R re-analysis in TCGA

1
2

a= read.table('plot.txt',sep = '\t', header = T)
colnames(a) = c('id', 'subtype', 'expression', 'mutations', 'CNA')

1
2
3
4
5

dat = a
library(ggstatsplot)
library(ggplot2)
ggbetweenstats(dat, x = subtype, y = expression)
ggsave('subtypeExp.png')

1
2
3
4

library(ggplot2)
ggplot(dat, aes(x=subtype, y=expression)) +
  geom_boxplot()
ggsave('boxplot.png')

1
2
3
4

ggplot(dat, aes(x=subtype, y=expression)) +
  geom_violin(trim=FALSE, fill='brown', color='red') +
geom_boxplot(width=0.1) + theme_minimal()
ggsave('violinplot.png')

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31

# variance
aov(expression ~ subtype, data=dat)

summary(aov(expression ~ subtype, data=dat))
# output
#              Df Sum Sq Mean Sq F value Pr(>F)
# subtype       4  289.5   72.38   115.5 <2e-16 ***
# Residuals   514  322.0    0.63
# ---
# Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1


TukeyHSD(aov(expression ~ subtype, data=dat))
# output
# Tukey multiple comparisons of means
#     95% family-wise confidence level
#
# Fit: aov(formula = expression ~ subtype, data = dat)
#
# $subtype
#                                  diff        lwr         upr     p adj
# HER2-enriched-Basal-like  -1.58328100 -1.9422245 -1.22433751 0.0000000
# Luminal A-Basal-like      -1.82596364 -2.0873307 -1.56459662 0.0000000
# Luminal B-Basal-like      -2.09467315 -2.3870044 -1.80234192 0.0000000
# Normal-like-Basal-like    -1.61344005 -2.4101234 -0.81675673 0.0000005
# Luminal A-HER2-enriched   -0.24268264 -0.5610358  0.07567051 0.2273188
# Luminal B-HER2-enriched   -0.51139215 -0.8556211 -0.16716323 0.0005268
# Normal-like-HER2-enriched -0.03015905 -0.8473128  0.78699474 0.9999764
# Luminal B-Luminal A       -0.26870951 -0.5094705 -0.02794855 0.0199389
# Normal-like-Luminal A      0.21252359 -0.5667149  0.99176206 0.9453257
# Normal-like-Luminal B      0.48123310 -0.3089298  1.27139601 0.4552733