Help with R code to get a unique list

Question

[Space_Life]

Hi,

I have gene ontology data that looks like this (sample data). I manually downloaded the gene ontology data from the Uniprot website.

I want to:

• Make a list of unique GO terms from the column 'Gene Ontology(molecular function)
• Make the list of genes associated with each GO term
• Also make the list of GO term descriptions in the next column
• Make the list of number of gene associated to a GO term

The output data would look like this

Please note that GO terms do not repeat in a specific cell, however, they repeat in other rows. Kindly help me to do this in R.

I did several unsuccessful trials. Hope to get help from someone. I will highly appreciate your help. Thank you so much.

[fracarb8]

df[!duplicated(df$'Gene Ontology (molecular function)'),] will work, and it will only keep the first entry of the duplicated values.

[cpad0112]

group by GO IDs, collapse gene_id column, expand GO IDs, group by GO IDs, and count genes.

[zx8754]

See this SO post as a start:

• Split delimited strings in a column and insert as new rows

Answer 1

[rpolicastro]

Tidyverse answer

library("tidyverse")

df %>%
  select(gene_id, `Gene ontology (molecular function)`) %>%
  distinct %>%
  group_by(`Gene ontology (molecular function)`) %>%
  summarize(gene_id=str_c(gene_id, collapse=", "), `No. of Genes`=n(), .groups="drop") %>%
  separate_rows(`Gene ontology (molecular function)`, sep="; ") %>%
  separate(
    `Gene ontology (molecular function)`, sep="\\s(?=\\[GO)",
    into=c("Gene ontology (molecular function)", "Gene Ontology IDs"))

[ponganta]

I'd strongly suggest making gene_id a list-column of vectors though. Else this ain't tidy.

[Space_Life]

Thank you so much for this code. I tried it. It works wonderfully. Some of the GO terms are repeating (present in multiple rows_ in the output table (GO:0003735,GO:0019843). Also, can we add corresponding UniprotKBs in one more column? Thank you again.

[Space_Life]

Also, I tried exporting it as CSV. It gives out the same file. How do you export the updated dataset as CSV? Thank you again for your kind help.

[rpolicastro]

You need to save the results to a variable, such as changing the first line to df2 <- df %>%, and then you can save df2 to a csv.

[Space_Life]

Thank you so much. It worked.