Correlation between gene methylation status and clinical features
Lung Adenocarcinoma (Primary solid tumor)
21 April 2013  |  analyses__2013_04_21
Maintainer Information
Citation Information
doi:10.7908/C19G5JSH
Maintained by Juok Cho (Broad Institute)
Cite as Broad Institute TCGA Genome Data Analysis Center (2013): Lung Adenocarcinoma (Primary solid tumor cohort) - 21 April 2013: Correlation between gene methylation status and clinical features. Broad Institute of MIT and Harvard. doi:10.7908/C19G5JSH
Overview
Introduction

This pipeline uses various statistical tests to identify genes whose promoter methylation levels correlated to selected clinical features.

Summary

Testing the association between 17358 genes and 14 clinical features across 286 samples, statistically thresholded by Q value < 0.05, 9 clinical features related to at least one genes.

  • 18 genes correlated to 'Time to Death'.

    • FUT1 ,  LARP1 ,  LOC650368 ,  RRAGD ,  RAD52 ,  ...

  • 1 gene correlated to 'AGE'.

    • KIF15

  • 84 genes correlated to 'GENDER'.

    • KIF4B ,  EIF4A1 ,  RNASEH2C ,  FRG1B ,  GPN1 ,  ...

  • 99 genes correlated to 'HISTOLOGICAL.TYPE'.

    • MURC ,  GNG10 ,  KRT39 ,  KAT5 ,  TTC32 ,  ...

  • 1 gene correlated to 'PATHOLOGY.N'.

    • KLHDC9

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

    • C2ORF73 ,  CARD6 ,  POT1 ,  ZSCAN20 ,  FAM190A

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

    • PRELP ,  CCNI ,  ZNF642 ,  MIIP ,  MAT2B ,  ...

  • 6 genes correlated to 'TOBACCOSMOKINGHISTORYINDICATOR'.

    • SRM ,  ZC3HAV1L ,  GIT1 ,  CMTM5 ,  C14ORF115 ,  ...

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

    • ANKIB1 ,  CDK13 ,  FAM177A1 ,  SLC6A16 ,  CWC22 ,  ...

  • No genes correlated to 'KARNOFSKY.PERFORMANCE.SCORE', 'PATHOLOGY.T', 'TUMOR.STAGE', 'NUMBERPACKYEARSSMOKED', and '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=18 shorter survival N=7 longer survival N=11
AGE Spearman correlation test N=1 older N=1 younger N=0
GENDER t test N=84 male N=7 female N=77
KARNOFSKY PERFORMANCE SCORE Spearman correlation test   N=0        
HISTOLOGICAL TYPE ANOVA test N=99        
PATHOLOGY T Spearman correlation test   N=0        
PATHOLOGY N Spearman correlation test N=1 higher pN N=1 lower pN N=0
PATHOLOGICSPREAD(M) ANOVA test N=5        
TUMOR STAGE Spearman correlation test   N=0        
RADIATIONS RADIATION REGIMENINDICATION t test N=17 yes N=12 no N=5
NUMBERPACKYEARSSMOKED Spearman correlation test   N=0        
TOBACCOSMOKINGHISTORYINDICATOR ANOVA test N=6        
YEAROFTOBACCOSMOKINGONSET Spearman correlation test   N=0        
COMPLETENESS OF RESECTION ANOVA test N=108        
Clinical variable #1: 'Time to Death'

18 genes related to 'Time to Death'.

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

Time to Death Duration (Months) 0-224 (median=7.3)
  censored N = 193
  death N = 60
     
  Significant markers N = 18
  associated with shorter survival 7
  associated with longer survival 11
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
FUT1 15001 5.881e-09 1e-04 0.675
LARP1 0 4.122e-08 0.00072 0.293
LOC650368 0 4.14e-08 0.00072 0.345
RRAGD 261 1.296e-07 0.0022 0.643
RAD52 0 7.091e-07 0.012 0.334
DPH2 180001 8.067e-07 0.014 0.573
TM4SF19 0.03 1.006e-06 0.017 0.365
ZFAND2A 0.02 1.009e-06 0.017 0.331
ZNF117 0 1.299e-06 0.023 0.394
APOBEC4 0 1.452e-06 0.025 0.387

Figure S1.  Get High-res Image As an example, this figure shows the association of FUT1 to 'Time to Death'. four curves present the cumulative survival rates of 4 quartile subsets of patients. P value = 5.88e-09 with univariate Cox regression analysis using continuous log-2 expression values.

Clinical variable #2: 'AGE'

One gene related to 'AGE'.

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

AGE Mean (SD) 65.12 (9.8)
  Significant markers N = 1
  pos. correlated 1
  neg. correlated 0
List of one gene significantly correlated to 'AGE' by Spearman correlation test

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

SpearmanCorr corrP Q
KIF15 0.3288 6.448e-08 0.00112

Figure S2.  Get High-res Image As an example, this figure shows the association of KIF15 to 'AGE'. P value = 6.45e-08 with Spearman correlation analysis. The straight line presents the best linear regression.

Clinical variable #3: 'GENDER'

84 genes related to 'GENDER'.

Table S5.  Basic characteristics of clinical feature: 'GENDER'

GENDER Labels N
  FEMALE 154
  MALE 132
     
  Significant markers N = 84
  Higher in MALE 7
  Higher in FEMALE 77
List of top 10 genes differentially expressed by 'GENDER'

Table S6.  Get Full Table List of top 10 genes differentially expressed by 'GENDER'

T(pos if higher in 'MALE') ttestP Q AUC
KIF4B -12.54 1.069e-28 1.86e-24 0.8564
EIF4A1 -9.67 5.745e-19 9.97e-15 0.808
RNASEH2C 8.65 4.22e-16 7.32e-12 0.7562
FRG1B -7.29 3.786e-12 6.57e-08 0.7556
GPN1 -7.19 7.547e-12 1.31e-07 0.7198
MACC1 6.93 2.802e-11 4.86e-07 0.7258
SPESP1 -6.72 1.45e-10 2.52e-06 0.7077
COX7C -6.54 3.065e-10 5.32e-06 0.7093
ATP5J 6.44 1.386e-09 2.41e-05 0.7402
GABPA 6.44 1.386e-09 2.41e-05 0.7402

Figure S3.  Get High-res Image As an example, this figure shows the association of KIF4B to 'GENDER'. P value = 1.07e-28 with T-test analysis.

Clinical variable #4: 'KARNOFSKY.PERFORMANCE.SCORE'

No gene related to 'KARNOFSKY.PERFORMANCE.SCORE'.

Table S7.  Basic characteristics of clinical feature: 'KARNOFSKY.PERFORMANCE.SCORE'

KARNOFSKY.PERFORMANCE.SCORE Mean (SD) 74.55 (32)
  Significant markers N = 0
Clinical variable #5: 'HISTOLOGICAL.TYPE'

99 genes related to 'HISTOLOGICAL.TYPE'.

Table S8.  Basic characteristics of clinical feature: 'HISTOLOGICAL.TYPE'

HISTOLOGICAL.TYPE Labels N
  LUNG ACINAR ADENOCARCINOMA 10
  LUNG ADENOCARCINOMA MIXED SUBTYPE 60
  LUNG ADENOCARCINOMA- NOT OTHERWISE SPECIFIED (NOS) 164
  LUNG BRONCHIOLOALVEOLAR CARCINOMA MUCINOUS 4
  LUNG BRONCHIOLOALVEOLAR CARCINOMA NONMUCINOUS 17
  LUNG CLEAR CELL ADENOCARCINOMA 1
  LUNG MICROPAPILLARY ADENOCARCINOMA 2
  LUNG MUCINOUS ADENOCARCINOMA 2
  LUNG PAPILLARY ADENOCARCINOMA 16
  LUNG SOLID PATTERN PREDOMINANT ADENOCARCINOMA 3
  MUCINOUS (COLLOID) ADENOCARCINOMA 4
  MUCINOUS (COLLOID) CARCINOMA 3
     
  Significant markers N = 99
List of top 10 genes differentially expressed by 'HISTOLOGICAL.TYPE'

Table S9.  Get Full Table List of top 10 genes differentially expressed by 'HISTOLOGICAL.TYPE'

ANOVA_P Q
MURC 9.283e-30 1.61e-25
GNG10 1.613e-28 2.8e-24
KRT39 7.989e-24 1.39e-19
KAT5 3.632e-22 6.3e-18
TTC32 1.537e-21 2.67e-17
MED17 3.649e-20 6.33e-16
USMG5 2.958e-16 5.13e-12
GOPC 4.01e-15 6.96e-11
C8ORF42 1.47e-14 2.55e-10
CRISPLD2 4.688e-14 8.13e-10

Figure S4.  Get High-res Image As an example, this figure shows the association of MURC to 'HISTOLOGICAL.TYPE'. P value = 9.28e-30 with ANOVA analysis.

Clinical variable #6: 'PATHOLOGY.T'

No gene related to 'PATHOLOGY.T'.

Table S10.  Basic characteristics of clinical feature: 'PATHOLOGY.T'

PATHOLOGY.T Mean (SD) 1.91 (0.77)
  N
  T1 62
  T2 126
  T3 17
  T4 13
     
  Significant markers N = 0
Clinical variable #7: 'PATHOLOGY.N'

One gene related to 'PATHOLOGY.N'.

Table S11.  Basic characteristics of clinical feature: 'PATHOLOGY.N'

PATHOLOGY.N Mean (SD) 0.59 (0.79)
  N
  N0 130
  N1 44
  N2 41
     
  Significant markers N = 1
  pos. correlated 1
  neg. correlated 0
List of one gene significantly correlated to 'PATHOLOGY.N' by Spearman correlation test

Table S12.  Get Full Table List of one gene significantly correlated to 'PATHOLOGY.N' by Spearman correlation test

SpearmanCorr corrP Q
KLHDC9 0.3169 2.116e-06 0.0367

Figure S5.  Get High-res Image As an example, this figure shows the association of KLHDC9 to 'PATHOLOGY.N'. P value = 2.12e-06 with Spearman correlation analysis.

Clinical variable #8: 'PATHOLOGICSPREAD(M)'

5 genes related to 'PATHOLOGICSPREAD(M)'.

Table S13.  Basic characteristics of clinical feature: 'PATHOLOGICSPREAD(M)'

PATHOLOGICSPREAD(M) Labels N
  M0 139
  M1 10
  MX 64
     
  Significant markers N = 5
List of 5 genes differentially expressed by 'PATHOLOGICSPREAD(M)'

Table S14.  Get Full Table List of 5 genes differentially expressed by 'PATHOLOGICSPREAD(M)'

ANOVA_P Q
C2ORF73 8.589e-08 0.00149
CARD6 6.235e-07 0.0108
POT1 7.72e-07 0.0134
ZSCAN20 1.214e-06 0.0211
FAM190A 1.416e-06 0.0246

Figure S6.  Get High-res Image As an example, this figure shows the association of C2ORF73 to 'PATHOLOGICSPREAD(M)'. P value = 8.59e-08 with ANOVA analysis.

Clinical variable #9: 'TUMOR.STAGE'

No gene related to 'TUMOR.STAGE'.

Table S15.  Basic characteristics of clinical feature: 'TUMOR.STAGE'

TUMOR.STAGE Mean (SD) 1.79 (0.94)
  N
  Stage 1 111
  Stage 2 47
  Stage 3 46
  Stage 4 10
     
  Significant markers N = 0
Clinical variable #10: 'RADIATIONS.RADIATION.REGIMENINDICATION'

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

Table S16.  Basic characteristics of clinical feature: 'RADIATIONS.RADIATION.REGIMENINDICATION'

RADIATIONS.RADIATION.REGIMENINDICATION Labels N
  NO 15
  YES 271
     
  Significant markers N = 17
  Higher in YES 12
  Higher in NO 5
List of top 10 genes differentially expressed by 'RADIATIONS.RADIATION.REGIMENINDICATION'

Table S17.  Get Full Table List of top 10 genes differentially expressed by 'RADIATIONS.RADIATION.REGIMENINDICATION'

T(pos if higher in 'YES') ttestP Q AUC
PRELP -6.15 2.601e-09 4.51e-05 0.6524
CCNI 6.58 4.936e-09 8.57e-05 0.7306
ZNF642 5.97 7.993e-09 0.000139 0.6716
MIIP -6.28 9.514e-09 0.000165 0.5697
MAT2B 6.07 1.175e-08 0.000204 0.5909
SLCO4C1 5.77 2.088e-08 0.000362 0.7262
C12ORF62 6.94 3.603e-08 0.000625 0.7759
LMX1B 5.73 5.774e-08 0.001 0.7087
ZNF506 5.68 1.815e-07 0.00315 0.5496
GALNT14 5.33 1.994e-07 0.00346 0.5748

Figure S7.  Get High-res Image As an example, this figure shows the association of PRELP to 'RADIATIONS.RADIATION.REGIMENINDICATION'. P value = 2.6e-09 with T-test analysis.

Clinical variable #11: 'NUMBERPACKYEARSSMOKED'

No gene related to 'NUMBERPACKYEARSSMOKED'.

Table S18.  Basic characteristics of clinical feature: 'NUMBERPACKYEARSSMOKED'

NUMBERPACKYEARSSMOKED Mean (SD) 39.69 (27)
  Significant markers N = 0
Clinical variable #12: 'TOBACCOSMOKINGHISTORYINDICATOR'

6 genes related to 'TOBACCOSMOKINGHISTORYINDICATOR'.

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

TOBACCOSMOKINGHISTORYINDICATOR Labels N
  CURRENT REFORMED SMOKER FOR < OR = 15 YEARS 77
  CURRENT REFORMED SMOKER FOR > 15 YEARS 59
  CURRENT SMOKER 45
  LIFELONG NON-SMOKER 26
     
  Significant markers N = 6
List of 6 genes differentially expressed by 'TOBACCOSMOKINGHISTORYINDICATOR'

Table S20.  Get Full Table List of 6 genes differentially expressed by 'TOBACCOSMOKINGHISTORYINDICATOR'

ANOVA_P Q
SRM 8.696e-08 0.00151
ZC3HAV1L 6.823e-07 0.0118
GIT1 9.344e-07 0.0162
CMTM5 1.166e-06 0.0202
C14ORF115 1.915e-06 0.0332
FAM92B 2.546e-06 0.0442

Figure S8.  Get High-res Image As an example, this figure shows the association of SRM to 'TOBACCOSMOKINGHISTORYINDICATOR'. P value = 8.7e-08 with ANOVA analysis.

Clinical variable #13: 'YEAROFTOBACCOSMOKINGONSET'

No gene related to 'YEAROFTOBACCOSMOKINGONSET'.

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

YEAROFTOBACCOSMOKINGONSET Mean (SD) 1965.24 (13)
  Significant markers N = 0
Clinical variable #14: 'COMPLETENESS.OF.RESECTION'

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

Table S22.  Basic characteristics of clinical feature: 'COMPLETENESS.OF.RESECTION'

COMPLETENESS.OF.RESECTION Labels N
  R0 146
  R1 7
  R2 1
  RX 10
     
  Significant markers N = 108
List of top 10 genes differentially expressed by 'COMPLETENESS.OF.RESECTION'

Table S23.  Get Full Table List of top 10 genes differentially expressed by 'COMPLETENESS.OF.RESECTION'

ANOVA_P Q
ANKIB1 4.024e-64 6.99e-60
CDK13 1.983e-51 3.44e-47
FAM177A1 5.311e-47 9.22e-43
SLC6A16 5.327e-46 9.25e-42
CWC22 2.778e-45 4.82e-41
ZBTB1 3.104e-39 5.39e-35
ZBTB25 3.104e-39 5.39e-35
ZNF619 4.874e-37 8.46e-33
THRAP3 2.132e-25 3.7e-21
MTMR4 4.649e-25 8.07e-21

Figure S9.  Get High-res Image As an example, this figure shows the association of ANKIB1 to 'COMPLETENESS.OF.RESECTION'. P value = 4.02e-64 with ANOVA analysis.

Methods & Data
Input

  • Expresson data file = LUAD-TP.meth.for_correlation.filtered_data.txt

  • Clinical data file = LUAD-TP.clin.merged.picked.txt

  • Number of patients = 286

  • Number of genes = 17358

  • Number of clinical features = 14

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

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

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] Lehmann and Romano, Testing Statistical Hypotheses (3E ed.), New York: Springer. ISBN 0387988645 (2005)
[4] Howell, D, Statistical Methods for Psychology. (5th ed.), Duxbury Press:324-5 (2002)
[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 © 2013 Broad Institute TCGA GDAC as part of the TCGA Research Network. All rights reserved.
Made with Nozzle