CODEWITHSUNDEEP
r
beginner
beginner

Reputation: 411

How to remove character with specific pattern form data frame in R

Im trying to remove the all the characters starting with the pattern "Gm" from last column of my data.frame

My data.frame looks like this

level   logp  chr   start    end     CNA      Genes                                                           
  3     1.4   3      100     110    gain     Gm5852,Gm5773,Tdpoz4,Tdpoz3,Gm911                      
  4     18.10 3      962      966   gain     Fcgr1,Terc,Gm5703

The result should look something like this

level   logp     chr   start    end     CNA           Genes                                                           
   3     1.4      3      100     110    gain    Tdpoz4,Tdpoz3                   
   4     18.10    3      962     966    gain    Fcgr1,Terc

Upvotes: 3

Views: 896

Answers (2)

G. Grothendieck
G. Grothendieck

Reputation: 270248

This uses a single gsub to remove the unwanted portions:

Genes <- c("Gm5852,Gm5773,Tdpoz4,Tdpoz3,Gm911", "Fcgr1,Terc,Gm5703") # test data
gsub(",?Gm[^,]*,?", "", Genes)

giving:

[1] "Tdpoz4,Tdpoz3" "Fcgr1,Terc"

Here is a visualization of the regular expression:

,?Gm[^,]*,?

Regular expression visualization

Debuggex Demo

Upvotes: 5

Marat Talipov
Marat Talipov

Reputation: 13314

Given a data frame d:

d$Genes_new <- sapply(strsplit(as.character(d$Genes),split=','),function(s) paste(s[!grepl('^Gm',s)],collapse=','))

#  level logp chr start end  CNA                             Genes     Genes_new
#1     3  1.4   3   100 110 gain Gm5852,Gm5773,Tdpoz4,Tdpoz3,Gm911 Tdpoz4,Tdpoz3
#2     4 18.1   3   962 966 gain                 Fcgr1,Terc,Gm5703    Fcgr1,Terc

Here, strsplit(as.character(d$Genes),split=',') creates a list of comma-separated gene names for each row, and sapply applies to each element of this list a function that excludes all gene names starting from Gm (s[!grepl('^Gm',s)]) and concatenates the remaining genes (paste(.,collapse=',').

Upvotes: 2

Related Questions