Correlation between mRNAseq expression and clinical features
Head and Neck Squamous Cell Carcinoma (Primary solid tumor)
16 April 2014  |  analyses__2014_04_16
Maintainer Information
Citation Information
doi:10.7908/C1FJ2FD4
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/C1FJ2FD4
Overview
Introduction

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

Summary

Testing the association between 18370 genes and 12 clinical features across 399 samples, statistically thresholded by Q value < 0.05, 11 clinical features related to at least one genes.

  • 6 genes correlated to 'Time to Death'.

    • CELSR3|1951 ,  FGD3|89846 ,  ZNF266|10781 ,  POLR2C|5432 ,  LOC728989|728989 ,  ...

  • 2 genes correlated to 'AGE'.

    • RNASE3|6037 ,  HN1L|90861

  • 1 gene correlated to 'NEOPLASM.DISEASESTAGE'.

    • ZC3HAV1|56829

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

    • LRRC8A|56262 ,  MSL3|10943 ,  C5ORF20|140947 ,  TCEANC|170082 ,  CX3CR1|1524 ,  ...

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

    • PDXP|57026 ,  MPZL2|10205 ,  TGM1|7051 ,  CRCT1|54544 ,  APOC1|341 ,  ...

  • 189 genes correlated to 'PATHOLOGY.M.STAGE'.

    • LOC649330|649330 ,  SNHG12|85028 ,  PIK3CG|5294 ,  SNHG3|8420 ,  ACYP1|97 ,  ...

  • 111 genes correlated to 'GENDER'.

    • XIST|7503 ,  ZFY|7544 ,  PRKY|5616 ,  RPS4Y1|6192 ,  DDX3Y|8653 ,  ...

  • 77 genes correlated to 'HISTOLOGICAL.TYPE'.

    • GPR45|11250 ,  DQX1|165545 ,  THBS3|7059 ,  MAL2|114569 ,  EHD2|30846 ,  ...

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

    • NFATC2|4773 ,  UHMK1|127933 ,  ZNF366|167465 ,  ZNF827|152485 ,  RAD54L2|23132 ,  ...

  • 30 genes correlated to 'NUMBERPACKYEARSSMOKED'.

    • TMTC3|160418 ,  TMSL3|7117 ,  LASS6|253782 ,  DPH3|285381 ,  LRP4|4038 ,  ...

  • 49 genes correlated to 'NUMBER.OF.LYMPH.NODES'.

    • ARL4C|10123 ,  ALS2CL|259173 ,  PPEF1|5475 ,  MPZL2|10205 ,  TGM1|7051 ,  ...

  • No genes correlated to 'YEAROFTOBACCOSMOKINGONSET'

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=6 shorter survival N=2 longer survival N=4
AGE Spearman correlation test N=2 older N=1 younger N=1
NEOPLASM DISEASESTAGE ANOVA test N=1        
PATHOLOGY T STAGE Spearman correlation test N=16 higher stage N=5 lower stage N=11
PATHOLOGY N STAGE Spearman correlation test N=25 higher stage N=6 lower stage N=19
PATHOLOGY M STAGE t test N=189 mx N=119 m0 N=70
GENDER t test N=111 male N=60 female N=51
HISTOLOGICAL TYPE ANOVA test N=77        
RADIATIONS RADIATION REGIMENINDICATION t test N=68 yes N=10 no N=58
NUMBERPACKYEARSSMOKED Spearman correlation test N=30 higher numberpackyearssmoked N=24 lower numberpackyearssmoked N=6
YEAROFTOBACCOSMOKINGONSET Spearman correlation test   N=0        
NUMBER OF LYMPH NODES Spearman correlation test N=49 higher number.of.lymph.nodes N=23 lower number.of.lymph.nodes N=26
Clinical variable #1: 'Time to Death'

6 genes related to 'Time to Death'.

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

Time to Death Duration (Months) 0.1-211 (median=16.7)
  censored N = 243
  death N = 151
     
  Significant markers N = 6
  associated with shorter survival 2
  associated with longer survival 4
List of 6 genes significantly associated with 'Time to Death' by Cox regression test

Table S2.  Get Full Table List of 6 genes significantly associated with 'Time to Death' by Cox regression test

HazardRatio Wald_P Q C_index
CELSR3|1951 0.76 5.226e-07 0.0096 0.373
FGD3|89846 0.7 5.246e-07 0.0096 0.361
ZNF266|10781 0.5 7.675e-07 0.014 0.379
POLR2C|5432 2.4 1.603e-06 0.029 0.608
LOC728989|728989 0.7 1.665e-06 0.031 0.38
CYB5B|80777 2.1 2.623e-06 0.048 0.59

Figure S1.  Get High-res Image As an example, this figure shows the association of CELSR3|1951 to 'Time to Death'. four curves present the cumulative survival rates of 4 quartile subsets of patients. P value = 5.23e-07 with univariate Cox regression analysis using continuous log-2 expression values.

Clinical variable #2: 'AGE'

2 genes related to 'AGE'.

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

AGE Mean (SD) 60.79 (12)
  Significant markers N = 2
  pos. correlated 1
  neg. correlated 1
List of 2 genes significantly correlated to 'AGE' by Spearman correlation test

Table S4.  Get Full Table List of 2 genes significantly correlated to 'AGE' by Spearman correlation test

SpearmanCorr corrP Q
RNASE3|6037 0.2975 1.376e-06 0.0253
HN1L|90861 -0.2345 2.243e-06 0.0412

Figure S2.  Get High-res Image As an example, this figure shows the association of RNASE3|6037 to 'AGE'. P value = 1.38e-06 with Spearman correlation analysis. The straight line presents the best linear regression.

Clinical variable #3: 'NEOPLASM.DISEASESTAGE'

One gene related to 'NEOPLASM.DISEASESTAGE'.

Table S5.  Basic characteristics of clinical feature: 'NEOPLASM.DISEASESTAGE'

NEOPLASM.DISEASESTAGE Labels N
  STAGE I 24
  STAGE II 61
  STAGE III 65
  STAGE IVA 187
  STAGE IVB 8
     
  Significant markers N = 1
List of one gene differentially expressed by 'NEOPLASM.DISEASESTAGE'

Table S6.  Get Full Table List of one gene differentially expressed by 'NEOPLASM.DISEASESTAGE'

ANOVA_P Q
ZC3HAV1|56829 1.863e-06 0.0342

Figure S3.  Get High-res Image As an example, this figure shows the association of ZC3HAV1|56829 to 'NEOPLASM.DISEASESTAGE'. P value = 1.86e-06 with ANOVA analysis.

Clinical variable #4: 'PATHOLOGY.T.STAGE'

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

Table S7.  Basic characteristics of clinical feature: 'PATHOLOGY.T.STAGE'

PATHOLOGY.T.STAGE Mean (SD) 2.85 (1)
  N
  1 38
  2 103
  3 83
  4 126
     
  Significant markers N = 16
  pos. correlated 5
  neg. correlated 11
List of top 10 genes significantly correlated to 'PATHOLOGY.T.STAGE' by Spearman correlation test

Table S8.  Get Full Table List of top 10 genes significantly correlated to 'PATHOLOGY.T.STAGE' by Spearman correlation test

SpearmanCorr corrP Q
LRRC8A|56262 0.3016 8.57e-09 0.000157
MSL3|10943 -0.2863 4.991e-08 0.000917
C5ORF20|140947 -0.2859 6.512e-08 0.0012
TCEANC|170082 -0.2738 1.953e-07 0.00359
CX3CR1|1524 -0.2723 2.381e-07 0.00437
WDR37|22884 -0.2719 2.399e-07 0.00441
ADM|133 0.2706 2.728e-07 0.00501
FRMD4A|55691 -0.264 5.431e-07 0.00997
DDIT4|54541 0.2572 1.076e-06 0.0198
LTB|4050 -0.256 1.214e-06 0.0223

Figure S4.  Get High-res Image As an example, this figure shows the association of LRRC8A|56262 to 'PATHOLOGY.T.STAGE'. P value = 8.57e-09 with Spearman correlation analysis.

Clinical variable #5: 'PATHOLOGY.N.STAGE'

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

Table S9.  Basic characteristics of clinical feature: 'PATHOLOGY.N.STAGE'

PATHOLOGY.N.STAGE Mean (SD) 1.03 (0.95)
  N
  0 131
  1 51
  2 125
  3 7
     
  Significant markers N = 25
  pos. correlated 6
  neg. correlated 19
List of top 10 genes significantly correlated to 'PATHOLOGY.N.STAGE' by Spearman correlation test

Table S10.  Get Full Table List of top 10 genes significantly correlated to 'PATHOLOGY.N.STAGE' by Spearman correlation test

SpearmanCorr corrP Q
PDXP|57026 0.3133 1.411e-08 0.000259
MPZL2|10205 -0.3103 1.958e-08 0.00036
TGM1|7051 -0.2974 7.805e-08 0.00143
CRCT1|54544 -0.2945 1.277e-07 0.00235
APOC1|341 0.2868 2.339e-07 0.0043
SPRR2F|6705 -0.289 2.348e-07 0.00431
GCOM1|145781 -0.2811 4.273e-07 0.00785
D4S234E|27065 -0.2784 5.35e-07 0.00982
ALS2CL|259173 -0.2777 5.745e-07 0.0105
EPS8L2|64787 -0.2773 5.938e-07 0.0109

Figure S5.  Get High-res Image As an example, this figure shows the association of PDXP|57026 to 'PATHOLOGY.N.STAGE'. P value = 1.41e-08 with Spearman correlation analysis.

Clinical variable #6: 'PATHOLOGY.M.STAGE'

189 genes related to 'PATHOLOGY.M.STAGE'.

Table S11.  Basic characteristics of clinical feature: 'PATHOLOGY.M.STAGE'

PATHOLOGY.M.STAGE Labels N
  M0 100
  MX 40
     
  Significant markers N = 189
  Higher in MX 119
  Higher in M0 70
List of top 10 genes differentially expressed by 'PATHOLOGY.M.STAGE'

Table S12.  Get Full Table List of top 10 genes differentially expressed by 'PATHOLOGY.M.STAGE'

T(pos if higher in 'MX') ttestP Q AUC
LOC649330|649330 7.75 4.089e-11 7.51e-07 0.8447
SNHG12|85028 7.13 7.312e-11 1.34e-06 0.79
PIK3CG|5294 -7.33 1.775e-10 3.26e-06 0.824
SNHG3|8420 6.72 6.231e-10 1.14e-05 0.7753
ACYP1|97 6.67 1.867e-09 3.43e-05 0.7952
ITGA8|8516 -6.72 2.757e-09 5.06e-05 0.8086
PABPN1|8106 6.63 2.864e-09 5.26e-05 0.8
CCNL2|81669 6.42 3.553e-09 6.53e-05 0.7678
ZNF692|55657 6.4 4.778e-09 8.77e-05 0.7797
RABGGTB|5876 6.33 5.767e-09 0.000106 0.7647

Figure S6.  Get High-res Image As an example, this figure shows the association of LOC649330|649330 to 'PATHOLOGY.M.STAGE'. P value = 4.09e-11 with T-test analysis.

Clinical variable #7: 'GENDER'

111 genes related to 'GENDER'.

Table S13.  Basic characteristics of clinical feature: 'GENDER'

GENDER Labels N
  FEMALE 112
  MALE 287
     
  Significant markers N = 111
  Higher in MALE 60
  Higher in FEMALE 51
List of top 10 genes differentially expressed by 'GENDER'

Table S14.  Get Full Table List of top 10 genes differentially expressed by 'GENDER'

T(pos if higher in 'MALE') ttestP Q AUC
XIST|7503 -48.14 1.109e-139 2.04e-135 0.9981
ZFY|7544 46.31 1.589e-114 2.92e-110 0.9965
PRKY|5616 33.52 5.36e-91 9.84e-87 0.9942
RPS4Y1|6192 44.28 1.694e-84 3.11e-80 0.9998
DDX3Y|8653 45.03 7.625e-78 1.4e-73 0.9967
USP9Y|8287 39.47 5.326e-66 9.78e-62 0.9987
KDM5D|8284 39.4 2.674e-56 4.91e-52 0.9981
UTY|7404 34.95 7.731e-51 1.42e-46 0.995
TSIX|9383 -19.11 5.569e-47 1.02e-42 0.97
NLGN4Y|22829 23.52 1.153e-33 2.12e-29 0.9852

Figure S7.  Get High-res Image As an example, this figure shows the association of XIST|7503 to 'GENDER'. P value = 1.11e-139 with T-test analysis.

Clinical variable #8: 'HISTOLOGICAL.TYPE'

77 genes related to 'HISTOLOGICAL.TYPE'.

Table S15.  Basic characteristics of clinical feature: 'HISTOLOGICAL.TYPE'

HISTOLOGICAL.TYPE Labels N
  HEAD AND NECK SQUAMOUS CELL CARCINOMA 392
  HEAD AND NECK SQUAMOUS CELL CARCINOMA SPINDLE CELL VARIANT 1
  HEAD AND NECK SQUAMOUS CELL CARCINOMA BASALOID TYPE 6
     
  Significant markers N = 77
List of top 10 genes differentially expressed by 'HISTOLOGICAL.TYPE'

Table S16.  Get Full Table List of top 10 genes differentially expressed by 'HISTOLOGICAL.TYPE'

ANOVA_P Q
GPR45|11250 4.181e-13 7.67e-09
DQX1|165545 4.792e-13 8.8e-09
THBS3|7059 2.107e-11 3.87e-07
MAL2|114569 5.282e-11 9.69e-07
EHD2|30846 1.214e-10 2.23e-06
TOM1L1|10040 4.232e-10 7.77e-06
MARVELD2|153562 6.436e-10 1.18e-05
KLKP1|606293 9.383e-10 1.72e-05
MAP7D3|79649 2.587e-09 4.75e-05
CBLC|23624 1.321e-08 0.000242

Figure S8.  Get High-res Image As an example, this figure shows the association of GPR45|11250 to 'HISTOLOGICAL.TYPE'. P value = 4.18e-13 with ANOVA analysis.

Clinical variable #9: 'RADIATIONS.RADIATION.REGIMENINDICATION'

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

Table S17.  Basic characteristics of clinical feature: 'RADIATIONS.RADIATION.REGIMENINDICATION'

RADIATIONS.RADIATION.REGIMENINDICATION Labels N
  NO 81
  YES 318
     
  Significant markers N = 68
  Higher in YES 10
  Higher in NO 58
List of top 10 genes differentially expressed by 'RADIATIONS.RADIATION.REGIMENINDICATION'

Table S18.  Get Full Table List of top 10 genes differentially expressed by 'RADIATIONS.RADIATION.REGIMENINDICATION'

T(pos if higher in 'YES') ttestP Q AUC
NFATC2|4773 -6.89 1.486e-10 2.73e-06 0.7213
UHMK1|127933 -6.33 1.559e-09 2.86e-05 0.6813
ZNF366|167465 -6.36 1.727e-09 3.17e-05 0.6827
ZNF827|152485 -6.34 1.928e-09 3.54e-05 0.702
RAD54L2|23132 -6.25 3.446e-09 6.33e-05 0.6933
NHSL2|340527 -6.07 1.201e-08 0.000221 0.7031
CTDSP2|10106 -6.03 1.474e-08 0.000271 0.6982
ITPRIPL2|162073 -5.86 2.941e-08 0.00054 0.68
PLXNC1|10154 -5.77 4.256e-08 0.000781 0.6845
TMPPE|643853 -5.71 6.453e-08 0.00118 0.6877

Figure S9.  Get High-res Image As an example, this figure shows the association of NFATC2|4773 to 'RADIATIONS.RADIATION.REGIMENINDICATION'. P value = 1.49e-10 with T-test analysis.

Clinical variable #10: 'NUMBERPACKYEARSSMOKED'

30 genes related to 'NUMBERPACKYEARSSMOKED'.

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

NUMBERPACKYEARSSMOKED Mean (SD) 46.91 (38)
  Significant markers N = 30
  pos. correlated 24
  neg. correlated 6
List of top 10 genes significantly correlated to 'NUMBERPACKYEARSSMOKED' by Spearman correlation test

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

SpearmanCorr corrP Q
TMTC3|160418 0.3769 4.867e-09 8.94e-05
TMSL3|7117 -0.3589 2.834e-08 0.000521
LASS6|253782 0.3447 1.057e-07 0.00194
DPH3|285381 -0.3326 3.067e-07 0.00563
LRP4|4038 0.3326 3.082e-07 0.00566
ZNF462|58499 0.3306 3.651e-07 0.0067
C12ORF66|144577 0.3297 3.931e-07 0.00722
ZNRF3|84133 0.3288 4.275e-07 0.00785
GLI2|2736 0.3265 5.202e-07 0.00955
ANAPC1|64682 0.3235 6.676e-07 0.0123

Figure S10.  Get High-res Image As an example, this figure shows the association of TMTC3|160418 to 'NUMBERPACKYEARSSMOKED'. P value = 4.87e-09 with Spearman correlation analysis. The straight line presents the best linear regression.

Clinical variable #11: 'YEAROFTOBACCOSMOKINGONSET'

No gene related to 'YEAROFTOBACCOSMOKINGONSET'.

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

YEAROFTOBACCOSMOKINGONSET Mean (SD) 1966.49 (13)
  Significant markers N = 0
Clinical variable #12: 'NUMBER.OF.LYMPH.NODES'

49 genes related to 'NUMBER.OF.LYMPH.NODES'.

Table S22.  Basic characteristics of clinical feature: 'NUMBER.OF.LYMPH.NODES'

NUMBER.OF.LYMPH.NODES Mean (SD) 2.34 (4.7)
  Significant markers N = 49
  pos. correlated 23
  neg. correlated 26
List of top 10 genes significantly correlated to 'NUMBER.OF.LYMPH.NODES' by Spearman correlation test

Table S23.  Get Full Table List of top 10 genes significantly correlated to 'NUMBER.OF.LYMPH.NODES' by Spearman correlation test

SpearmanCorr corrP Q
ARL4C|10123 0.3275 4.404e-09 8.09e-05
ALS2CL|259173 -0.3165 1.514e-08 0.000278
PPEF1|5475 0.3146 2.188e-08 0.000402
MPZL2|10205 -0.3105 2.918e-08 0.000536
TGM1|7051 -0.2918 2.02e-07 0.00371
SLC13A4|26266 -0.2904 2.547e-07 0.00468
GSC|145258 0.2894 2.806e-07 0.00515
ZNF275|10838 0.2881 2.925e-07 0.00537
DEFB1|1672 -0.2878 3.014e-07 0.00553
EPS8L2|64787 -0.2872 3.211e-07 0.00589

Figure S11.  Get High-res Image As an example, this figure shows the association of ARL4C|10123 to 'NUMBER.OF.LYMPH.NODES'. P value = 4.4e-09 with Spearman correlation analysis. The straight line presents the best linear regression.

Methods & Data
Input

  • Expresson data file = HNSC-TP.uncv2.mRNAseq_RSEM_normalized_log2.txt

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

  • Number of patients = 399

  • Number of genes = 18370

  • 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

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] 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 © 2014 Broad Institute TCGA GDAC as part of the TCGA Research Network. All rights reserved.
Made with Nozzle