how many geneIDsets can be plot at maximum using gseaplot2?

[dksldjlskdfjlsdkf]

deg<- FindMarkers(
integrated_indAll,
ident.1 = "Dedifferentiated cells",
layer = "data", # In Seurat 5.0
test.use = "wilcox",
min.pct = 0.1,
logfc.threshold = 0.25,
only.pos = FALSE
)
deg <- data.frame(gene = rownames(deg), deg)#add gene names to deg table

deg1 <- deg
k1 = (deg1$p_val_adj < 0.05)&(deg1$avg_log2FC < -0.5)
k2 = (deg1$p_val_adj < 0.05)&(deg1$avg_log2FC > 0.5)

change = ifelse(k1,"down",ifelse(k2,"up","stable"))#If k1 is TRUE → "down",Else if k2 is TRUE → "up";Else→"stable"
deg1$change <- change

#Gene name conversion
s2e <- bitr(deg1$gene,
fromType = "SYMBOL",
toType = "ENTREZID",
OrgDb = org.Hs.eg.db) #Human, convert to ENTREZID

deg1 <- inner_join(deg1,s2e,by=c("gene"="SYMBOL"))
geneList = deg1$avg_log2FC
names(geneList) = deg1$ENTREZID
geneList = sort(geneList,decreasing = T)

gsea-KEGG enrichment analysis

kk_gse <- gseKEGG( #gseKEGG() Find KEGG pathways enriched across the entire ranked gene list
geneList = geneList,
organism = "hsa",
keyType = 'kegg',
minGSSize = 10,
pvalueCutoff = 0.05,
verbose = FALSE #The details are not displayed at runtime
)

kk_gse <- DOSE::setReadable(kk_gse, OrgDb='org.Hs.eg.db',keyType='ENTREZID') #Gene ID conversion
sortkk <- kk_gse[order(kk_gse$enrichmentScore, decreasing = T),]
gseaplot2(kk_gse, row.names(sortkk)[1:10])
the last row works if it's less than or equals 10, but cannot be 30. What is the maximum number that it can plot? Thank you very much.

[guidohooiveld]

In principle there is no limit!

Yet, if you print your object to the screen, and the size of the plotting window in R is (too) small, not everything can be plotted. So it seems not all sets are plotted, but actually they are! You can check this if you assign your plot to an object, and then export it using e.g. ggsave().

Having said this, you could wonder how to interpret the information on such information dense plot, but that is up to you! 😏

> ## load libraries
> library(clusterProfiler)
> library(enrichplot)
> library(org.Hs.eg.db)
> library(ggplot2)  ## for function ggsave
> 
> 
> ## reproducible example
> data(geneList, package="DOSE")
> 
> res <- gseGO(geneList      = geneList,
+              OrgDb         = org.Hs.eg.db,
+              keyType       = "ENTREZID",
+              ont           = "CC",
+              minGSSize     = 15,
+              maxGSSize     = 100,
+              eps           = 0,
+              pvalueCutoff  = 0.05,
+              pAdjustMethod = "BH")
using 'fgsea' for GSEA analysis, please cite Korotkevich et al (2019).

preparing geneSet collections...
GSEA analysis...
leading edge analysis...
done...
> 
> res <- pairwise_termsim(res)
> res <- setReadable(res, OrgDb = org.Hs.eg.db, keyType="ENTREZID")
> 
> ## using the last 2 lines of your code on 'res'
> sortkk <- res[order(res$enrichmentScore, decreasing = T),]
> 
> ## check: 42 sets should be plotted
> dim(sortkk)
[1] 42 11
> 
> ## assing plot to object
> plot.all <- gseaplot2(res, row.names(sortkk)[1 : dim(sortkk)[1] ] )
> 
> ## when plotting output to screen,
> ## it seems not all data is plotted...
> print(plot.all)
> 

> ## yet, when exported/saved plot as png,
> ## all data is there!
> ## note large values for width and height,
> ## to allow printing all information
> 
> ggsave(plot.all, file="./gseaplot2.all.sets.png", width = 30, height = 30, units = "in")
 >

[dksldjlskdfjlsdkf]

thank you so much!