Issue with Emapplot category node size increase with CompareCluster results

[KR-15]

Hi,

I would like to follow-up on this issue #304.

I am currently using enrichplot v1.27.4.

The size node are now proportional to number of gene present inside (pie = "Count" is now working well) but Is it possible to increase proportionally the size of the node ? On my side, changing the size_category parameter does not improve/change anything.

Reproducible example:

data(gcSample)

GO <- compareCluster(geneCluster = gcSample, 
                                  fun = enrichGO,
                                  ont = "MF",
                                  OrgDb ="org.Hs.eg.db",
                                  pAdjustMethod = "BH",
                                  keyType = "ENTREZID",
                                  pvalueCutoff  = 0.05,
                                  qvalueCutoff  = 0.05,
                                  minGSSize = 10,
                                  maxGSSize = 500,
                                  readable=TRUE)

GO_sim <- pairwise_termsim(GO) 
 
emapplot(GO_sim, pie="Count", 
         size_edge = 1.5,
         size_category = 0.5,
         showCategory = 8)

emapplot(GO_sim, pie="Count", 
         size_edge = 1.5,
         size_category = 1.0,
         showCategory = 8)

emapplot_size_categoery_0.5:

emapplot_size_categoery_1.0:

Thank you,

KR

[guidohooiveld]

I can reproduce your issue. That is, changing the value of the argument size_category has no impact at all.

As far as I can see it happens at this line of the code:

enrichplot/R/emapplot.R

Line 89 in 2ee77c6

p <- add_node_pie(p, gg$data, pie, pie_scale=size_category)

Moreover, argument size_category also doesn't work with cnetplot on the same compareClusterResult object... ??

enrichplot/R/cnetplot.R

Line 114 in 2ee77c6

p <- add_node_pie(p, d, pie, pie_scale=size_category)

Both call the internal function add_node_pie.
https://github.com/YuLab-SMU/enrichplot/blob/2ee77c68af4717856a0fabe50302fc5c28bd722c/R/cnetplot.R#L123C1-L145C1

@GuangchuangYu : could you please have a look at this? Thanks!