pkgs <- c("ggthemes", "jhtools", "glue", "ggsci", "patchwork", "tidyverse",
"circlize", "ComplexHeatmap", "SummarizedExperiment", "jhuanglabRNAseq",
"viridis","ggrepel", "tidygraph","ggraph")
for (pkg in pkgs){
suppressPackageStartupMessages(library(pkg, character.only = T))
}
out_dir <- "./results/figure1" %>% checkdir
in_dir <- "./data" %>% checkdirfigure1
figure1A-1
mx <- glue("{in_dir}/raw_big_heatmap.rds") %>% read_rds
cluster_annotation <- glue("{in_dir}/raw_big_heatmap_anno.rds") %>% read_rds
group_color <- glue("{in_dir}/group_color.rds") %>% read_rds
bigcod <- glue("{in_dir}/bigcod.rds") %>% read_rds
ft <- cluster_annotation$label != "others"
ha <- HeatmapAnnotation(
label = cluster_annotation$biglabel[ft],
col = list(
label = group_color
),
simple_anno_size = unit(1.8, "mm"),
show_legend = F,
show_annotation_name = F
)
ht <- Heatmap(t(mx)[,ft],
column_split = factor(cluster_annotation$biglabel[ft], levels = bigcod),
clustering_method_columns = "ward.D2",
clustering_distance_columns = "spearman",
cluster_rows = F, cluster_columns = T,
show_row_names = F, show_column_names = F, top_annotation = ha,
column_title_rot = 45,
use_raster = F,
heatmap_width = unit(200, "mm"),
column_dend_height = unit(5, "mm"),
column_title_gp = gpar(fontsize = 9),
column_title = NULL,
border = T,
border_gp = gpar(lwd = 0.5),
column_gap = unit(0, "mm"), show_heatmap_legend = F)
pdf(glue("{out_dir}/figure1A1.pdf"), width = 8.6, height = 2.5)
draw(ht)
d <- dev.off()
ht
figure1A-2
mx2 <- read_rds(glue("{in_dir}/raw_sub_heatmap.rds"))
cluster_annotation2 <- read_rds(glue("{in_dir}/raw_subheatmap_anno.rds"))
colDendrogram <- read_rds(glue("{in_dir}/colDendrogram.rds"))
spv<- read_rds(glue("{in_dir}/raw_sub_ids.rds"))
ha2 <- HeatmapAnnotation(
label = cluster_annotation2$label,
IDH = cluster_annotation2$IDH,
FH_A273T = cluster_annotation2$FH_A273T,
col = list(
label = group_color,
IDH = c("IDH1_132" = "black", "IDH2_140" = "black"),
FH_A273T = c("FH_A273T" = "black")
),
simple_anno_size = unit(1.8, "mm"),
annotation_name_gp = grid::gpar(fontsize = 7),
show_legend = F,
show_annotation_name = T
)
ht <- Heatmap(t(mx2)[,spv],
cluster_columns = colDendrogram,
clustering_method_columns = "ward.D2",
clustering_distance_columns = "pearson",
cluster_rows = F,
show_row_names = F, show_column_names = F, top_annotation = ha2,
column_title_rot = 45,
column_dend_height = unit(3, "mm"),
heatmap_width = unit(50, "mm"),
use_raster = F,
column_title_gp = gpar(fontsize = 8.5),
border = T,
border_gp = gpar(lwd = 0.5),
column_gap = unit(0, "mm"), show_heatmap_legend = F)
pdf(glue("{out_dir}/figure1A2.pdf"), width = 4, height = 3.5)
draw(ht)
d <- dev.off()
ht
figure1B-1
umap_frame <- glue("{in_dir}/umap_frame.rds") %>% read_rds()
umap_label_frame <- glue("{in_dir}/umap_label_frame.rds") %>% read_rds()
group_color <- glue("{in_dir}/group_color.rds") %>% read_rds()
subcod <- glue("{in_dir}/subcod.rds") %>% read_rds()
a <- umap_frame %>%
dplyr::filter(sub_groups != "others") %>%
ggplot(aes(x = V1, y = V2, color = sub_groups)) +
geom_point(size = 0.5) +
theme_void() +
scale_color_manual(values = group_color) +
theme(legend.position = "none") +
geom_text_repel(data = umap_label_frame %>%
dplyr::filter(!sub_groups %in% subcod) %>%
dplyr::filter(sub_groups != "others"),
mapping = aes(x = x, y = y, label = sub_groups),
color = "black",size = 3,
max.overlaps = 1000)+
geom_rect(xmin = 3.4, xmax = 5.3, ymin = -1.3, ymax = 0.026,
fill = NA, linewidth = 0.5, color = "red", linetype = "dashed")
pdf(glue("{out_dir}/figure1B1.pdf"), width = 4, height = 3.5)
print(a)
d <- dev.off()
a
figure1B-2
umap_frame <- glue("{in_dir}/umap_frame.rds") %>% read_rds()
subcod <- glue("{in_dir}/subcod.rds") %>% read_rds()
umap_label_frame_sub <- glue("{in_dir}/umap_label_frame_sub.rds") %>% read_rds()
group_color <- glue("{in_dir}/group_color.rds") %>% read_rds
a <- umap_frame %>%
dplyr::filter(sub_groups %in% subcod) %>%
dplyr::filter(V1 > 2.3) %>%
ggplot(aes(x = V1, y = V2, color = sub_groups)) +
geom_point(size = 0.3) +
theme_void() +
scale_color_manual(values = group_color) +
theme(legend.position = "none") +
geom_text_repel(data = umap_label_frame_sub,
mapping = aes(x = x, y = y, label = sub_groups),
color = "black",size = 1.5,
max.overlaps = 100000)
pdf(glue("{out_dir}/figure1B2.pdf"), width = 3.5*0.4, height = 3*0.4)
print(a)
d <- dev.off()
a
figure1C
boxplotdf <- glue("{in_dir}/boxplot_twist2.rds") %>% read_rds
group_color <- glue("{in_dir}/group_color.rds") %>% read_rds
boxplotp <- boxplotdf %>%
ggplot(aes(x = sub_groups, y = TWIST2, fill = sub_groups)) +
geom_boxplot(outlier.size = 0.5) +
scale_fill_manual(values = group_color) +
theme_few() +
ylab(label = "Expression level of TWIST2") +
theme(
axis.text.y = element_text(size = 5,colour = "black"),
axis.text.x = element_text(size = 8, angle = 90, hjust = 1, vjust = 0.5,colour = "black"),
axis.ticks = element_line(linewidth = 0.8,colour = "black"),
axis.ticks.length = unit(0.05, "lines"),
axis.title.x = element_blank(),
axis.title.y = element_text(size = 8),
legend.position = "none",
panel.border = element_rect(colour = "black", linewidth = 1)
)
pdf(glue("{out_dir}/figure1C.pdf"), width = 5, height = 2.5)
print(boxplotp)
d <- dev.off()
boxplotp
figure1D
full_rankframe <- glue("{in_dir}/generank_twist2.rds") %>% read_rds
group_color <- glue("{in_dir}/group_color.rds") %>% read_rds
sub_rankframe <- full_rankframe %>% dplyr::filter(rank <= 5)
rank_plot <- ggplot(full_rankframe, aes(x = rank, y = logFC)) +
geom_point(size = 0.2)+
theme_few() +
ylab(label = "logFC") +
theme(
axis.text.y = element_text(size = 8,colour = "black"),
axis.text.x = element_blank(),
axis.ticks.y = element_line(linewidth = 0.8,colour = "black"),
axis.ticks.x = element_blank(),
axis.ticks.length = unit(0.05, "lines"),
axis.title.x = element_blank(),
axis.title.y = element_text(size = 8),
legend.position = "none",
panel.border = element_rect(colour = "black", linewidth = 1)
) +
geom_text_repel(data = sub_rankframe,
mapping = aes(label = gene),
color = "black",size = 2,
max.overlaps = 10)
pdf(glue("{out_dir}/figure1D.pdf"), width = 1.7, height = 2.5)
print(rank_plot)
d <- dev.off()
rank_plot