Correlation between mRNAseq expression and clinical features
Overview
Introduction

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

Summary

Testing the association between 18086 genes and 12 clinical features across 258 samples, statistically thresholded by Q value < 0.05, 7 clinical features related to at least one genes.

  • 299 genes correlated to 'Time from Specimen Diagnosis to Death'.

    • CD38|952 ,  EAF2|55840 ,  SAMSN1|64092 ,  ANKRD22|118932 ,  CXCL10|3627 ,  ...

  • 7 genes correlated to 'AGE'.

    • ACOX2|8309 ,  MAOB|4129 ,  FAM84B|157638 ,  PPP1R1B|84152 ,  CMBL|134147 ,  ...

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

    • TMEM156|80008 ,  POU2AF1|5450 ,  IGJ|3512 ,  CD38|952 ,  FCRL5|83416 ,  ...

  • 1 gene correlated to 'PATHOLOGY.M.STAGE'.

    • LRRC28|123355

  • 1 gene correlated to 'MELANOMA.ULCERATION'.

    • CDKL2|8999

  • 8 genes correlated to 'BRESLOW.THICKNESS'.

    • GBP4|115361 ,  REEP6|92840 ,  TNFSF13B|10673 ,  SLC7A8|23428 ,  GCNT1|2650 ,  ...

  • 25 genes correlated to 'GENDER'.

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

  • No genes correlated to 'Time to Death', 'NEOPLASM.DISEASESTAGE', 'PATHOLOGY.T.STAGE', 'PATHOLOGY.N.STAGE', and 'MELANOMA.PRIMARY.KNOWN'.

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 from Specimen Diagnosis to Death Cox regression test N=299 shorter survival N=13 longer survival N=286
Time to Death Cox regression test   N=0        
AGE Spearman correlation test N=7 older N=0 younger N=7
PRIMARY SITE OF DISEASE ANOVA test N=141        
NEOPLASM DISEASESTAGE ANOVA test   N=0        
PATHOLOGY T STAGE Spearman correlation test   N=0        
PATHOLOGY N STAGE Spearman correlation test   N=0        
PATHOLOGY M STAGE ANOVA test N=1        
MELANOMA ULCERATION t test N=1 yes N=1 no N=0
MELANOMA PRIMARY KNOWN t test   N=0        
BRESLOW THICKNESS Spearman correlation test N=8 higher breslow.thickness N=4 lower breslow.thickness N=4
GENDER t test N=25 male N=14 female N=11
Clinical variable #1: 'Time from Specimen Diagnosis to Death'

299 genes related to 'Time from Specimen Diagnosis to Death'.

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

Time from Specimen Diagnosis to Death Duration (Months) 0.1-124.3 (median=13.9)
  censored N = 126
  death N = 120
     
  Significant markers N = 299
  associated with shorter survival 13
  associated with longer survival 286
List of top 10 genes significantly associated with 'Time from Specimen Diagnosis to Death' by Cox regression test

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

HazardRatio Wald_P Q C_index
CD38|952 0.8 8.296e-11 1.5e-06 0.318
EAF2|55840 0.65 1.876e-10 3.4e-06 0.317
SAMSN1|64092 0.77 3.927e-10 7.1e-06 0.322
ANKRD22|118932 0.8 6.361e-10 1.2e-05 0.329
CXCL10|3627 0.8 7.827e-10 1.4e-05 0.317
CCL8|6355 0.77 8.584e-10 1.6e-05 0.327
SRGN|5552 0.65 8.785e-10 1.6e-05 0.332
GCH1|2643 0.67 9.341e-10 1.7e-05 0.319
GBP2|2634 0.72 1.025e-09 1.9e-05 0.328
C21ORF7|56911 0.66 1.127e-09 2e-05 0.334

Figure S1.  Get High-res Image As an example, this figure shows the association of CD38|952 to 'Time from Specimen Diagnosis to Death'. four curves present the cumulative survival rates of 4 quartile subsets of patients. P value = 8.3e-11 with univariate Cox regression analysis using continuous log-2 expression values.

Clinical variable #2: 'Time to Death'

No gene related to 'Time to Death'.

Table S3.  Basic characteristics of clinical feature: 'Time to Death'

Time to Death Duration (Months) 0.2-357.4 (median=48.7)
  censored N = 130
  death N = 122
     
  Significant markers N = 0
Clinical variable #3: 'AGE'

7 genes related to 'AGE'.

Table S4.  Basic characteristics of clinical feature: 'AGE'

AGE Mean (SD) 55.53 (16)
  Significant markers N = 7
  pos. correlated 0
  neg. correlated 7
List of 7 genes significantly correlated to 'AGE' by Spearman correlation test

Table S5.  Get Full Table List of 7 genes significantly correlated to 'AGE' by Spearman correlation test

SpearmanCorr corrP Q
ACOX2|8309 -0.3509 8.993e-09 0.000163
MAOB|4129 -0.3049 7.282e-07 0.0132
FAM84B|157638 -0.2982 1.302e-06 0.0235
PPP1R1B|84152 -0.3103 1.506e-06 0.0272
CMBL|134147 -0.2943 1.816e-06 0.0328
NMNAT3|349565 -0.2922 2.16e-06 0.0391
MGST2|4258 -0.2902 2.558e-06 0.0462

Figure S2.  Get High-res Image As an example, this figure shows the association of ACOX2|8309 to 'AGE'. P value = 8.99e-09 with Spearman correlation analysis. The straight line presents the best linear regression.

Clinical variable #4: 'PRIMARY.SITE.OF.DISEASE'

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

Table S6.  Basic characteristics of clinical feature: 'PRIMARY.SITE.OF.DISEASE'

PRIMARY.SITE.OF.DISEASE Labels N
  DISTANT METASTASIS 33
  PRIMARY TUMOR 1
  REGIONAL CUTANEOUS OR SUBCUTANEOUS TISSUE 55
  REGIONAL LYMPH NODE 168
     
  Significant markers N = 141
List of top 10 genes differentially expressed by 'PRIMARY.SITE.OF.DISEASE'

Table S7.  Get Full Table List of top 10 genes differentially expressed by 'PRIMARY.SITE.OF.DISEASE'

ANOVA_P Q
TMEM156|80008 1.749e-11 3.16e-07
POU2AF1|5450 8.76e-11 1.58e-06
IGJ|3512 2.725e-10 4.93e-06
CD38|952 3.602e-10 6.51e-06
FCRL5|83416 4e-10 7.23e-06
EXOC3|11336 9.001e-10 1.63e-05
RBP5|83758 2.16e-09 3.91e-05
MGC29506|51237 3.651e-09 6.6e-05
ADAM6|8755 4.495e-09 8.13e-05
KIAA0748|9840 4.88e-09 8.82e-05

Figure S3.  Get High-res Image As an example, this figure shows the association of TMEM156|80008 to 'PRIMARY.SITE.OF.DISEASE'. P value = 1.75e-11 with ANOVA analysis.

Clinical variable #5: 'NEOPLASM.DISEASESTAGE'

No gene related to 'NEOPLASM.DISEASESTAGE'.

Table S8.  Basic characteristics of clinical feature: 'NEOPLASM.DISEASESTAGE'

NEOPLASM.DISEASESTAGE Labels N
  I OR II NOS 10
  STAGE 0 5
  STAGE I 21
  STAGE IA 10
  STAGE IB 24
  STAGE II 16
  STAGE IIA 10
  STAGE IIB 13
  STAGE IIC 10
  STAGE III 30
  STAGE IIIA 11
  STAGE IIIB 22
  STAGE IIIC 41
  STAGE IV 12
     
  Significant markers N = 0
Clinical variable #6: 'PATHOLOGY.T.STAGE'

No gene related to 'PATHOLOGY.T.STAGE'.

Table S9.  Basic characteristics of clinical feature: 'PATHOLOGY.T.STAGE'

PATHOLOGY.T.STAGE Mean (SD) 2.4 (1.3)
  N
  0 21
  1 30
  2 57
  3 49
  4 53
     
  Significant markers N = 0
Clinical variable #7: 'PATHOLOGY.N.STAGE'

No gene related to 'PATHOLOGY.N.STAGE'.

Table S10.  Basic characteristics of clinical feature: 'PATHOLOGY.N.STAGE'

PATHOLOGY.N.STAGE Mean (SD) 0.83 (1.1)
  N
  0 134
  1 47
  2 31
  3 31
     
  Significant markers N = 0
Clinical variable #8: 'PATHOLOGY.M.STAGE'

One gene related to 'PATHOLOGY.M.STAGE'.

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

PATHOLOGY.M.STAGE Labels N
  M0 232
  M1 4
  M1A 2
  M1B 2
  M1C 5
     
  Significant markers N = 1
List of one gene differentially expressed by 'PATHOLOGY.M.STAGE'

Table S12.  Get Full Table List of one gene differentially expressed by 'PATHOLOGY.M.STAGE'

ANOVA_P Q
LRRC28|123355 2.605e-07 0.00471

Figure S4.  Get High-res Image As an example, this figure shows the association of LRRC28|123355 to 'PATHOLOGY.M.STAGE'. P value = 2.6e-07 with ANOVA analysis.

Clinical variable #9: 'MELANOMA.ULCERATION'

One gene related to 'MELANOMA.ULCERATION'.

Table S13.  Basic characteristics of clinical feature: 'MELANOMA.ULCERATION'

MELANOMA.ULCERATION Labels N
  NO 98
  YES 64
     
  Significant markers N = 1
  Higher in YES 1
  Higher in NO 0
List of one gene differentially expressed by 'MELANOMA.ULCERATION'

Table S14.  Get Full Table List of one gene differentially expressed by 'MELANOMA.ULCERATION'

T(pos if higher in 'YES') ttestP Q AUC
CDKL2|8999 4.89 2.746e-06 0.0497 0.7148

Figure S5.  Get High-res Image As an example, this figure shows the association of CDKL2|8999 to 'MELANOMA.ULCERATION'. P value = 2.75e-06 with T-test analysis.

Clinical variable #10: 'MELANOMA.PRIMARY.KNOWN'

No gene related to 'MELANOMA.PRIMARY.KNOWN'.

Table S15.  Basic characteristics of clinical feature: 'MELANOMA.PRIMARY.KNOWN'

MELANOMA.PRIMARY.KNOWN Labels N
  NO 31
  YES 227
     
  Significant markers N = 0
Clinical variable #11: 'BRESLOW.THICKNESS'

8 genes related to 'BRESLOW.THICKNESS'.

Table S16.  Basic characteristics of clinical feature: 'BRESLOW.THICKNESS'

BRESLOW.THICKNESS Mean (SD) 3.58 (5.1)
  Significant markers N = 8
  pos. correlated 4
  neg. correlated 4
List of 8 genes significantly correlated to 'BRESLOW.THICKNESS' by Spearman correlation test

Table S17.  Get Full Table List of 8 genes significantly correlated to 'BRESLOW.THICKNESS' by Spearman correlation test

SpearmanCorr corrP Q
GBP4|115361 -0.3627 2.349e-07 0.00425
REEP6|92840 0.3578 3.493e-07 0.00632
TNFSF13B|10673 -0.3411 1.289e-06 0.0233
SLC7A8|23428 0.3364 1.832e-06 0.0331
GCNT1|2650 -0.3357 1.928e-06 0.0349
FGL2|10875 -0.3348 2.064e-06 0.0373
C6ORF218|221718 0.3341 2.317e-06 0.0419
CRTAP|10491 0.3328 2.393e-06 0.0433

Figure S6.  Get High-res Image As an example, this figure shows the association of GBP4|115361 to 'BRESLOW.THICKNESS'. P value = 2.35e-07 with Spearman correlation analysis. The straight line presents the best linear regression.

Clinical variable #12: 'GENDER'

25 genes related to 'GENDER'.

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

GENDER Labels N
  FEMALE 100
  MALE 158
     
  Significant markers N = 25
  Higher in MALE 14
  Higher in FEMALE 11
List of top 10 genes differentially expressed by 'GENDER'

Table S19.  Get Full Table List of top 10 genes differentially expressed by 'GENDER'

T(pos if higher in 'MALE') ttestP Q AUC
ZFY|7544 41.76 5.567e-107 1.01e-102 0.9989
PRKY|5616 34.5 6.015e-87 1.09e-82 0.9976
DDX3Y|8653 40.58 7.55e-64 1.36e-59 1
XIST|7503 -24.33 1.131e-63 2.04e-59 0.963
RPS4Y1|6192 35.29 1.186e-61 2.14e-57 1
EIF1AY|9086 41.82 1.888e-55 3.41e-51 1
KDM5D|8284 38.5 2.424e-53 4.38e-49 1
UTY|7404 33.98 2.284e-44 4.13e-40 1
USP9Y|8287 31.8 1.025e-39 1.85e-35 0.9996
TSIX|9383 -16.8 1.8e-39 3.25e-35 0.9563

Figure S7.  Get High-res Image As an example, this figure shows the association of ZFY|7544 to 'GENDER'. P value = 5.57e-107 with T-test analysis.

Methods & Data
Input

  • Expresson data file = SKCM-TM.uncv2.mRNAseq_RSEM_normalized_log2.txt

  • Clinical data file = SKCM-TM.merged_data.txt

  • Number of patients = 258

  • Number of genes = 18086

  • 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.

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)
Made with Nozzle