CODEWITHSUNDEEP
r
kleurless
kleurless

Reputation: 5

change rownames with a dataframe that holds the old and the new rowname

I've got a dataframe (over 9000 rows) that holds the average gene expression of certain genes (rows) per cellcluster (columns) and now I need to change the gene names (rownames) to the orthologs. It looks like this:

          Cluster1   Cluster2  Cluster3
[Tppp2]   10.32      0.14      2.56
[Mtx1]    6.32       8.77      0.30
[Vps37c]  225.02     132.87    9.52
[Slc39a9] 52.13      18.42     4.12

And I have another dataframe (over 13000 rows) that holds the orthologs; gene name as stated (old) - ortholog gene name (new). It looks like this:

    GeneName    NewGeneName
[1] Vps37c      VPS37C
[2] Tppp2       TPPP3
[3] Slc39a9     SLC39A9
[4] Mtx1        MTX1B

So for each of the rows in the first dataframe, the rownames will be matched with the name in dataframe2$GeneName and then the name in dataframe2$NewGeneName will be saved as the new rowname (or in a vector that holds all the new rownames in the correct order). For instance Tppp2 will be TPPP3, and so on to result in the following:

rownames(expr_df) <- c("TPPP3", "MTX1B", "VPS37C", "SLC39A9".

I have tried a lot of things, and its really bugging me that I can't make it work. I don't remember all the ways that I have tried and failed, sorry.

FYI: The orthologs and the rownames of the expression dataframe do NOT have the same order, and the ortholog dataframe holds more genes than there are present in the expression dataframe.

Upvotes: 0

Views: 44

Answers (1)

Ahorn
Ahorn

Reputation: 3876

A tidyverse solution:

df1 %>% 
  rownames_to_column() %>% 
  left_join(df2, by = c("rowname" = "GeneName")) 

  rowname Cluster1 Cluster2 Cluster3 NewGeneName
1   Tppp2    10.32     0.14     2.56       TPPP3
2    Mtx1     6.32     8.77     0.30       MTX1B
3  Vps37c   225.02   132.87     9.52      VPS37C
4 Slc39a9    52.13    18.42     4.12     SLC39A9

Data

df1 <- tibble::tribble(
         ~Cluster1, ~Cluster2, ~Cluster3,
             10.32,      0.14,      2.56,
              6.32,      8.77,       0.3,
            225.02,    132.87,      9.52,
             52.13,     18.42,      4.12
         )

df1 <- as.data.frame(df1)

rownames(df1) <- c("Tppp2", "Mtx1", "Vps37c", "Slc39a9")

df2 <- tibble::tribble(
         ~GeneName, ~NewGeneName,
          "Vps37c",     "VPS37C",
           "Tppp2",      "TPPP3",
         "Slc39a9",    "SLC39A9",
            "Mtx1",      "MTX1B"
         )

Upvotes: 0

Related Questions