/
F3D_PMD_depth_by_conservation.R
171 lines (129 loc) · 6.83 KB
/
F3D_PMD_depth_by_conservation.R
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
#!/usr/bin/Rscript
library(tidyverse)
library(hexbin)
###########################################
# Observed/expected ratio statistics: gene-level depletion
###########################################
data <- read.table("~/Desktop/Decato-PMD-revision-analysis/OE_statistics/human_genes_inside_pmds_OE",header=TRUE)
ggplot(data,aes(x=log(OE))) +
geom_density()
summary(data$OE) # There is a median OE ratio of 0.67 across all PMD containing samples.
data <- data %>%
filter(ExpOverlap>0)
data <- data %>%
rowwise() %>%
mutate(BinomTestP = (binom.test(ObsOverlap, NumGenes, ExpOverlap/NumGenes))$p.value) %>%
mutate(adjP = p.adjust(BinomTestP)) %>% mutate(sig = ifelse(adjP < 0.05,"Yes","No"))
table(data$sig)
write.table(data, file = "~/Desktop/Decato-PMD-revision-analysis/Gene_OE.tsv", quote = FALSE,
append = FALSE, sep = "\t", row.names = FALSE)
###########################################
# Retrotransposon family-level depletion
###########################################
# F3A suppTable build from output of repeats_in_pmds.sh:
human <- read.table("~/Desktop/Decato-PMD-revision-analysis/F3A_OE_data/Roadmap-2015_Human_Placenta.pmd_PMD_family_OE")
human <- human %>%
mutate(Species = "Human")
mouse <- read.table("~/Desktop/Decato-PMD-revision-analysis/F3A_OE_data/Schroeder-2015-Mouse_Placenta.pmd_PMD_family_OE")
mouse <- mouse %>%
mutate(Species = "Mouse")
sm <- read.table("~/Desktop/Decato-PMD-revision-analysis/F3A_OE_data/Schroeder-2015-SquirrelMonkey_Placenta.pmd_PMD_family_OE")
sm <- sm %>%
mutate(Species = "Squirrel Monkey")
cow <- read.table("~/Desktop/Decato-PMD-revision-analysis/F3A_OE_data/Schroeder-2015-Cow_Placenta.pmd_PMD_family_OE")
cow <- cow %>%
mutate(Species = "Cow")
dog <- read.table("~/Desktop/Decato-PMD-revision-analysis/F3A_OE_data/Schroeder-2015-Dog_Placenta.pmd_PMD_family_OE")
dog <- dog %>%
mutate(Species = "Dog")
horse <- read.table("~/Desktop/Decato-PMD-revision-analysis/F3A_OE_data/Schroeder-2015-Horse_Placenta.pmd_PMD_family_OE")
horse <- horse %>%
mutate(Species = "Horse")
rhesus <- read.table("~/Desktop/Decato-PMD-revision-analysis/F3A_OE_data/Schroeder-2015-Rhesus_Placenta.pmd_PMD_family_OE")
rhesus <- rhesus %>%
mutate(Species = "Rhesus")
fullTable <- rbind(human, cow, dog, horse, rhesus, sm, mouse)
colnames(fullTable) <- c("Retrotransposon family", "Number of PMDs",
"Number of retros in family", "Obs. retros in PMDs", "Exp. retros in PMDs", "O/E ratio", "Species")
# Binomial tests for significance
fullTable <- fullTable %>%
rowwise() %>%
mutate(BinomTestP = (binom.test(`Obs. retros in PMDs`, `Number of retros in family`,
`Exp. retros in PMDs`/`Number of retros in family`))$p.value) %>%
mutate(adjP = p.adjust(BinomTestP)) %>%
mutate(sig = ifelse(adjP < 0.05,"Yes","No"))
table(fullTable$sig)
write.table(fullTable, file = "~/Desktop/Decato-PMD-revision-analysis/F3A_OE_table.tsv", quote = FALSE,
append = FALSE, sep = "\t", row.names = FALSE)
rm(cow, dog, horse, human, mouse, rhesus, sm)
#####################################
# Boundary associations with CpG islands, TSS/TES, CTCF, and chromatin loops
#####################################
cgi <- read.table("~/Desktop/Decato-PMD-revision-analysis/OE_statistics/human_cgi_boundary_OE", header = TRUE)
summary(cgi$OE) # There is a median OE ratio of 1.9 across all PMD containing samples.
cgi <- cgi %>%
rowwise() %>%
mutate(BinomTestP = (binom.test(ObsOverlap, NumCGIs, ExpOverlap/NumCGIs))$p.value) %>%
mutate(adjP = p.adjust(BinomTestP)) %>% mutate(sig = ifelse(adjP < 0.05,"Yes","No")) %>%
mutate(Region = "CGI")
table(cgi$sig)
write.table(cgi, file = "~/Desktop/Decato-PMD-revision-analysis/cgi_OE_table.tsv", quote = FALSE,
append = FALSE, sep = "\t", row.names = FALSE)
#### TSS
tss <- read.table("~/Desktop/Decato-PMD-revision-analysis/OE_statistics/human_TSS_boundary_OE", header = TRUE)
summary(tss$OE) # There is a median OE ratio of 1.65 across all PMD containing samples.
tss <- tss %>%
rowwise() %>%
mutate(BinomTestP = (binom.test(ObsOverlap, NumTSSs, ExpOverlap/NumTSSs))$p.value) %>%
mutate(adjP = p.adjust(BinomTestP)) %>% mutate(sig = ifelse(adjP < 0.05,"Yes","No")) %>%
mutate(Region = "TSS")
table(tss$sig)
write.table(tss, file = "~/Desktop/Decato-PMD-revision-analysis/tss_OE_table.tsv", quote = FALSE,
append = FALSE, sep = "\t", row.names = FALSE)
#### TES
tes <- read.table("~/Desktop/Decato-PMD-revision-analysis/OE_statistics/human_TES_boundary_OE", header = TRUE)
summary(tes$OE) # There is a median OE ratio of 1.65 across all PMD containing samples.
tes <- tes %>%
rowwise() %>%
mutate(BinomTestP = (binom.test(ObsOverlap, NumTESs, ExpOverlap/NumTESs))$p.value) %>%
mutate(adjP = p.adjust(BinomTestP)) %>% mutate(sig = ifelse(adjP < 0.05,"Yes","No")) %>%
mutate(Region = "TES")
table(tes$sig)
write.table(tes, file = "~/Desktop/Decato-PMD-revision-analysis/tes_OE_table.tsv", quote = FALSE,
append = FALSE, sep = "\t", row.names = FALSE)
#### CTCF bound sites
ctcf <- read.table("~/Desktop/Decato-PMD-revision-analysis/CTCF_enrichment/human_ctcf_boundary_OE",header=TRUE)
summary(ctcf$OE)
ctcf <- ctcf %>%
rowwise() %>%
mutate(BinomTestP = (binom.test(ObsOverlap, NumCTCFboundsites, ExpOverlap/NumCTCFboundsites))$p.value) %>%
mutate(adjP = p.adjust(BinomTestP)) %>% mutate(sig = ifelse(adjP < 0.05,"Yes","No"))
table(ctcf$sig)
write.table(ctcf, file = "~/Desktop/Decato-PMD-revision-analysis/ctcf_OE_table.tsv", quote = FALSE,
append = FALSE, sep = "\t", row.names = FALSE)
#### Chromatin loop boundaries
loops <- read.table("~/Desktop/Decato-PMD-revision-analysis/OE_statistics/loop_OE", header = TRUE)
summary(loops$OE) # There is a median OE ratio of 1.65 across all PMD containing samples.
loops <- loops %>%
rowwise() %>%
mutate(BinomTestP = (binom.test(ObsOverlap, NumLoopBoundaries, ExpOverlap/NumLoopBoundaries))$p.value) %>%
mutate(adjP = p.adjust(BinomTestP)) %>% mutate(sig = ifelse(adjP < 0.05,"Yes","No"))
table(loops$sig)
write.table(loops, file = "~/Desktop/Decato-PMD-revision-analysis/loops_OE_table.tsv", quote = FALSE,
append = FALSE, sep = "\t", row.names = FALSE)
#####################################
# Figure 3D.
#####################################
data<-read.table("~/Desktop/Decato-PMD-revision-analysis/F3D_depth_by_conservation.txt",header=TRUE)
smoothScatter(data$meanMeth~data$numSamples)
model <- lm(data=data, meanMeth~numSamples)
summary(model)
# Figure 3D.
#smoothScatter(data$numSamples,data$meanMeth)
ggplot(data,aes(x=numSamples,y=meanMeth)) +
geom_hex() +
stat_smooth(method=lm, color="red") +
theme_bw() +
theme(legend.position = "bottom", text = element_text(size=14), axis.text = element_text(size = 10),
axis.text.x = element_text(angle = 45, hjust = 1), strip.background = element_blank(),
strip.placement = "outside")