ggplot low pvalues not detected

Question

[kra277]

Hi,

I want to use GAGE GO results and GGplot to output a dot plot. I tried it using the following code.

graph_ke_test <- 
      gage_ke_test %>% 
      subset(p.val < 0.05) %>% 
      head(5) %>% 
      ggplot(aes(x= set.size, y= Kegg_pathway)) + 
      geom_point(aes(color = p.val), size = 5) +
      theme_bw()+ labs(title = "GAGE KEGG 2 Directional Top 5 Genesets", 
      x = "Number of Differentially Expressed Genes", y = NULL, color = "p value")

The p-value scale is not representative of the values of GAGE GO. In the GAGE GO Table, the pvalue is 2.595794e-14. The scale in ggplot ranges from 0.005 to 0.02.

Below is the short version of the table

1 hsa03010 Ribosome 2.595794e-14

2 hsa03008 Ribosome biogenesis in eukaryotes 1.435528e-02

3 hsa03018 RNA degradation 1.457412e-02

4 hsa00190 Oxidative phosphorylation 1.887497e-02

5 hsa03013 RNA transport 2.044700e-02

I checked the structure of the data to make sure the value column is numeric, and it is. Did anyone come across this issue and know how to solve this? It would be very helpful to resolve this issue.

Thank you in advance

Answer 1

[Sam]

You used head(5)

which only show the first 5 pathway. You might want to remove that if you want to plot all pathways. Or you might want to sort the results by p-value before you plot so that you are plotting the 5 most significant pathways

Answer 2

[Alex Nesmelov]

It may be just a problem with p.val visualization via color. You can see the real plotted value of p.val, plotting it as factor:

    geom_point(aes(color = factor(p.val)), size = 5) +