-
Notifications
You must be signed in to change notification settings - Fork 42
/
TNMplot_miRNA.Rmd
206 lines (173 loc) · 7.39 KB
/
TNMplot_miRNA.Rmd
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
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
---
title: "Breast Cancer, miRNA expression in Tumor, Normal, and Metastatic samples, in PAM50 subtypes"
author: "Mikhail Dozmorov"
date: "`r Sys.Date()`"
output:
pdf_document:
toc: no
html_document:
toc: no
theme: united
bibliography: data.TCGA/TCGA.bib
csl: styles.ref/genomebiology.csl
editor_options:
chunk_output_type: console
---
```{r setup, echo=FALSE, message=FALSE, warning=FALSE}
# Set up the environment
library(knitr)
opts_chunk$set(cache.path='cache/', fig.path='img/', cache=F, tidy=T, fig.keep='high', echo=F, dpi=100, warnings=F, message=F, comment=NA, warning=F, results='as.is', fig.width = 10, fig.height = 6) #out.width=700,
library(pander)
panderOptions('table.split.table', Inf)
set.seed(1)
library(dplyr)
options(stringsAsFactors = FALSE)
```
```{r}
library(curatedTCGAData)
library(TCGAutils)
library(ggplot2)
library("ggsci")
library(scales)
# scales::show_col(pal_lancet("lanonc")(8))
mycols = pal_lancet("lanonc")(8)
library(grid)
library(gridExtra)
library(ggprism)
library(readr)
library(stringr)
library(writexl)
```
```{r echo=TRUE}
selected_genes <- c("hsa-let-7a-1")
```
```{r}
dir_data <- "/Users/mdozmorov/Documents/Work/GitHub/TCGAsurvival"
# Create results directory if not exist
if (!dir.exists(file.path(dir_data, "results"))) dir.create(file.path(dir_data, "results"))
# Save the data
fileNameOut1 <- file.path(dir_data, "results/TCGA_BRCA_miRNA.xlsx")
```
```{r results='hide'}
brca <- curatedTCGAData(diseaseCode = "BRCA", assays = "miRNASeqGene", FALSE, version = "2.0.1")
# https://gdc.cancer.gov/resources-tcga-users/tcga-code-tables/sample-type-codes
# 01 Primary Solid Tumor
# 11 Solid Tissue Normal
sampleTables(brca)
# 01 06 11
# 755 7 87
```
# Tumor-normal pairs comparison
```{r message=FALSE, results='hide'}
(tnmae <- TCGAsplitAssays(brca, c("01", "11")))
(matchmae <- as(tnmae, "MatchedAssayExperiment"))
selected_expr_tumor <- assay(matchmae[[1]])[selected_genes, ]
selected_expr_normal <- assay(matchmae[[2]])[selected_genes, ]
# T-test to put on a plot
res <- t.test(selected_expr_normal, selected_expr_tumor, paired = TRUE)$p.value
grob <- grobTree(textGrob(paste0("p-value: ", formatC(res, format = "e")), x=0.1, y=0.90, hjust=0, gp=gpar(col="black", fontsize=10, fontface="italic")))
mtx_to_plot <- data.frame(Status = c(rep("Tumor", length(selected_expr_tumor)), rep("Normal", length(selected_expr_normal))), Expression = c(selected_expr_tumor, selected_expr_normal))
# ggplot(mtx_to_plot, aes(x = Status, y = Expression, fill = Status)) +
# geom_boxplot() +
# theme_bw() +
# scale_fill_manual(values = mycols[c(3, 2, 1)]) +
# geom_jitter(shape=20, position=position_jitter(0.2), size = 1, alpha = 0.2) +
# annotation_custom(grob)
```
```{r fig.height=4, fig.width=4}
p1 <- ggplot(mtx_to_plot, aes(x = Status, y = Expression, fill = Status)) +
geom_boxplot() +
scale_fill_prism(palette = "prism_light") +
geom_jitter(shape=20, position=position_jitter(0.2), size = 1, alpha = 0.2) +
annotation_custom(grob) +
theme_prism(palette = "prism_light")
# ggsave("results/Figure_TN_paired.png", plot = p1, width = 4, height = 4, dpi = 300)
```
# Tumor-normal-metastatic comparison
```{r message=FALSE, results='hide'}
(tnmae <- TCGAsplitAssays(brca, c("01", "11", "06")))
selected_expr_tumor <- assay(tnmae[[1]])[selected_genes, ]
selected_expr_met <- assay(tnmae[[2]])[selected_genes, ]
selected_expr_normal <- assay(tnmae[[3]])[selected_genes, ]
mtx_to_plot <- data.frame(Status = c(rep("Tumor", length(selected_expr_tumor)), rep("Normal", length(selected_expr_normal)), rep("Metastatic", length(selected_expr_met))), Expression = c(selected_expr_tumor, selected_expr_normal, selected_expr_met))
mtx_to_plot$Status <- factor(mtx_to_plot$Status, levels = c("Normal", "Tumor", "Metastatic"))
res_tn <- t.test(selected_expr_tumor, selected_expr_normal)$p.value %>% formatC(., format = "e")
res_tm <- t.test(selected_expr_tumor, selected_expr_met)$p.value %>% formatC(., format = "e")
res_mn <- t.test(selected_expr_met, selected_expr_normal)$p.value %>% formatC(., format = "e")
grob <- textGrob(paste0("T vs. N: ", res_tn, ", T vs. M: ", res_tm, ", M vs. N: ", res_mn)) #, gp=gpar(col="black", fontsize=8, fontface="italic")))
my_text <- paste0("T vs. N: ", res_tn, ", T vs. M: ", res_tm, ", M vs. N: ", res_mn)
my_grob <- grid.text(my_text, x=0.5, y=0.9, gp=gpar(col="black", fontsize=6, fontface="italic"))
# ggplot(mtx_to_plot, aes(x = Status, y = Expression, fill = Status)) +
# geom_boxplot() +
# theme_bw() +
# scale_fill_manual(values = mycols[c(3, 2, 1)]) +
# geom_jitter(shape=20, position=position_jitter(0.2), size = 1, alpha = 0.2) +
# annotation_custom(my_grob)
```
```{r fig.height=4, fig.width=5}
p2 <- ggplot(mtx_to_plot, aes(x = Status, y = Expression, fill = Status)) +
geom_boxplot() +
# scale_fill_prism(palette = "prism_light") +
scale_fill_manual(values = c(prism_color_pal(palette = "prism_light")(10)[6], prism_color_pal(palette = "prism_light")(10)[7], prism_color_pal(palette = "prism_light")(10)[9])) +
geom_jitter(shape=20, position=position_jitter(0.2), size = 1, alpha = 0.2) +
annotation_custom(my_grob) +
theme_prism(palette = "prism_light")
# ggsave("results/Figure_TNM.png", plot = p2, width = 5, height = 4, dpi = 300)
```
```{r fig.height=4}
grid.arrange(p1, p2, ncol = 2)
```
# PAM50
```{r fig.height=4, fig.width=5}
# PAM50 annotations
pam50 <- read_tsv("data.TCGA/PAM50_classification.txt", col_types = c("cc"))
# Get tumor only
tnmae_tumor <- TCGAsplitAssays(brca, c("01"))
# Get selected gene
selected_expr_tumor <- assay(tnmae[[1]])[selected_genes, ]
# Matrix for plotting, with Sample ID shortened to match PAM50
mtx_to_plot <- data.frame(Sample = str_sub(names(selected_expr_tumor), start = 1, end = 16),
Expression = selected_expr_tumor)
mtx_to_plot <- left_join(mtx_to_plot, pam50, by = c("Sample" = "Particpant ID"))
mtx_to_plot <- mtx_to_plot[!is.na(mtx_to_plot$Subtype), ]
# table(mtx_to_plot$Subtype)
# Basal Her2 LumA LumB
# 71 31 233 100
p3 <- ggplot(mtx_to_plot, aes(x = Subtype, y = log2(Expression), fill = Subtype)) +
geom_boxplot() +
# scale_fill_prism(palette = "prism_light") +
scale_fill_manual(values = c(prism_color_pal(palette = "prism_light")(10)[6:9])) +
geom_jitter(shape=20, position=position_jitter(0.2), size = 1, alpha = 0.2) +
# annotation_custom(my_grob) +
theme_prism(palette = "prism_light")
p3
```
# Save the data
- Original data description: "Gene-level log2 RPM miRNA expression values"
- The data seems NOT to be log2-transformed. I log2-transformed it manually
- The data is saved in `r basename(fileNameOut1)`
```{r}
# Tumor-specific expression
mtx <- log2(assay(tnmae_tumor) + 1)
# Transpose and make a data frame
mtx_to_save <- t(mtx)
mtx_to_save <- data.frame(Sample = rownames(mtx_to_save),
`Participant ID` = str_sub(rownames(mtx_to_save), start = 1, end = 16),
mtx_to_save)
# Attach PAM50 and sort
mtx_to_save <- left_join(mtx_to_save, pam50, by = c("Participant.ID" = "Particpant ID"))
mtx_to_save <- mtx_to_save %>% relocate("Subtype", .after = "Participant.ID") %>% dplyr::arrange(Subtype)
write_xlsx(mtx_to_save, fileNameOut1)
```
```{r eval=FALSE}
library(scales)
show_palette <- function(palette) {
scales::show_col(
prism_colour_pal(palette = palette)(
attr(prism_colour_pal(palette = palette), "max_n")
)
)
}
# show the colours in the palette "pearl"
show_palette("prism_light")
```