-
Notifications
You must be signed in to change notification settings - Fork 2
/
DE.Rmd
372 lines (285 loc) · 12.6 KB
/
DE.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
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
---
title: "Differential expression"
output:
html_document:
code_folding: hide
---
This Rmarkdown document presents results from a bulk RNA-seq experiment that has been processed using mapping (e.g. hisat2) and then features are counted using featurecounts. For data that has been processed using pseudoalignment please refer to DE_pseudoalignment.
To run the Rmarkdown please save your data as
```{r, include=FALSE}
source('functions.R')
library(org.Hs.eg.db)
library(org.Mm.eg.db)
library("org.Mmu.eg.db")
library(DESeq2)
library(pheatmap)
library(tidyverse)
library(DT)
library(limma)
library(knitr)
library(kableExtra)
library("ggplot2") #Best plots
library("ggrepel") #Avoid overlapping labels
library(yaml)
library(rhdf5)
library(tximport)
library(ensembldb)
library(EnsDb.Hsapiens.v86)
library(EnsDb.Mmusculus.v79)
library(biomaRt)
knitr::opts_chunk$set(cache=TRUE, warning=FALSE, message=FALSE)
```
```{r yaml, echo=TRUE,warning=FALSE,message=FALSE,error=FALSE, include=FALSE}
params <- read_yaml("config.yml")
```
```{r input, echo=TRUE,warning=FALSE,message=FALSE,error=FALSE, include=FALSE}
if(params$kallisto){
print("kallisto input detected")
design_files <- list.files(path='designs_kallisto/', pattern = "design_")
# load in the sample metadata file
sample_table <- read.csv("kallisto_input.csv")
if (dir.exists(params$kallisto_dir)) {
path <- file.path(params$kallisto_dir, sample_table$Sample, "abundance.h5")
sample_table <- dplyr::mutate(sample_table, path = path)
if(params$species == "human"){
# Create the tx2gene file that will map transcripts to genes
Tx <- transcripts(EnsDb.Hsapiens.v86, columns=c("tx_id", "gene_id", "symbol"))
Tx <- as_tibble(Tx)
Tx <- dplyr::rename(Tx, target_id = tx_id, ens_gene = gene_id, ext_gene = symbol)
Tx <- dplyr::select(Tx, "target_id", "ens_gene", "ext_gene")
}else if(params$species == "mouse"){
Tx <- transcripts(EnsDb.Mmusculus.v79, columns=c("tx_id", "gene_id", "symbol"))
Tx <- as_tibble(Tx)
Tx <- dplyr::rename(Tx, target_id = tx_id, ens_gene = gene_id, ext_gene = symbol)
Tx <- dplyr::select(Tx, "target_id", "ens_gene", "ext_gene")
}else if(params$species == "macaque"){
orgSymbols <- keys(org.Mmu.eg.db, keytype="ENSEMBL")
mart <- useMart(dataset = "mmulatta_gene_ensembl", biomart='ensembl')
tx2gene <- getBM(attributes = c('ensembl_gene_id', 'ensembl_gene_id_version', 'ensembl_transcript_id', 'ensembl_transcript_id_version','entrezgene_id'),
filters = 'ensembl_gene_id',
values = orgSymbols,
mart = mart)
Tx <- tx2gene %>% dplyr::select(ensembl_transcript_id, ensembl_gene_id)
colnames(Tx) <- c("TXNAME", "GENEID")
}else if(params$species == "pig"){
ensembl = useMart(biomart="ENSEMBL_MART_ENSEMBL",
dataset="sscrofa_gene_ensembl",
host="uswest.ensembl.org",
ensemblRedirect = FALSE)
orgSymbols <- unlist(getBM(attributes = 'ensembl_gene_id', mart=ensembl))
mart <- useMart(dataset = "sscrofa_gene_ensembl", biomart='ensembl', host="uswest.ensembl.org")
tx2gene <- getBM(attributes = c('ensembl_gene_id', 'ensembl_gene_id_version', 'ensembl_transcript_id', 'ensembl_transcript_id_version','entrezgene_id'),
filters = 'ensembl_gene_id',
values = orgSymbols,
mart = mart)
Tx <- tx2gene %>% dplyr::select(ensembl_transcript_id, ensembl_gene_id)
colnames(Tx) <- c("TXNAME", "GENEID")
}else if(params$species == "rabbit"){
ensembl = useMart(biomart="ENSEMBL_MART_ENSEMBL",
dataset="ocuniculus_gene_ensembl",
host="uswest.ensembl.org",
ensemblRedirect = FALSE)
orgSymbols <- unlist(getBM(attributes = 'ensembl_gene_id', mart=ensembl))
mart <- useMart(dataset = "ocuniculus_gene_ensembl", biomart='ensembl', host="uswest.ensembl.org")
tx2gene <- getBM(attributes = c('ensembl_gene_id', 'ensembl_gene_id_version', 'ensembl_transcript_id', 'ensembl_transcript_id_version','entrezgene_id'),
filters = 'ensembl_gene_id',
values = orgSymbols,
mart = mart)
Tx <- tx2gene %>% dplyr::select(ensembl_transcript_id, ensembl_gene_id)
colnames(Tx) <- c("TXNAME", "GENEID")
}else{ print('please supply a valid species within the config.yml file')}
# import Kallisto transcript counts into R using Tximport
Txi_gene <- tximport(path,
type = "kallisto",
tx2gene = Tx,
txOut = FALSE, # TRUE outputs transcripts, FALSE outputs gene-level data
countsFromAbundance = "lengthScaledTPM",
ignoreTxVersion = TRUE)
# Write the counts to an object (used for metadata and clustering)
df_mRNA <- Txi_gene$counts %>%
round() %>%
data.frame()
colnames(df_mRNA) <- sample_table$Sample
meta_data <- sample_table
rownames(meta_data) <- meta_data$Sample
if (file.exists(paste0('designs_kallisto/',design_files[1]))) {
for (i in design_files){
meta_data <- read.table(paste0('designs_kallisto/',i), sep=",", header = TRUE)
rownames(meta_data) <- meta_data$Sample
df_mRNA_tmp = df_mRNA[,rownames(meta_data)]
all(rownames(meta_data) %in% colnames(df_mRNA_tmp))
assign(paste("meta_data", i, sep = "."), meta_data)}
} else {
print("No design files were detected please add a file called design_<test>_<control>_<test>_<column>.csv. Please refer to documentation on github for more ifnormation")
}
} else {
print("Please add path to the kallisto dir in the config.yml file as it seems to be missing")
}
}else{
design_files <- list.files(path='designs_featurecounts/', pattern = "design_")
if (file.exists("featurecounts.tsv.gz")) {
df_mRNA <- read.table(gzfile("featurecounts.tsv.gz"), sep = "\t", header = TRUE, row.names = 1)
colnames(df_mRNA) <- gsub(".", "-", x = colnames(df_mRNA), fixed = T)
} else {
print("Please add featurecounts.tsv.gz into the project folder as it seems to be missing")
}
if (file.exists(paste0('designs_featurecounts/',design_files[1]))) {
for (i in design_files){
meta_data <- read.table(paste0('designs_featurecounts/',i), sep=",", header = TRUE)
rownames(meta_data) <- meta_data$Sample
df_mRNA_tmp = df_mRNA[,rownames(meta_data)]
all(rownames(meta_data) %in% colnames(df_mRNA_tmp))
assign(paste("meta_data", i, sep = "."), meta_data)
}
} else {
print("No design files were detected please add a file called design_<test>_<control>_<test>_<column>.csv. Please refer to documentation on github for more ifnormation")
}
}
```
```{r}
## Set up the DESeq2 object
for (i in design_files) {
x <- strsplit(i, "_")
stat.test <- x[[1]][2]
control <- x[[1]][3]
test <- x[[1]][4]
value <- x[[1]][5]
value <- gsub(".csv","",value)
meta_data <- get(gsub("SAMPLE_FILE",i , "meta_data.SAMPLE_FILE"))
rownames(meta_data) <- meta_data$Sample
model <- as.character(meta_data$model[[1]])
if ("wald" == stat.test){
deseq_results <- run_deseq2(as.data.frame(df_mRNA), meta_data, control = control, test=test, value=value, model = model)
} else {
reduced_model <- as.character(meta_data$reduced[[1]])
deseq_results <- run_deseq2_lrt(as.data.frame(df_mRNA), meta_data, control = control, test=test, value=value, model = model, reduced = reduced_model)
}
res = deseq_results@res
dds = deseq_results@dds
assign(paste("res", i, sep = "."), res)
assign(paste("dds", i, sep = "."), dds)
}
```
# Model fitting
This section of the report describes figures that can be used to assess how well the DESeq2 model has fitted the data.
## Dispersion {.tabset .tabset-fade}
Plotting the dispersion estimates is a useful diagnostic. The dispersion plot below is typical, with the final estimates shrunk from the gene-wise estimates towards the fitted estimates. Some gene-wise estimates are flagged as outliers and not shrunk towards the fitted value, (this outlier detection is described in the manual page for estimateDispersionsMAP). The amount of shrinkage can be more or less than seen here, depending on the sample size, the number of coefficients, the row mean and the variability of the gene-wise estimates.
```{r, results='asis', echo = FALSE}
for (i in design_files){
name <- gsub(".csv","",i)
dds <- get(gsub("SAMPLE_FILE",i , "dds.SAMPLE_FILE"))
cat("### ",name,"\n")
p <- plotDispEsts(dds)
print(p)
cat('\n\n')
}
```
# Summary of the data {.tabset .tabset-fade}
```{r, results='asis', echo = FALSE}
k <- list()
for (i in design_files){
name <- gsub(".csv","",i)
res <- get(gsub("SAMPLE_FILE",i , "res.SAMPLE_FILE"))
cat("## ",name,"\n","\n")
res <- as.data.frame(res)
cat("Number of significant genes:","\n")
significance <- res %>% dplyr::filter(padj < 0.05) %>% count()
cat(significance[1,],"\n")
cat("Number of upregulated significant genes log2 > 1:","\n")
significance <- res %>% dplyr::filter(padj < 0.05 & log2FoldChange > 1) %>% count()
cat(significance[1,],"\n")
cat("Number of downregulated significant genes log2 < -1:","\n")
significance <- res %>% dplyr::filter(padj < 0.05 & log2FoldChange < -1) %>% count()
cat(significance[1,],"\n")
cat('\n\n')
}
```
# MA plots {.tabset .tabset-fade}
In DESeq2, the function plotMA shows the log2 fold changes attributable to a given variable over the mean of normalized counts for all the samples in the DESeqDataSet. Points will be colored red if the adjusted p value is less than 0.01. Points which fall out of the window are plotted as open triangles pointing either up or down.
```{r, results='asis', echo = FALSE}
for (i in design_files){
name <- gsub(".csv","",i)
res <- get(gsub("SAMPLE_FILE",i , "res.SAMPLE_FILE"))
cat("## ",name,"\n")
plt <- DESeq2::plotMA(res)
print(plt)
cat('\n\n')
}
```
# Volcano plots {.tabset .tabset-fade}
```{r, results='asis', echo = FALSE}
for (i in design_files){
name <- gsub(".csv","",i)
res <- get(gsub("SAMPLE_FILE",i , "res.SAMPLE_FILE"))
cat("## ",name,"\n")
plt <- plot_volcano(res, species=params$species)
print(plt)
cat('\n\n')
}
```
# Histogram of pvalues {.tabset .tabset-fade}
```{r, results='asis'}
use <- res$baseMean > metadata(res)$filterThreshold
h1 <- hist(res$pvalue[!use], breaks=0:50/50, plot=FALSE)
h2 <- hist(res$pvalue[use], breaks=0:50/50, plot=FALSE)
colori <- c(`do not pass`="khaki", `pass`="powderblue")
barplot(height = rbind(h1$counts, h2$counts), beside = FALSE,
col = colori, space = 0, main = "", ylab="frequency")
text(x = c(0, length(h1$counts)), y = 0, label = paste(c(0,1)),
adj = c(0.5,1.7), xpd=NA)
legend("topright", fill=rev(colori), legend=rev(names(colori)))
```
## Results tables {.tabset .tabset-fade}
The following results tables show the significant genes. Filtering has been performed with a log2 fold change +/- 2.
```{r data.setup}
new_list <- list()
names <- c()
for (i in design_files){
name.start <- gsub(".csv","",i)
res <- get(gsub("SAMPLE_FILE",i , "res.SAMPLE_FILE"))
name <- paste0(name.start)
dt <- filter_genes(as.data.frame(res), name=name, species=params$species)
table <- dt@sig
new_list[[paste0("dt_",name,sep="")]] <- table
names <- c(names, name)
assign(paste("sig", i, sep = "."), dt@sig)
assign(paste("res", i, sep = "."), dt@res)
}
df <- data.frame(cbind(new_list, classification=names))
```
```{r , include=FALSE, results = 'asis'}
#for(j in unique(df$classification)){ # You were using level() here, so your for-loop never got off the ground
# df.j <- df[df$classification == j, ]
# cat(paste("\n\n###", j, "\n"))
# df_final <- as.data.frame(df.j$new_list)
# colnames(df_final) <- c("Log2FoldChage", "Symbol", "GeneName", "baseMean", "padj", "Ensembl")
# print( htmltools::tagList(datatable(df_final) ))
#}
```
# Annotate results
```{r annotate}
for (i in design_files){
name.start <- gsub(".csv","",i)
res <- get(gsub("SAMPLE_FILE",i , "sig.SAMPLE_FILE"))
if(any(dim(res)[1] == 0)){
}else{
rownames(res) <- make.unique(res$Row.names)
res_tmp <- Annotate_genes_results(res, species=params$species)
sig_name = paste("results/", name.start,"_annotate_sig.csv", sep="")
write.csv(res_tmp, sig_name)
assign(paste("sig", i, sep = "."), res)
}
}
for (i in design_files){
name.start <- gsub(".csv","",i)
res <- as.data.frame(get(gsub("SAMPLE_FILE",i , "res.SAMPLE_FILE")))
if(any(dim(res)[1] == 0)){
}else{
rownames(res) <- make.unique(res$Row.names)
res_tmp <- Annotate_genes_results(res, species=params$species)
sig_name = paste("results/", name.start,"_annotate_res.csv", sep="")
write.csv(res_tmp, sig_name)
assign(paste("res", i, sep = "."), res)
}
}
```