Skip to content

Commit

Permalink
add manuscript figure scripts
Browse files Browse the repository at this point in the history
  • Loading branch information
ngerstner committed Apr 13, 2022
1 parent 8340144 commit 9ebf49a
Show file tree
Hide file tree
Showing 34 changed files with 1,563 additions and 0 deletions.
89 changes: 89 additions & 0 deletions 04_PlotsManuscript/01_FigureII_C.R
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,89 @@
##################################################
## Project: DexStim Mouse Brain
## Date: 21.11.2021
## Author: Nathalie
##################################################
# GO Enrichment plot for manuscript
# Unique DE and hub genes in PFC

# set working directory to source file location
setwd("~/Documents/ownCloud/DexStim_RNAseq_Mouse/scripts/07_PlotsManuscript")

library(readxl)
library(dplyr)
library(stringr)
library(ggplot2)


# 1. GO enrichment unique DE and hub genes in PFC

data_hub <- read_xlsx(path = "~/Documents/ownCloud/DexStim_RNAseq_Mouse/manuscript/Figures/Figure II_PFC/FUMA_diff unique hubgenes_gene2func67664/PFC diff unique hubgenes_FUMA.xlsx",
sheet = "GO bp")

data_DE <- read_xlsx(path = "~/Documents/ownCloud/DexStim_RNAseq_Mouse/manuscript/Figures/Figure II_PFC/DE PFC/FUMA_gene2func66314/PFC unique DE genes _FUMA.xlsx",
sheet = "GO bp")


# data for left panel with top 14 terms for DE
top_DE <- data_DE %>%
dplyr::filter(N_overlap_A > 24)
top_DE <- top_DE[order(top_DE$`my adj p BH`, decreasing = FALSE),][1:14,]

top_DE_hub <- data_hub %>%
filter(GeneSet %in% top_DE$GeneSet)

top_DE_comb <- bind_rows("de" = top_DE, "hub" = top_DE_hub, .id = "group")


# data for right panel with top 14 terms for hubs
top_hub <- data_hub %>%
dplyr::filter(N_overlap_A > 2)
top_hub <- top_hub[order(top_hub$`my adj p BH`, decreasing = FALSE),][1:14,]

top_hub_DE <- data_DE %>%
filter(GeneSet %in% top_hub$GeneSet)

missing_term <- setdiff(top_hub$GeneSet, top_hub_DE$GeneSet)
add_entry <- data.frame("GeneSet" = missing_term)
top_hub_DE <- bind_rows(top_hub_DE, add_entry)

top_hub_comb <- bind_rows("hub" = top_hub, "de" = top_hub_DE, .id = "group")


# combine everything into one df for plotting
top_data <- bind_rows("de" = top_DE_comb, "hub" = top_hub_comb, .id = "panel")
# remove the "GO_" from GO terms
top_data$GeneSet <- str_replace_all(top_data$GeneSet, '^GO_', ' ')
# remove the "_" from GO terms
top_data$GeneSet <- str_replace_all(top_data$GeneSet, "_", " ")
# only capitalize first letter per word
top_data$GeneSet <- str_to_title(top_data$GeneSet)
# of and to should start with lower case
top_data$GeneSet <- str_replace_all(top_data$GeneSet, "Of", "of")
top_data$GeneSet <- str_replace_all(top_data$GeneSet, "To", "to")

# labeller function for facet titles
facet_names <- c(
'de'="Top 14 GO terms for DE genes",
'hub'="Top 14 GO terms for hub genes"
)

# barplot
g <- ggplot(top_data, aes(x = GeneSet, y = `[-log10 fdr`, fill = group)) +
geom_bar(position = position_dodge2(reverse=TRUE), stat="identity") +

scale_colour_manual(values = c("grey", "black"), guide = FALSE) +
scale_fill_manual(name = "Geneset",
values = c("orange", "darkred"),
labels = c("DE genes", "hub genes")) +
coord_flip() +
scale_x_discrete(limits = rev) +
xlab("GOterm") +
ylab("-log10(FDR)") +
ggtitle("GO terms enriched for unique DE and hub genes in PFC") +
facet_wrap(~panel, scales="free", labeller = as_labeller(facet_names)) +
theme_light() +
theme(text = element_text(size= 14))
print(g)
ggsave("01_FigureII_C.pdf", g, width = 14, height = 8)

83 changes: 83 additions & 0 deletions 04_PlotsManuscript/02_FigureII_F.R
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,83 @@
##################################################
## Project: DexStim Mouse Brain
## Date: 22.11.2021
## Author: Nathalie
##################################################
# Pathway enrichment plot for manuscript
# Abcd1 and neighbourhood in diff network of PFC

library(readxl)

# 1. Pathway enrichment for Abcd1 and neighbours

data <- read_xlsx(path = "~/Documents/ownCloud/DexStim_RNAseq_Mouse/manuscript/Figures/Figure II_PFC/Abcd1 gene neighbourhood_FUMA_gene2func67210/Abcd1_diffnetwork_FUMA.xlsx",
sheet = "KEGG AND REACTOME")

data <- arrange(data, desc(`[-log10FDR]`))
data <- data[1:20,]


# remove the "_" from GO terms
data$GeneSet <- str_replace_all(data$GeneSet, "_", " ")
# only capitalize first letter per word
data$GeneSet <- str_to_title(data$GeneSet)
# of and to should start with lower case
data$GeneSet <- str_replace_all(data$GeneSet, "Of", "of")
data$GeneSet <- str_replace_all(data$GeneSet, "To", "to")
data$GeneSet <- str_replace_all(data$GeneSet, " In ", " in ")
data$GeneSet <- str_replace_all(data$GeneSet, " Into ", " into ")
# Ii from "Polymerase II" back to upper case
data$GeneSet <- str_replace_all(data$GeneSet, "Reactome", "Reactome:")
data$GeneSet <- str_replace_all(data$GeneSet, "Kegg", "KEGG:")
# GeneSet as factor
data$GeneSet <- factor(data$GeneSet, levels = data$GeneSet)



# bar plot gene ratio
gr <- ggplot(data, aes(x = GeneSet, y = `Genes ratio`) ) +
geom_bar(position = position_dodge2(reverse=TRUE), stat="identity", fill = "#0F5057") +
scale_fill_manual() +
theme_light() +
coord_flip() +
scale_x_discrete(limits = rev) +
theme(text = element_text(size= 14)) +
ylab("% Gene Ratio") +
labs(x = NULL)

# bar plot odds ratio
or <- ggplot(data, aes(x = GeneSet, y = `OR[log2(a*d)-log2(b*c)]`) ) +
geom_bar(position = position_dodge2(reverse=TRUE), stat="identity", fill = "#FAA916") +
scale_fill_manual() +
theme_light() +
coord_flip() +
scale_x_discrete(limits = rev) +
theme(text = element_text(size= 14)) +
ylab("Odds Ratio") +
labs(x = NULL)

# barplot fdr p-value
fdr <- ggplot(data, aes(x = GeneSet, y = `[-log10FDR]`) ) +
geom_bar(position = position_dodge2(reverse=TRUE), stat="identity", fill = "#D45E60") +
geom_hline(yintercept = -log10(0.05),linetype="dashed", color = "red") +
theme_light() +
coord_flip() +
scale_x_discrete(limits = rev) +
theme(text = element_text(size= 14)) +
ylab("-log10(FDR)") +
labs(x = NULL)

# combined barplot
comb <- ggarrange(gr,
or +
theme(axis.text.y = element_blank(),
axis.ticks.y = element_blank(),
axis.title.y = element_blank() ),
fdr +
theme(axis.text.y = element_blank(),
axis.ticks.y = element_blank(),
axis.title.y = element_blank() ),
nrow = 1,
widths = c(1.9,1,1))

ggexport(comb, filename = "02_FigureII_F.pdf", width = 14, height = 8)
83 changes: 83 additions & 0 deletions 04_PlotsManuscript/03_FigureIII_B.R
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,83 @@
##################################################
## Project: DexStim Mouse Brain
## Date: 29.11.2021
## Author: Nathalie
##################################################
# GO enrichment plot for manuscript
# vCA1 unique DE genes enrichments

library(readxl)
library(dplyr)
library(stringr)
library(ggpubr)

# 1. GO enrichment for vCA unique DE genes

data <- read_xlsx(path = "~/Documents/ownCloud/DexStim_RNAseq_Mouse/manuscript/Figures/Figure III_vCA1/DE vCA1/FUMA_gene2func67310 (2)/vCA1 Unique DE genes_FUMA enrichments.xlsx",
sheet = "GO bp")

data <- arrange(data, desc(`[-log10 of FDR`))
data <- data[1:20,]


# remove the "GO_" from GO terms
data$GeneSet <- str_replace_all(data$GeneSet, '^GO_', ' ')
# remove the "_" from GO terms
data$GeneSet <- str_replace_all(data$GeneSet, "_", " ")
# only capitalize first letter per word
data$GeneSet <- str_to_title(data$GeneSet)
# of and to should start with lower case
data$GeneSet <- str_replace_all(data$GeneSet, "Of", "of")
data$GeneSet <- str_replace_all(data$GeneSet, "To", "to")
data$GeneSet <- str_replace_all(data$GeneSet, " In ", " in ")
# GeneSet as factor
data$GeneSet <- factor(data$GeneSet, levels = data$GeneSet)


# bar plot gene ratio
gr <- ggplot(data, aes(x = GeneSet, y = `Genes ratio`) ) +
geom_bar(position = position_dodge2(reverse=TRUE), stat="identity", fill = "#0F5057") +
scale_fill_manual() +
theme_light() +
coord_flip() +
scale_x_discrete(limits = rev) +
theme(text = element_text(size= 14)) +
ylab("% Gene Ratio") +
labs(x = NULL)

# bar plot odds ratio
or <- ggplot(data, aes(x = GeneSet, y = `OR[log2(a*d)-log2(b*c)]`) ) +
geom_bar(position = position_dodge2(reverse=TRUE), stat="identity", fill = "#FAA916") +
scale_fill_manual() +
theme_light() +
coord_flip() +
scale_x_discrete(limits = rev) +
theme(text = element_text(size= 14)) +
ylab("Odds Ratio") +
labs(x = NULL)

# barplot fdr p-value
fdr <- ggplot(data, aes(x = GeneSet, y = `[-log10 of FDR`) ) +
geom_bar(position = position_dodge2(reverse=TRUE), stat="identity", fill = "#D45E60") +
geom_hline(yintercept = -log10(0.05),linetype="dashed", color = "red") +
theme_light() +
coord_flip() +
scale_x_discrete(limits = rev) +
theme(text = element_text(size= 14)) +
ylab("-log10(FDR)") +
labs(x = NULL)

# combined barplot
comb <- ggarrange(gr,
or +
theme(axis.text.y = element_blank(),
axis.ticks.y = element_blank(),
axis.title.y = element_blank() ),
fdr +
theme(axis.text.y = element_blank(),
axis.ticks.y = element_blank(),
axis.title.y = element_blank() ),
nrow = 1,
widths = c(2.5,1,1))

ggexport(comb, filename = "03_FigureIII_B.pdf", width = 14, height = 8)
84 changes: 84 additions & 0 deletions 04_PlotsManuscript/04_FigureIII_C.R
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,84 @@
##################################################
## Project: DexStim Mouse Brain
## Date: 29.11.2021
## Author: Nathalie
##################################################
# GO enrichment plot for manuscript
# vCA1 unique DE genes with neighbours enrichments


library(readxl)
library(dplyr)
library(stringr)
library(ggpubr)

# 1. GO enrichment for vCA unique DE genes and their diff neighbours

data <- read_xlsx(path = "~/Documents/ownCloud/DexStim_RNAseq_Mouse/manuscript/Figures/Figure III_vCA1/network vCA1/Unique DEs and their diff neighbours_FUMA_gene2func67313/vCA1 Unique DE genes+diff neighbours_FUMA enrichments.xlsx",
sheet = "GO bp")

data <- arrange(data, desc(`[-LOG10 of FDR`))
data <- data[1:20,]

# remove the "GO_" from GO terms
data$GeneSet <- str_replace_all(data$GeneSet, '^GO_', ' ')
# remove the "_" from GO terms
data$GeneSet <- str_replace_all(data$GeneSet, "_", " ")
# only capitalize first letter per word
data$GeneSet <- str_to_title(data$GeneSet)
# of and to should start with lower case
data$GeneSet <- str_replace_all(data$GeneSet, "Of", "of")
data$GeneSet <- str_replace_all(data$GeneSet, "To", "to")
data$GeneSet <- str_replace_all(data$GeneSet, " In ", " in ")
data$GeneSet <- str_replace_all(data$GeneSet, " Into ", " into ")
# GeneSet as factor
data$GeneSet <- factor(data$GeneSet, levels = data$GeneSet)


# bar plot gene ratio
gr <- ggplot(data, aes(x = GeneSet, y = `Genes ratio`) ) +
geom_bar(position = position_dodge2(reverse=TRUE), stat="identity", fill = "#0F5057") +
scale_fill_manual() +
theme_light() +
coord_flip() +
scale_x_discrete(limits = rev) +
theme(text = element_text(size= 14)) +
ylab("% Gene Ratio") +
labs(x = NULL)

# bar plot odds ratio
or <- ggplot(data, aes(x = GeneSet, y = `OR[log2(a*d)-log2(b*c)]`) ) +
geom_bar(position = position_dodge2(reverse=TRUE), stat="identity", fill = "#FAA916") +
scale_fill_manual() +
theme_light() +
coord_flip() +
scale_x_discrete(limits = rev) +
theme(text = element_text(size= 14)) +
ylab("Odds Ratio") +
labs(x = NULL)

# barplot fdr p-value
fdr <- ggplot(data, aes(x = GeneSet, y = `[-LOG10 of FDR`) ) +
geom_bar(position = position_dodge2(reverse=TRUE), stat="identity", fill = "#D45E60") +
geom_hline(yintercept = -log10(0.05),linetype="dashed", color = "red") +
theme_light() +
coord_flip() +
scale_x_discrete(limits = rev) +
theme(text = element_text(size= 14)) +
ylab("-log10(FDR)") +
labs(x = NULL)

# combined barplot
comb <- ggarrange(gr,
or +
theme(axis.text.y = element_blank(),
axis.ticks.y = element_blank(),
axis.title.y = element_blank() ),
fdr +
theme(axis.text.y = element_blank(),
axis.ticks.y = element_blank(),
axis.title.y = element_blank() ),
nrow = 1,
widths = c(1.8,1,1))

ggexport(comb, filename = "03_FigureIII_C.pdf", width = 12, height = 8)
Loading

0 comments on commit 9ebf49a

Please sign in to comment.