-
Notifications
You must be signed in to change notification settings - Fork 1
/
DESeq.Rmd
308 lines (257 loc) · 9.87 KB
/
DESeq.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
---
title: "DESeq2"
output: html_document
---
Rmd to use `DESeq2` package on 4 libraries count matrix. Differential gene expression between infection statuses, then influence of temperature on the expression of those differentially expressed genes.
Library info --> all from sampling day 2
380822_cold_uninfected
380823_cold_infected
380824_warm_uninfected
380825_warm_infected
Load packages:
```{r}
library(DESeq2)
library(tidyverse)
library(pheatmap)
library(RColorBrewer)
library(data.table)
```
### Read in count matrix for the 4 libraries from transcriptome v 3.1
file created using https://github.com/RobertsLab/paper-tanner-crab/blob/master/notebooks/kallisto-4libraries.ipynb
```{r}
countmatrix <- read.delim("../analyses/kallisto-0812/kallisto-0812.isoform.counts.matrix", header = TRUE, sep = '\t')
rownames(countmatrix) <- countmatrix$X
countmatrix <- countmatrix[,-1]
head(countmatrix)
```
Rename columns just to remove the "X" before each sample name:
```{r}
colnames(countmatrix) <- c("380822_cold_uninfected", "380823_cold_infected", "380824_warm_uninfected", "380825_warm_infected")
head(countmatrix)
```
Round integers up to hole numbers for further analysis:
```{r}
countmatrix <- round(countmatrix, 0)
head(countmatrix)
```
## Get DEGs based on infection
Not including temperature as part of the design
```{r}
colData <- data.frame(condition=factor(c("uninfected", "infected", "uninfected", "infected")),
type=factor(rep("paired-end", 4)))
rownames(colData) <- colnames(countmatrix)
dds <- DESeqDataSetFromMatrix(countData = countmatrix,
colData = colData,
design = ~ condition)
```
```{r}
dds <- DESeq(dds)
res <- results(dds)
res <- res[order(rownames(res)), ]
```
```{r}
head(res)
```
```{r}
# Count number of hits with adjusted p-value less then 0.05
dim(res[!is.na(res$padj) & res$padj <= 0.05, ])
```
1343 rows! DEGs just based on infection status (irrespective of temperature)
```{r}
infection_fig <- res
# The main plot
plot(infection_fig$baseMean, infection_fig$log2FoldChange, pch=20, cex=0.45, ylim=c(-15, 15), log="x", col="darkgray",
main="Infection Status (pval </= 0.05)",
xlab="mean of normalized counts",
ylab="Log2 Fold Change")
# Getting the significant points and plotting them again so they're a different color
infection_fig.sig <- res[!is.na(res$padj) & res$padj <= 0.05, ]
points(infection_fig.sig$baseMean, infection_fig.sig$log2FoldChange, pch=20, cex=0.45, col="red")
# 2 FC lines
abline(h=c(-1,1), col="blue")
```
```{r}
#write.table(infection_fig.sig, "../analyses/DESeq2/DEGlist-infectionONLY.tab", sep = "\t", row.names = T, quote = FALSE, col.names = TRUE)
```
Wrote out file 08/12/2020. Commented out code.
## Get influence of temperature on the DEGs from above
Taking temperature into account in the design.
Code based on the Multi-factor design section of the manual: http://bioconductor.org/packages/release/bioc/vignettes/DESeq2/inst/doc/DESeq2.html#multi-factor-designs
```{r}
deseq2.colData <- data.frame(condition=factor(c("uninfected", "infected", "uninfected", "infected")),
type=factor(rep("paired-end", 4)),
temperature=factor(c("decreased", "decreased", "elevated", "elevated")))
rownames(deseq2.colData) <- colnames(countmatrix)
deseq2.dds <- DESeqDataSetFromMatrix(countData = countmatrix,
colData = deseq2.colData,
design = ~ condition + temperature)
```
Check levels of temperature and condition (infection status)
```{r}
levels(deseq2.dds$temperature)
```
```{r}
levels(deseq2.dds$condition)
```
`DESeq2` automatically puts the levels in alphabetical order and the first listed level is the reference level for the factor.
We want uninfected to be the reference.
```{r}
deseq2.dds$condition = relevel(deseq2.dds$condition, "uninfected")
levels(deseq2.dds$condition)
```
The following will pull the results from `condition` (infection status) because that is our variable of interest. This tells us how temperature contributes to the infection DEGs
```{r}
design(deseq2.dds) <- formula(~ temperature + condition)
deseq2.dds <- DESeq(deseq2.dds)
```
Access results:
```{r}
deseq2.res <- results(deseq2.dds)
head(deseq2.res)
```
```{r}
summary(deseq2.res)
```
```{r}
# Count number of hits with adjusted p-value less then 0.05
dim(deseq2.res[!is.na(deseq2.res$padj) & deseq2.res$padj <= 0.05, ])
```
408 DEGs between infection statuses that are influence by temperature
```{r}
inf <- deseq2.res
# The main plot
plot(inf$baseMean, inf$log2FoldChange, pch=20, cex=0.45, ylim=c(-15, 15), log="x", col="darkgray",
#main="Infection Status (pval </= 0.05)",
xlab="mean of normalized counts",
ylab="Log2 Fold Change")
# Getting the significant points and plotting them again so they're a different color
inf.sig <- deseq2.res[!is.na(deseq2.res$padj) & deseq2.res$padj <= 0.05, ]
points(inf.sig$baseMean, inf.sig$log2FoldChange, pch=20, cex=0.45, col="red")
# 2 FC lines
abline(h=c(-1,1), col="blue")
```
```{r}
#write.table(inf.sig, "../analyses/DESeq2/DEGlist-infection-with-temp.tab", sep = "\t", row.names = T, quote = FALSE, col.names = TRUE)
```
Wrote out table 08/13/2020. Commented out code
## Now to figure out what these results mean:
In the multifactor section of the `DESeq2` manual:
The contrast argument of the function _results_ needs a character vector of three componenets: the name of the variable (in this case "temperature"), and the name of the factor level for the numerator of the log2 ratio (elevated) and the denominator (decreased)
A **contrast** is a linear combination of estimated log2 fold changes. Can be used to test if differences between groups are equal to zero.
```{r}
resultsNames(deseq2.dds)
```
```{r}
deseq2.resTemp <- results(deseq2.dds,
contrast = c("temperature", "elevated", "decreased"))
head(deseq2.resTemp)
```
```{r}
tmp <- deseq2.resTemp
# The main plot
plot(tmp$baseMean, tmp$log2FoldChange, pch=20, cex=0.45, ylim=c(-15, 15), log="x", col="darkgray",
#main="Infection Status (pval </= 0.05)",
xlab="mean of normalized counts",
ylab="Log2 Fold Change")
# Getting the significant points and plotting them again so they're a different color
tmp.sig <- deseq2.resTemp[!is.na(deseq2.resTemp$padj) & deseq2.resTemp$padj <= 0.05, ]
points(tmp.sig$baseMean, tmp.sig$log2FoldChange, pch=20, cex=0.45, col="red")
# 2 FC lines
abline(h=c(-1,1), col="blue")
```
```{r}
# Count number of hits with adjusted p-value less then 0.05
dim(deseq2.resTemp[!is.na(deseq2.resTemp$padj) & deseq2.resTemp$padj <= 0.05, ])
```
123 DEGs - genes that are influenced by temperature that are associated with infection status
```{r}
summary(deseq2.resTemp)
```
```{r}
#write.table(tmp.sig, "../analyses/DESeq2/DEGlist-contrast_temperature.tab", sep = "\t", row.names = T, quote = FALSE, col.names = TRUE)
```
Wrote out 08/12/2020. Comment out code.
## `join` count matrix with lists of DEGs
Make rownames into a column called "Trinity_ID" for the count matrix:
```{r}
countmatrix <- tibble::rownames_to_column(countmatrix, "Trinity_ID")
head(countmatrix)
```
### Read in list of 1343 DEGs:
```{r}
degs_infONLY <- read.delim("../analyses/DESeq2/DEGlist-infectionONLY.tab", sep = '\t')
head(degs_infONLY)
```
Make rownames into a column called "Trinity_ID":
```{r}
degs_infONLY <- tibble::rownames_to_column(degs_infONLY, "Trinity_ID")
head(degs_infONLY)
```
`join` count matrix with degs_infONLY:
```{r}
degs_infONLY_counts <- left_join(countmatrix, degs_infONLY, by = "Trinity_ID")
head(degs_infONLY_counts)
```
Only want the ones that match!
```{r}
degs_infONLY_counts_match <- filter(degs_infONLY_counts, baseMean != "NA")
head(degs_infONLY_counts_match)
```
1343 rows. Perfect!
Write out list with count data:
```{r}
#write.table(degs_infONLY_counts_match, "../analyses/DESeq2/DEGlist-infectionONLY-counts.tab", sep = '\t', quote = FALSE, row.names = FALSE)
```
wrote out 08/13/2020. Comment out code.
### Read in list of 183 DEGs:
```{r}
degs_inftemp <- read.delim("../analyses/DESeq2/DEGlist-infection_influenced-by-temp.tab", sep = '\t')
head(degs_inftemp)
```
Make rownames into a column called "Trinity_ID":
```{r}
degs_inftemp <- tibble::rownames_to_column(degs_inftemp, "Trinity_ID")
head(degs_inftemp)
```
`join` count matrix with degs_inftemp:
```{r}
degs_inftemp_counts <- left_join(countmatrix, degs_inftemp, by = "Trinity_ID")
head(degs_inftemp_counts)
```
Only want the ones that match!
```{r}
degs_inftemp_counts_match <- filter(degs_inftemp_counts, baseMean != "NA")
head(degs_inftemp_counts_match)
```
Write out list with count data:
```{r}
#write.table(degs_inftemp_counts_match, "../analyses/DESeq2/DEGlist-infection-with-temp-counts.tab", sep = '\t', quote = FALSE, row.names = FALSE)
```
Wrote out 08/13/2020. Commented out code.
### Read in list of 123 DEGs:
DEG list of genes influenced by temperature associated with infection status comparison.
```{r}
degs_contrasttemp <- read.delim("../analyses/DESeq2/DEGlist-contrast_temperature.tab", sep = '\t')
head(degs_contrasttemp)
```
Make rownames into a column called "Trinity_ID":
```{r}
degs_contrasttemp <- tibble::rownames_to_column(degs_contrasttemp, "Trinity_ID")
head(degs_contrasttemp)
```
`join` count matrix with degs_inftemp:
```{r}
degs_contrasttemp_counts <- left_join(countmatrix, degs_contrasttemp, by = "Trinity_ID")
head(degs_contrasttemp_counts)
```
Only want the ones that match!
```{r}
degs_contrasttemp_counts_match <- filter(degs_contrasttemp_counts, baseMean != "NA")
head(degs_contrasttemp_counts_match)
```
123 rows! Good.
Write out list with count data:
```{r}
#write.table(degs_contrasttemp_counts_match, "../analyses/DESeq2/DEGlist-contrast_temperature-counts.tab", sep = '\t', quote = FALSE, row.names = FALSE)
```
wrote out 08/13/2020. commented out code.