0.000128
JOSD2|126119 -0.4139 7.666e-09 0.00014
ARFRP1|10139 -0.4128 8.49e-09 0.000155
MED25|81857 -0.4124 8.801e-09 0.00016
RNF138|51444 0.4104 1.054e-08 0.000192
RGPD4|285190 0.4103 1.063e-08 0.000194

Figure S9.  Get High-res Image As an example, this figure shows the association of CCDC124|115098 to 'DAYS_TO_PREOP_PSA'. P value = 1.55e-09 with Spearman correlation analysis. The straight line presents the best linear regression.

Clinical variable #12: 'PSA_VALUE'

No gene related to 'PSA_VALUE'.

Table S21.  Basic characteristics of clinical feature: 'PSA_VALUE'

PSA_VALUE Mean (SD) 1.46 (4.5)
  Significant markers N = 0
Clinical variable #13: 'DAYS_TO_PSA'

12 genes related to 'DAYS_TO_PSA'.

Table S22.  Basic characteristics of clinical feature: 'DAYS_TO_PSA'

DAYS_TO_PSA Mean (SD) 534.61 (500)
  Significant markers N = 12
  pos. correlated 0
  neg. correlated 12
List of top 10 genes significantly correlated to 'DAYS_TO_PSA' by Spearman correlation test

Table S23.  Get Full Table List of top 10 genes significantly correlated to 'DAYS_TO_PSA' by Spearman correlation test

SpearmanCorr corrP Q
HOOK2|29911 -0.5054 2.821e-12 5.14e-08
PAN2|9924 -0.4028 6.22e-08 0.00113
WDR73|84942 -0.375 5.483e-07 0.00999
RGL2|5863 -0.3717 7.015e-07 0.0128
GTF2IRD2B|389524 -0.3628 1.347e-06 0.0245
TSSK3|81629 -0.3619 1.436e-06 0.0262
LOC100128842|100128842 -0.3575 1.966e-06 0.0358
DHX16|8449 -0.3574 1.981e-06 0.0361
ANKMY1|51281 -0.3572 2.007e-06 0.0366
C15ORF62|643338 -0.3569 2.041e-06 0.0372

Figure S10.  Get High-res Image As an example, this figure shows the association of HOOK2|29911 to 'DAYS_TO_PSA'. P value = 2.82e-12 with Spearman correlation analysis. The straight line presents the best linear regression.

Methods & Data
Input
  • Expresson data file = PRAD-TP.uncv2.mRNAseq_RSEM_normalized_log2.txt

  • Clinical data file = PRAD-TP.merged_data.txt

  • Number of patients = 185

  • Number of genes = 18226

  • Number of clinical features = 13

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

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

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

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] Spearman, C, The proof and measurement of association between two things, Amer. J. Psychol 15:72-101 (1904)
[2] Lehmann and Romano, Testing Statistical Hypotheses (3E ed.), New York: Springer. ISBN 0387988645 (2005)
[3] Howell, D, Statistical Methods for Psychology. (5th ed.), Duxbury Press:324-5 (2002)
[4] 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)