This is a test version of Biostars. For the public version, visit https://www.biostars.org.
GG Sankey plot - plot that does not show any flow between columns

Hi,

I am trying to create a Sankey plot that illustrates the relationship of some GO terms, and I have a dataset that looks like this:

structure(list(parent_id = c("GO:0015727", "GO:0035873", "GO:0035873", 
"GO:0015129", "GO:0035879"), parent_name = c("lactate transport", 
"lactate transmembrane transport", "lactate transmembrane transport", 
"lactate transmembrane transporter activity", "plasma membrane lactate transport"
), aspect_parent = c("biological_process", "biological_process", 
"biological_process", "molecular_function", "biological_process"
), id = c("GO:0035873", "GO:0015129", "GO:0035879", "GO:0015650", 
"GO:0046722"), name = c("lactate transmembrane transport", "lactate transmembrane transporter activity", 
"plasma membrane lactate transport", "lactate:proton symporter activity", 
"lactic acid secretion"), relation = c("is_a", "part_of", "is_a", 
"is_a", "is_a"), hasChildren = c("TRUE", "TRUE", "TRUE", "FALSE", 
"FALSE"), aspect = c("biological_process", "molecular_function", 
"biological_process", "molecular_function", "biological_process"
), recursive_pass = structure(c(1L, 2L, 2L, 3L, 3L), levels = c("first", 
"second", "third"), class = "factor"), next_pass = structure(c(1L, 
2L, 2L, NA, NA), levels = c("first", "second", "third"), class = "factor")), row.names = c(NA, 
-5L), class = "data.frame")

Now i have used the ggsankey R package to make a Sankey plot and the ordering of the data, everything looks fine to me but I get a plot that does not show any flow between columns, and I do not understand where my code or my data structure is falling.

Thank you !

The ggplot code:

pathways_connections <- ggplot(data = GOs_childs_recursive,
                               aes(x = recursive_pass, 
                                   next_x = next_pass, 
                                   node = parent_id, 
                                   next_node = parent_id,
                                   fill = factor(parent_id))) + 
  geom_sankey(flow_alpha = 1,
              node.color = "black",           
              show.legend = TRUE)
pathways_connections

The graph I get:

enter image description here

ggplot2 sankey-plot r go
• 2,718 views • link
updated by zx8754 zx8754 • written by ijarne • ijarne

Use dput to share the dataset, not to show what it looks like - use a simple print() to do that, like so:

   parent_id                                parent_name      aspect_parent         id                                       name relation hasChildren             aspect recursive_pass next_pass
1 GO:0015727                          lactate transport biological_process GO:0035873            lactate transmembrane transport     is_a        TRUE biological_process          first     first
2 GO:0035873            lactate transmembrane transport biological_process GO:0015129 lactate transmembrane transporter activity  part_of        TRUE molecular_function         second    second
3 GO:0035873            lactate transmembrane transport biological_process GO:0035879          plasma membrane lactate transport     is_a        TRUE biological_process         second    second
4 GO:0015129 lactate transmembrane transporter activity molecular_function GO:0015650          lactate:proton symporter activity     is_a       FALSE molecular_function          third      <NA>
5 GO:0035879          plasma membrane lactate transport biological_process GO:0046722                      lactic acid secretion     is_a       FALSE biological_process          third      <NA>
• 0 views • link
written by Ram Ram

1 answer

You need to change your data table into a 'long' format. You can do something like this-

GOs_childs_recursive <- GOs_childs_recursive %>% make_long(parent_id, id)

ggplot(data=GOs_childs_recursive, 
       aes(x=x, next_x=next_x, 
           node=node, 
           next_node=next_node, 
           fill=factor(node), 
           label=node, 
           color=factor(node)))+
  geom_sankey()

You will get something like below:

enter image description here

• 0 views • link
updated by Ram Ram • written by bk11 • bk11

bk11 i changed this comment to an answer. if you disagree please let me know. thanks for sharing your knowledge. VAL

• 0 views • link
written by LauferVA LauferVA

Hey I am sorry for the late response, thank you for the help ! I will post the latest version of the plots when finished as it is something I am not working on now ! But thanks !

• 0 views • link
written by ijarne • ijarne

We don't need your latest version, just whether or not bk11's post helped you. Also, use Comments, not Answers to reply to people.

• 0 views • link
written by Ram Ram

upvote_bookmark_accept

• 0 views • link
written by Ram Ram

Log in to answer this question.

More posts like this