-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathDESeq2.Rmd
216 lines (197 loc) · 13.9 KB
/
DESeq2.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
---
title: "DESeq2"
author: "QD"
output: html_document
---
## Here differential abundance analysis with DESeq2 will be performed. First, filter genus table such that genera should be present in at least 25% of samples
```{r}
set.seed(1000)
genus_25 <- core(genus_005_samples,detection=0, prevalence = 25/100)
```
## To account for repeated measures, compare per time point (so basically just stratify per time point)
```{r}
genus_25_tp1<-subset_samples(genus_25,Time_point=="1")
DESeq2_test = phyloseq_to_deseq2(genus_25_tp1, ~ MDRO_cultured)
DESeq2_test = DESeq(DESeq2_test, test="Wald", fitType="parametric")
res = results(DESeq2_test, cooksCutoff = FALSE)
alpha = 0.1
sigtab = res[which(res$padj < alpha), ]
sigtab = cbind(as(sigtab, "data.frame"), as(tax_table(genus_25_tp1)[rownames(sigtab), ], "matrix"))
head(sigtab)
x = tapply(sigtab$log2FoldChange, sigtab$Phylum, function(x) max(x))
x = sort(x, TRUE)
sigtab$Phylum = factor(as.character(sigtab$Phylum), levels=names(x))
# Genus order
x = tapply(sigtab$log2FoldChange, sigtab$Genus, function(x) max(x))
x = sort(x, TRUE)
sigtab$Genus = factor(as.character(sigtab$Genus), levels=names(x))
ggplot(sigtab, aes(x=Genus, y=log2FoldChange, color=Phylum)) + geom_point(size=6) +theme(axis.text.x = element_text(angle = -90, hjust = 0, vjust=0.5))
x = tapply(sigtab$log2FoldChange, sigtab$Genus, function(x) max(x))
x = sort(x, TRUE)
sigtab$Genus = factor(as.character(sigtab$Genus), levels=names(x))
ggplot(sigtab, aes(x=Genus, y=log2FoldChange, color="MDRO.positive")) + geom_point(size=6) +
theme(axis.text.x = element_text(angle = -45, hjust = 0, vjust=0.5)) + labs(title = "Differential abundance between MDRO positive and negative fecal samples", x="Genus") + guides(color=guide_legend("Colonization status")) + scale_color_hue(labels = c("MDRO positive", "MDRO negative"))
sigtab_MDRO_positive_vs_negative_samples<-sigtab
sigtab_MDRO_positive_vs_negative_samples<-dplyr::select(sigtab_MDRO_positive_vs_negative_samples,-c(baseMean,lfcSE,stat,pvalue,Kingdom))
sigtab_MDRO_positive_vs_negative_samples$Phylum<-gsub("p__","",sigtab_MDRO_positive_vs_negative_samples$Phylum)
sigtab_MDRO_positive_vs_negative_samples$Class<-gsub("c__","",sigtab_MDRO_positive_vs_negative_samples$Class)
sigtab_MDRO_positive_vs_negative_samples$Order<-gsub("o__","",sigtab_MDRO_positive_vs_negative_samples$Order)
sigtab_MDRO_positive_vs_negative_samples$Family<-gsub("f__","",sigtab_MDRO_positive_vs_negative_samples$Family)
sigtab_MDRO_positive_vs_negative_samples$Genus<-gsub("g__","",sigtab_MDRO_positive_vs_negative_samples$Genus)
##Now order the data frame on log2fold change and adjusted p-value
sigtab_MDRO_positive_vs_negative_samples<-sigtab_MDRO_positive_vs_negative_samples[order(sigtab_MDRO_positive_vs_negative_samples$log2FoldChange, sigtab_MDRO_positive_vs_negative_samples$padj),]
sigtab_MDRO_positive_vs_negative_samples$Timepoint<-"Time point 1"
sigtab_MDRO_positive_vs_negative_samples$otu_nr<-rownames(sigtab_MDRO_positive_vs_negative_samples)
write.csv(sigtab_MDRO_positive_vs_negative_samples,"Timepoint1_deseq2.csv", row.names = FALSE)
```
## Timepoint 2
```{r}
genus_25_tp2<-subset_samples(genus_25,Time_point=="2")
DESeq2_test = phyloseq_to_deseq2(genus_25_tp2, ~ MDRO_cultured)
DESeq2_test = DESeq(DESeq2_test, test="Wald", fitType="parametric")
res = results(DESeq2_test, cooksCutoff = FALSE)
alpha = 0.1
sigtab = res[which(res$padj < alpha), ]
sigtab = cbind(as(sigtab, "data.frame"), as(tax_table(genus_25_tp2)[rownames(sigtab), ], "matrix"))
head(sigtab)
x = tapply(sigtab$log2FoldChange, sigtab$Phylum, function(x) max(x))
x = sort(x, TRUE)
sigtab$Phylum = factor(as.character(sigtab$Phylum), levels=names(x))
# Genus order
x = tapply(sigtab$log2FoldChange, sigtab$Genus, function(x) max(x))
x = sort(x, TRUE)
sigtab$Genus = factor(as.character(sigtab$Genus), levels=names(x))
ggplot(sigtab, aes(x=Genus, y=log2FoldChange, color=Phylum)) + geom_point(size=6) +theme(axis.text.x = element_text(angle = -90, hjust = 0, vjust=0.5))
x = tapply(sigtab$log2FoldChange, sigtab$Genus, function(x) max(x))
x = sort(x, TRUE)
sigtab$Genus = factor(as.character(sigtab$Genus), levels=names(x))
ggplot(sigtab, aes(x=Genus, y=log2FoldChange, color="MDRO.positive")) + geom_point(size=6) +
theme(axis.text.x = element_text(angle = -45, hjust = 0, vjust=0.5)) + labs(title = "Differential abundance between MDRO positive and negative fecal samples", x="Genus") + guides(color=guide_legend("Colonization status")) + scale_color_hue(labels = c("MDRO positive", "MDRO negative"))
sigtab_MDRO_positive_vs_negative_samples<-sigtab
sigtab_MDRO_positive_vs_negative_samples<-dplyr::select(sigtab_MDRO_positive_vs_negative_samples,-c(baseMean,lfcSE,stat,pvalue,Kingdom))
sigtab_MDRO_positive_vs_negative_samples$Phylum<-gsub("p__","",sigtab_MDRO_positive_vs_negative_samples$Phylum)
sigtab_MDRO_positive_vs_negative_samples$Class<-gsub("c__","",sigtab_MDRO_positive_vs_negative_samples$Class)
sigtab_MDRO_positive_vs_negative_samples$Order<-gsub("o__","",sigtab_MDRO_positive_vs_negative_samples$Order)
sigtab_MDRO_positive_vs_negative_samples$Family<-gsub("f__","",sigtab_MDRO_positive_vs_negative_samples$Family)
sigtab_MDRO_positive_vs_negative_samples$Genus<-gsub("g__","",sigtab_MDRO_positive_vs_negative_samples$Genus)
##Now order the data frame on log2fold change and adjusted p-value
sigtab_MDRO_positive_vs_negative_samples<-sigtab_MDRO_positive_vs_negative_samples[order(sigtab_MDRO_positive_vs_negative_samples$log2FoldChange, sigtab_MDRO_positive_vs_negative_samples$padj),]
sigtab_MDRO_positive_vs_negative_samples$Timepoint<-"Time point 2"
sigtab_MDRO_positive_vs_negative_samples$otu_nr<-rownames(sigtab_MDRO_positive_vs_negative_samples)
write.csv(sigtab_MDRO_positive_vs_negative_samples,"Timepoint2_deseq2.csv", row.names = FALSE)
```
```{r}
genus_25_tp3<-subset_samples(genus_25,Time_point=="3")
DESeq2_test = phyloseq_to_deseq2(genus_25_tp3, ~ MDRO_cultured)
DESeq2_test = DESeq(DESeq2_test, test="Wald", fitType="parametric")
res = results(DESeq2_test, cooksCutoff = FALSE)
alpha = 0.1
sigtab = res[which(res$padj < alpha), ]
sigtab = cbind(as(sigtab, "data.frame"), as(tax_table(genus_25_tp3)[rownames(sigtab), ], "matrix"))
head(sigtab)
x = tapply(sigtab$log2FoldChange, sigtab$Phylum, function(x) max(x))
x = sort(x, TRUE)
sigtab$Phylum = factor(as.character(sigtab$Phylum), levels=names(x))
# Genus order
x = tapply(sigtab$log2FoldChange, sigtab$Genus, function(x) max(x))
x = sort(x, TRUE)
sigtab$Genus = factor(as.character(sigtab$Genus), levels=names(x))
ggplot(sigtab, aes(x=Genus, y=log2FoldChange, color=Phylum)) + geom_point(size=6) +theme(axis.text.x = element_text(angle = -90, hjust = 0, vjust=0.5))
x = tapply(sigtab$log2FoldChange, sigtab$Genus, function(x) max(x))
x = sort(x, TRUE)
sigtab$Genus = factor(as.character(sigtab$Genus), levels=names(x))
ggplot(sigtab, aes(x=Genus, y=log2FoldChange, color="MDRO.positive")) + geom_point(size=6) +
theme(axis.text.x = element_text(angle = -45, hjust = 0, vjust=0.5)) + labs(title = "Differential abundance between MDRO positive and negative fecal samples", x="Genus") + guides(color=guide_legend("Colonization status")) + scale_color_hue(labels = c("MDRO positive", "MDRO negative"))
sigtab_MDRO_positive_vs_negative_samples<-sigtab
sigtab_MDRO_positive_vs_negative_samples<-dplyr::select(sigtab_MDRO_positive_vs_negative_samples,-c(baseMean,lfcSE,stat,pvalue,Kingdom))
sigtab_MDRO_positive_vs_negative_samples$Phylum<-gsub("p__","",sigtab_MDRO_positive_vs_negative_samples$Phylum)
sigtab_MDRO_positive_vs_negative_samples$Class<-gsub("c__","",sigtab_MDRO_positive_vs_negative_samples$Class)
sigtab_MDRO_positive_vs_negative_samples$Order<-gsub("o__","",sigtab_MDRO_positive_vs_negative_samples$Order)
sigtab_MDRO_positive_vs_negative_samples$Family<-gsub("f__","",sigtab_MDRO_positive_vs_negative_samples$Family)
sigtab_MDRO_positive_vs_negative_samples$Genus<-gsub("g__","",sigtab_MDRO_positive_vs_negative_samples$Genus)
##Now order the data frame on log2fold change and adjusted p-value
sigtab_MDRO_positive_vs_negative_samples<-sigtab_MDRO_positive_vs_negative_samples[order(sigtab_MDRO_positive_vs_negative_samples$log2FoldChange, sigtab_MDRO_positive_vs_negative_samples$padj),]
sigtab_MDRO_positive_vs_negative_samples$Timepoint<-"Time point 3"
sigtab_MDRO_positive_vs_negative_samples$otu_nr<-rownames(sigtab_MDRO_positive_vs_negative_samples)
write.csv(sigtab_MDRO_positive_vs_negative_samples,"Timepoint3_deseq2.csv", row.names = FALSE)
```
```{r}
genus_25_tp4<-subset_samples(genus_25,Time_point=="4")
DESeq2_test = phyloseq_to_deseq2(genus_25_tp4, ~ MDRO_cultured)
DESeq2_test = DESeq(DESeq2_test, test="Wald", fitType="parametric")
res = results(DESeq2_test, cooksCutoff = FALSE)
alpha = 0.1
sigtab = res[which(res$padj < alpha), ]
sigtab = cbind(as(sigtab, "data.frame"), as(tax_table(genus_25_tp4)[rownames(sigtab), ], "matrix"))
head(sigtab)
x = tapply(sigtab$log2FoldChange, sigtab$Phylum, function(x) max(x))
x = sort(x, TRUE)
sigtab$Phylum = factor(as.character(sigtab$Phylum), levels=names(x))
# Genus order
x = tapply(sigtab$log2FoldChange, sigtab$Genus, function(x) max(x))
x = sort(x, TRUE)
sigtab$Genus = factor(as.character(sigtab$Genus), levels=names(x))
ggplot(sigtab, aes(x=Genus, y=log2FoldChange, color=Phylum)) + geom_point(size=6) +theme(axis.text.x = element_text(angle = -90, hjust = 0, vjust=0.5))
x = tapply(sigtab$log2FoldChange, sigtab$Genus, function(x) max(x))
x = sort(x, TRUE)
sigtab$Genus = factor(as.character(sigtab$Genus), levels=names(x))
ggplot(sigtab, aes(x=Genus, y=log2FoldChange, color="MDRO.positive")) + geom_point(size=6) +
theme(axis.text.x = element_text(angle = -45, hjust = 0, vjust=0.5)) + labs(title = "Differential abundance between MDRO positive and negative fecal samples", x="Genus") + guides(color=guide_legend("Colonization status")) + scale_color_hue(labels = c("MDRO positive", "MDRO negative"))
sigtab_MDRO_positive_vs_negative_samples<-sigtab
sigtab_MDRO_positive_vs_negative_samples<-dplyr::select(sigtab_MDRO_positive_vs_negative_samples,-c(baseMean,lfcSE,stat,pvalue,Kingdom))
sigtab_MDRO_positive_vs_negative_samples$Phylum<-gsub("p__","",sigtab_MDRO_positive_vs_negative_samples$Phylum)
sigtab_MDRO_positive_vs_negative_samples$Class<-gsub("c__","",sigtab_MDRO_positive_vs_negative_samples$Class)
sigtab_MDRO_positive_vs_negative_samples$Order<-gsub("o__","",sigtab_MDRO_positive_vs_negative_samples$Order)
sigtab_MDRO_positive_vs_negative_samples$Family<-gsub("f__","",sigtab_MDRO_positive_vs_negative_samples$Family)
sigtab_MDRO_positive_vs_negative_samples$Genus<-gsub("g__","",sigtab_MDRO_positive_vs_negative_samples$Genus)
##Now order the data frame on log2fold change and adjusted p-value
sigtab_MDRO_positive_vs_negative_samples<-sigtab_MDRO_positive_vs_negative_samples[order(sigtab_MDRO_positive_vs_negative_samples$log2FoldChange, sigtab_MDRO_positive_vs_negative_samples$padj),]
sigtab_MDRO_positive_vs_negative_samples$Timepoint<-"Time point 4"
sigtab_MDRO_positive_vs_negative_samples$otu_nr<-rownames(sigtab_MDRO_positive_vs_negative_samples)
write.csv(sigtab_MDRO_positive_vs_negative_samples,"Timepoint4_deseq2.csv", row.names = FALSE)
## Since Phascolarctobacterium has such a big log2fold change, quickly check visually
genus_25_tp4_rel<-microbiome::transform(genus_25_tp4,"compositional")
check_phasco<-psmelt(genus_25_tp4_rel)
check_phasco<-subset(check_phasco,OTU=="322282315")
```
## Visualize the different time point results from DESeq2. Here a p-value of 0.1 has been used for visualization purposes, since the sample size is quite limited
```{r}
all_timepoints_deseq2<-read_excel("Timepoints_combined_deseq2.xlsx")
all_timepoints_deseq2$Genus_OTU_nr<-paste(all_timepoints_deseq2$Genus,all_timepoints_deseq2$otu_nr,sep="_")
fct_relevel(all_timepoints_deseq2$Genus_OTU_nr)
my_palette<-colorRampPalette(rev(brewer.pal(n = 7, name ="RdYlBu")))(100) ## Filter out the NA, as it doesn't say anything
all_timepoints_deseq2<-subset(all_timepoints_deseq2,otu_nr!="322282318") ## Since we do not want to visualize unassigned reads.
deseq2_heatmap<-ggplot(all_timepoints_deseq2)+
aes(x=Timepoint,y=forcats::fct_rev(factor(Genus)),fill=log2FoldChange)+
geom_tile(colour="black")+
geom_text(aes(label=round(padj,digits=3)))+
scale_fill_gradientn(colours=my_palette,na.value="white")+
labs(fill="Log2 Fold Change")+
theme_classic()+
ylab("Genus")+
xlab("")+
theme(axis.text.y = element_text(face = "italic"))
ggsave(file="FigS4_heatmap_deseq2.svg", plot=deseq2_heatmap, width=12, height=8,dpi=1600)
```
## Compare the intra-individual groups
```{r}
genus_25_transient_pre<-subset_samples(genus_25,Transient_colonization_pre=="pre colonisation"|Transient_colonization_pre=="colonisation")
DESeq2_test = phyloseq_to_deseq2(genus_25_transient_pre, ~ Residentnumber + Transient_colonization_pre) ## Include Residentnumber for paired analysis
DESeq2_test = DESeq(DESeq2_test, test="Wald", fitType="parametric")
res = results(DESeq2_test, cooksCutoff = FALSE)
alpha = 0.05 ##
sigtab = res[which(res$padj < alpha), ]
sigtab = cbind(as(sigtab, "data.frame"), as(tax_table(genus_25_transient_pre)[rownames(sigtab), ], "matrix"))
head(sigtab) ## So there are simply no differences when correcting for individual, lowest p-value is 0.77......
```
##Now intra-individual group
```{r}
genus_25_transient_post<-subset_samples(genus_25,Transient_colonization_post=="post colonisation"|Transient_colonization_post=="colonisation")
DESeq2_test = phyloseq_to_deseq2(genus_25_transient_post, ~ Residentnumber + Transient_colonization_post) ## Include Residentnumber for paired analysis
DESeq2_test = DESeq(DESeq2_test, test="Wald", fitType="parametric")
res = results(DESeq2_test, cooksCutoff = FALSE)
alpha = 0.05 ##
sigtab = res[which(res$padj < alpha), ]
sigtab = cbind(as(sigtab, "data.frame"), as(tax_table(genus_25_transient_post)[rownames(sigtab), ], "matrix"))
head(sigtab) ## So there are simply no differences when correcting for individual, lowest p-value is 0.98......
```