-
Notifications
You must be signed in to change notification settings - Fork 0
/
Figure3_SCENIC.Rmd
417 lines (367 loc) · 18.6 KB
/
Figure3_SCENIC.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
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
---
title: "Figure3_SCENIC"
output: html_document
date: "2023-07-05"
---
```{r}
library(knitr)
library(dplyr)
library(tidyverse)
library(purrr)
library(xlsx)
library(Seurat)
library(igraph)
library(ggraph)
library(graphlayouts)
library(VennDiagram)
library(ggplot2)
library(msigdbr)
library(AnnotationDbi)
library(clusterProfiler)
library(org.Hs.eg.db)
```
```{r}
out_dir <- "/g/scb2/zaugg/amathiou/2020Dec_scBM_Tcells/6.ManuscriptFigures/Figure3/"
setwd(out_dir)
seurat_integrated <- readRDS("/g/scb2/zaugg/amathiou/2020Dec_scBM_Tcells/3.Downtream_demultiplexed/20220816_merged_SCT_xyexcluded_CompleteAnnot.rds")
input_dir <- "/g/scb2/zaugg/amathiou/2020Dec_scBM_Tcells/3.Downtream_demultiplexed/Final_20210826/DiffExp/wo_label/Pseudocount_0.001/"
#SCENIC was ran using an inhouse Snakemake pipeline; see Supplemental methods
scenic_df_wide <- read.csv("/g/scb2/zaugg/amathiou/2020Dec_scBM_Tcells/4.SCENIC/output/new_aucell_mtx.tsv",
sep = "\t",
row.names = "Cell")
```
```{r}
# SCENIC output-target genes are assigned to each TF
#Note: This is what we use for network re-construction
reg_files <- list.files("/g/scb2/zaugg/amathiou/2020Dec_scBM_Tcells/4.SCENIC/output/regulon",
pattern = ".*\\(\\+\\)\\.tsv$",
full.names = T)
df_list <- list()
for (file in reg_files) {
# the regex matches any characters except "/" that are right before a "(+).tsv" and thereby fetches the TF-names
TF_name <- str_extract(file, "[^\\/]+(?=\\(\\+\\)\\.tsv)")
regulon_df <- read.csv(file, sep = "\t", header = F, col.names = c("target", "count"))
regulon_df <- mutate(regulon_df, TF = TF_name)
df_list[[TF_name]] <- regulon_df
}
# targene_df_raw contains all target genes for the TFs, even the ones with
# counts < 80 % that were not used for the activity calculation
empty_df <- data.frame(TF = character(), target = character(), count = numeric())
targene_df_raw <- reduce(df_list, bind_rows, .init = empty_df)
# make another copy with only the target genes that were used for the activity calculation (observed in >80% of runs; set in min-regulon-gene-occurrence in config.yaml
targene_df <- filter(targene_df_raw, count > 40) #n = 50
targene_df <- readRDS("/g/scb2/zaugg/amathiou/2020Dec_scBM_Tcells/4.SCENIC/202203_GRN.rds")
```
```{r}
DE_genes <- list()
for (p in c("CD4", "CD8", "HSPCs")){
tmp_de <- read.table(paste0("/g/scb2/zaugg/amathiou/2020Dec_scBM_Tcells/4.DiffExpr/202208_SCT/for_scenic/", p, "_DEgenes_MAST_RNA_relVScr_202208.tsv"))
print(tmp_de)
DE_genes[[paste0(p, " CR")]] <- setdiff(rownames(tmp_de[which(tmp_de$avg_log2FC < -1 & tmp_de$p_val_adj < 0.0005), ]), noise_genes)
DE_genes[[paste0(p, " REL")]] <- setdiff(rownames(tmp_de[which(tmp_de$avg_log2FC > 1 & tmp_de$p_val_adj < 0.0005), ]), noise_genes)
}
pop_de <- list()
pop_de[["CD4"]] <- unique(unlist(DE_genes[c("CD4 CR", "CD4 REL")]))
pop_de[["CD8"]] <- unique(unlist(DE_genes[c("CD8 CR", "CD8 REL")]))
pop_de[["HSPCs"]] <- unique(unlist(DE_genes[c("HSPCs CR", "HSPCs REL")]))
```
```{r}
sig_TFs <- list()
fisher_df1 <- list()
fisher_df2 <- list()
for (p in c("CD4", "CD8", "HSPCs")){
tmp_genes <- intersect(pop_de[[p]], unique(targene_df$target)) #Take those that are in the network
tmp_targene_df <- targene_df[which(targene_df$target %in% tmp_genes), ] #Lets look only at all the DE genes, both conditions
cr_genes <- intersect(tmp_genes, DE_genes[[paste0(p, " CR")]])
rel_genes <- intersect(tmp_genes, DE_genes[[paste0(p, " REL")]])
count_tfs_genes <- data.frame(TF = unique(tmp_targene_df[which(tmp_targene_df$target %in% tmp_genes), "TF"])) #how many of the condition genes these TFs target
rownames(count_tfs_genes) <- count_tfs_genes$TF
table_tfs <- data.frame() #Fisher test input
for (t in rownames(count_tfs_genes)){
#A bit repeatitive with the cont table, but i like the visualization of this table
tf_targets <- tmp_targene_df[which(tmp_targene_df$TF == t & tmp_targene_df$target %in% tmp_genes), "target"]
count_tfs_genes[t, "CR_TF"] <- length(intersect(tf_targets, cr_genes)) + 1
count_tfs_genes[t, "REL_TF"] <- length(intersect(tf_targets, rel_genes)) + 1
count_tfs_genes[t, "CR_rest"] <- length(setdiff(cr_genes, tf_targets)) + 1
count_tfs_genes[t, "REL_rest"] <- length(setdiff(rel_genes, tf_targets)) + 1
#Contigency table
cont_table <- data.frame()
cont_table['CR', t] <- count_tfs_genes[t, "CR_TF"]
cont_table['CR', 'rest'] <- count_tfs_genes[t, "CR_rest"]
cont_table['REL', t] <- count_tfs_genes[t, "REL_TF"]
cont_table['REL', 'rest'] <- count_tfs_genes[t, "REL_rest"]
table_tfs[t, 'TF'] <- t
table_tfs[t, 'Fisher_pvalue'] <- fisher.test(cont_table, alternative='two.sided', conf.int = TRUE)$p.value
table_tfs[t, 'OR'] <- fisher.test(cont_table, alternative='two.sided', conf.int = TRUE)$estimate
}
table_tfs$fdr <- p.adjust(table_tfs$Fisher_pvalue, method="fdr")
table_tfs$padj <- ifelse(table_tfs$fdr < 0.05,
"<0.05",
"n.s.")
#print(table_tfs)
table_tfs$log2OR <- log2(table_tfs$OR)
table_tfs <- table_tfs[order(table_tfs$log2OR), ]
levels_plot <- rownames(table_tfs)
#Supplemental Table 4
#write.table(table_tfs, file = paste0("/g/scb2/zaugg/amathiou/2020Dec_scBM_Tcells/4.SCENIC/2022_SubGenesForFisher/SuppFig/FisherTF_targets_", p, ".tsv"))
write.xlsx(tmp_de, paste0("/g/scb2/zaugg/amathiou/2020Dec_scBM_Tcells/4.SCENIC/2022_SubGenesForFisher/SuppFig/FisherTF_targets.xlsx"), sheetName = p, col.names = TRUE, row.names = TRUE, append = TRUE)
q3 <- ggplot(table_tfs, aes(x = factor(TF, levels = levels_plot), y = log2OR, fill = padj)) +
geom_bar(stat = "identity") +
theme_bw() +
theme(text = element_text(size = 15),
axis.text.x = element_text(angle = 90, vjust = 0.5, hjust = 1),
legend.position = "right") + xlab("TFs") +
ggtitle(p) +
scale_fill_manual("padj", values = c("n.s." = "#BF812D", "<0.05" = "#35978F")) # + scale_fill_gradientn(colors = head(viridis(10), 9)
#For Supplemental Figure 7
pdf(paste0("/g/scb2/zaugg/amathiou/2020Dec_scBM_Tcells/4.SCENIC/2022_SubGenesForFisher/SuppFig/FisherTF_targets_", p, "_3.pdf"), width = 15, height = 4)
plot(q3)
dev.off()
sig_TFs[[paste0(p, " CR")]] <- table_tfs[which(table_tfs$padj < 0.05 & table_tfs$log2OR > 0), "TF"]
sig_TFs[[paste0(p, " REL")]] <- table_tfs[which(table_tfs$padj < 0.05 & table_tfs$log2OR < 0), "TF"]
}
```
```{r}
#For Supplemental Figure 7
pdf("/g/scb2/zaugg/amathiou/2020Dec_scBM_Tcells/4.SCENIC/2022_SubGenesForFisher/SuppFig/Upset_DATFs.pdf", width = 5, height = 4)
upset(fromList(sig_TFs), nsets = 15, order.by = "freq", point.size = 1.5)
dev.off()
```
```{r}
plot_tfs <- unique(unlist(sig_TFs))
cd34_or <- read.table(paste0("/g/scb2/zaugg/amathiou/2020Dec_scBM_Tcells/4.SCENIC/2022_SubGenesForFisher/SuppFig/FisherTF_targets_HSPCs.tsv"))
cd4_or <- read.table(paste0("/g/scb2/zaugg/amathiou/2020Dec_scBM_Tcells/4.SCENIC/2022_SubGenesForFisher/SuppFig/FisherTF_targets_CD4.tsv"))
cd8_or <- read.table(paste0("/g/scb2/zaugg/amathiou/2020Dec_scBM_Tcells/4.SCENIC/2022_SubGenesForFisher/SuppFig/FisherTF_targets_CD8.tsv"))
#tile plot but with log2fc
or_df <- data.frame(Gene = plot_tfs,
CD8 = NA,
CD4 = NA,
HSPCs = NA)
rownames(or_df) <- or_df$Gene
or_df$Gene <- NULL
```
```{r}
for (g in rownames(or_df)){
if (g %in% c(sig_TFs$`HSPCs CR`, sig_TFs$`HSPCs REL`)){
or_df[g, "HSPCs"] <- -log2(cd34_or[g, "OR"])
}
if(g %in% c(sig_TFs$`CD4 CR`, sig_TFs$`CD4 REL`)){
or_df[g, "CD4"] <- -log2(cd4_or[g, "OR"])
}
if(g %in% c(sig_TFs$`CD8 CR`, sig_TFs$`CD8 REL`)){
or_df[g, "CD8"] <- -log2(cd8_or[g, "OR"])
}
}
levels <- rownames(or_df)
levels <- c("SPI1", "STAT1", "FOSB", "JUN", "ETS1", "TAL1", "GATA1", "KLF1", "GATA2", "TBX21", "NFKB1", "RELB", "ELF1", "NFKB2", "MAFF",
"IRF1", "JUNB", "JUND", "NFE2L2", "FOS", "REL", "FOSL2", "CREM")
or_df$gene <- rownames(or_df)
or_df <- gather(or_df, "Cell", "OR", 1:3)
```
```{r, fig.width=3, fig.height=4}
pdf("/g/scb2/zaugg/amathiou/2020Dec_scBM_Tcells/4.SCENIC/2022_SubGenesForFisher/202209_Fisher_TF_fdr0.05.pdf",
height = 8,
width = 5)
ggplot(or_df, aes(x = factor(Cell, levels = c("HSPCs", "CD8", "CD4")),
y = factor(gene, levels = levels))) +
geom_tile(data = subset(or_df, !is.na(OR)), aes(fill = OR), color = "white",
lwd = 0.4,
linetype = 1) +
theme_classic2() +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1)) +
scale_fill_gradientn(colours = rev(c("#B2DF8A", "white", "#1F78B4"))) + xlab("") + ylab("TF")
dev.off()
#And some target gene numbers
table(targene_df[which(targene_df$TF %in% unique(unlist(sig_TFs)) &
targene_df$target %in% unique(unlist(pop_de))), c("TF")])
```
```{r}
target_genes_list <- list()
for (i in levels){
tmp_target_genes <- unique(targene_df[which(targene_df$TF == i & targene_df$target %in% unlist(DE_genes)), c("target")])
target_genes_list[[i]] <- tmp_target_genes
}
pdf("/g/scb2/zaugg/amathiou/2020Dec_scBM_Tcells/4.SCENIC/2022_SubGenesForFisher/SuppFig/Upset_TargetsPerTF.pdf", width = 10, height = 8)
upset(fromList(target_genes_list), nsets = 30)
dev.off()
```
```{r}
#For Supplemental Figure 8
#Venn diagram for shared genes across shared tfs
display_venn <- function(x, ...){
grid.newpage()
venn_object <- venn.diagram(x, filename = NULL, lwd = 2, lty = 'blank',
fill = head(myCol, 2), cex = 1, fontface = "bold", fontfamily = "sans", cat.cex = 1)
pdf(paste0("/g/scb2/zaugg/amathiou/2020Dec_scBM_Tcells/4.SCENIC/2022_SubGenesForFisher/TFoverlap/", i, "_venn.pdf"), width = 3, height = 3)
grid.draw(venn_object)
dev.off()
}
myCol <- brewer.paired(10)[6:8]
for (i in levels){
sig_genes_per_tf <- list()
tmp_CD4 <- unique(targene_df[which(targene_df$TF == i &
targene_df$target %in% unique(unlist(c(pop_de$CD4)))), c("target")])
tmp_CD8 <- unique(targene_df[which(targene_df$TF == i &
targene_df$target %in% unique(unlist(pop_de$CD8))), c("target")])
sig_genes_per_tf[["CD4"]] <- tmp_CD4
sig_genes_per_tf[["CD8"]] <- tmp_CD8
v <- display_venn(sig_genes_per_tf)
}
```
```{r, fig.height=5.5, fig.width=7}
#Hallmark analysis for the 'functional annotation' of target genes (For Figure 3)
tmp_bckgrnd <- as.data.frame(rowSums(seurat_integrated@assays$RNA@data))
colnames(tmp_bckgrnd) <- "Expression"
tmp_bckgrnd$gene <- rownames(tmp_bckgrnd)
background <- tmp_bckgrnd[which(tmp_bckgrnd$Expression >0), "gene"]
#DEgenes
disease_genes <- unique(unlist(pop_de))
go_enrich <- enrichGO(gene = disease_genes,
OrgDb = 'org.Hs.eg.db',
keyType = "SYMBOL",
ont = "BP",
universe = background)
plot(dotplot(go_enrich, showCategory = 30, title = "BP", font.size=12))
go_df <- as.data.frame(go_enrich)
```
```{r}
hallmark_gene_sets = msigdbr(species = "Homo sapiens", category = "H")
hallmark_t2g <- hallmark_gene_sets %>% dplyr::distinct(gs_name, gene_symbol) %>% as.data.frame()
hallmark_terms <- enricher(gene = disease_genes, TERM2GENE = hallmark_t2g)
hallmark_terms_df <- as.data.frame(hallmark_terms)
hallmark_terms_df$Hallmark <- str_replace_all(hallmark_terms_df$ID, "HALLMARK_", "")
ggplot(hallmark_terms_df) + geom_point(aes(x = GeneRatio, y = factor(Hallmark,
levels = hallmark_terms_df$Hallmark[order(GeneRatio)]),
color = -log10(pvalue), size = GeneRatio)) +
theme_minimal() +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1)) +
xlab("Gene Ratio") +
ylab("Hallmark terms") +
scale_colour_gradientn(colors = viridis(10))
```
```{r}
tnf_genes <- unique(unlist(strsplit(hallmark_terms_df[which(hallmark_terms_df$Description == "HALLMARK_TNFA_SIGNALING_VIA_NFKB"), "geneID"], split = "/")))
ifn_genes <- unique(unlist(strsplit(hallmark_terms_df[which(hallmark_terms_df$Description %in%
c("HALLMARK_INTERFERON_GAMMA_RESPONSE", "HALLMARK_INTERFERON_ALPHA_RESPONSE")), "geneID"], split = "/")))
Immune_activation <- unique(unlist(strsplit(go_df[which(go_df$Description %in%
c("cell activation involved in immune response", "regulation of immune effector process")), "geneID"],
split = "/")))
```
```{r}
tnf_final <- tnf_genes
ifn_final <- setdiff(ifn_genes, c(tnf_final))
activation_final <- setdiff(Immune_activation, c(tnf_final, ifn_final))
```
```{r}
disease_genes <- list(ifn_final, tnf_final, activation_final)
names(disease_genes) <- c("IFN", "TNF", "Activation")
plot_df <- targene_df[which(targene_df$TF %in% plot_tfs &
targene_df$target %in% unique(c(unique(unlist(c(DE_genes$`CD4 CR`, DE_genes$`CD8 CR`,
DE_genes$`CD4 REL`, DE_genes$`CD8 REL`))), plot_tfs))), ]
```
```{r}
#Some numbers
TF_counts <- as.data.frame(table(plot_df$TF))
colnames(TF_counts) <- c("TF", "Freq")
TF_counts <- TF_counts[rev(order(TF_counts$Freq)),]
sig_TFs <- TF_counts[which(TF_counts$Freq > 0), ]$TF
plot_df2 <- plot_df[, c("TF", "target")]
colnames(plot_df2) <- c("source", "target")
plot_df2 <- plot_df2[which(plot_df2$source %in% sig_TFs), ]
plot_df2 <- mutate(plot_df2, goterm = case_when(target %in% ifn_final ~ "IFN",
target %in% tnf_final ~ "TNF",
target %in% activation_final ~ "Activation"))
plot_df2$goterm <- ifelse(is.na(plot_df2$goterm),
"Others",
plot_df2$goterm)
plot_df2$goterm <- ifelse(plot_df2$target %in% c("HSPA1A", "HSPA1B"),
"Others",
plot_df2$goterm)
terms_counts <- as.data.frame(table(distinct(plot_df2[, c("target", "goterm")])$goterm))
colnames(terms_counts) <- c("TF", "Freq")
plot_df2 <- as.data.frame(table(plot_df2[, c("source", "goterm")]))
colnames(plot_df2) <- c("source", "target", "Freq")
```
```{r, fig.width=10, fig.height=3.5}
levels_TFs <- rev(levels)
cols <- c("lightgrey", "#D8C1DE", "#fdbf6f", "#fb9a99")
pdf("/g/scb2/zaugg/amathiou/2020Dec_scBM_Tcells/4.SCENIC/2022_SubGenesForFisher/202209_tf_goterms.pdf", width = 10, height = 3)
ggplot(plot_df2, aes(x = factor(source, levels = levels_TFs),
fill = factor(target, levels = c("Others", "Activation", "IFN", "TNF")),
y = Freq)) + geom_bar(stat = "identity", position = "fill") + theme_classic() + theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1)) + scale_y_continuous(labels = scales::percent) + scale_fill_manual("Term", values = cols)
dev.off()
```
```{r}
#Network visualization
cd4_all_genes <- unique(unlist(c(DE_genes[["CD4 REL"]], DE_genes[["CD4 CR"]])))
cd8_all_genes <- unique(unlist(c(DE_genes[["CD8 REL"]], DE_genes[["CD8 CR"]])))
shared_cd4_cd8 <- intersect(cd4_all_genes, cd8_all_genes)
```
```{r}
#For Figure 3 network
plot_df <- targene_df[which(targene_df$TF %in% c("TBX21", "REL", "FOS") &
targene_df$target %in% unique(cd8_all_genes)), ]
TF_counts <- as.data.frame(table(plot_df$TF))
colnames(TF_counts) <- c("TF", "Freq")
plot_df <- plot_df[, c("TF", "target")]
colnames(plot_df) <- c("source", "target")
#write.table(plot_df, paste0("/g/scb2/zaugg/amathiou/2020Dec_scBM_Tcells/4.SCENIC/", p, "network.csv"), sep = ",", quote = F, row.names = F)
network <- plot_df %>% graph_from_data_frame(directed = F)
#Label only TFs
V(network)$name <- ifelse(V(network)$name %in% unique(plot_df$source),
V(network)$name,
V(network)$name)
V(network)$label<- ifelse(V(network)$name %in% unique(plot_df$source),
"TF",
ifelse(V(network)$name %in% unique(unlist(c(DE_genes[["CD8 REL"]]))),
"Rel",
"CR"))
V(network)$size <- unlist(lapply(V(network)$name, function(x) ifelse(x %in% plot_df$source,
TF_counts[which(TF_counts$TF == x), "Freq"],
14)))
p <- network %>%
ggraph(layout = "stress") +
geom_edge_link(alpha = .5, color = "darkgrey") +
geom_node_point(aes(colour = label, size = size)) +
geom_node_text(aes(label = name, colour = label), repel = TRUE, check_overlap = T) +
theme_graph() +
scale_color_manual(values = c("#1F78B4", "#33A02C", "#FF7F00")) +
theme(legend.position="none")
pdf("/g/scb2/zaugg/amathiou/2020Dec_scBM_Tcells/4.SCENIC/202209_cd8_reduced_network_fig3.pdf", width = 8, height = 5)
plot(p)
dev.off()
```
```{r}
#For Supplemental Figure 8
hspcs_all_genes <- unique(unlist(c(DE_genes[["HSPCs REL"]], DE_genes[["HSPCs CR"]])))
plot_df <- targene_df[which(targene_df$TF %in% c("GATA2", "GATA1", "KLF1", "TAL1", "JUN", "FOSB", "FOSL2", "JUNB") &
targene_df$target %in% unique(hspcs_all_genes)), ]
TF_counts <- as.data.frame(table(plot_df$TF))
colnames(TF_counts) <- c("TF", "Freq")
plot_df <- plot_df[, c("TF", "target")]
colnames(plot_df) <- c("source", "target")
network <- plot_df %>% graph_from_data_frame(directed = F)
#Label only TFs
V(network)$name <- ifelse(V(network)$name %in% unique(plot_df$source),
V(network)$name,
V(network)$name)
V(network)$label<- ifelse(V(network)$name %in% unique(plot_df$source),
"TF",
ifelse(V(network)$name %in% unique(unlist(c(DE_genes[["HSPCs REL"]]))),
"Rel",
"CR"))
V(network)$size <- unlist(lapply(V(network)$name, function(x) ifelse(x %in% plot_df$source,
TF_counts[which(TF_counts$TF == x), "Freq"],
14)))
p <- network %>%
ggraph(layout = "stress") +
geom_edge_link(alpha = .5, color = "darkgrey") +
geom_node_point(aes(colour = label, size = size)) +
geom_node_text(aes(label = name, colour = label), repel = TRUE, check_overlap = T) +
theme_graph() +
scale_color_manual(values = c("#1F78B4", "#33A02C", "#FF7F00")) +
theme(legend.position="none")
pdf("/g/scb2/zaugg/amathiou/2020Dec_scBM_Tcells/4.SCENIC/202209_hspcs_reduced_network_supfig8.pdf", width = 8, height = 5)
plot(p)
dev.off()
```