This is a test version of Biostars. For the public version, visit https://www.biostars.org.
In R, how do I apply regexes to specific parts of a string that contains a pattern?

I have a dataframe (trip) that contains a column (SNP). It looks like this (but longer, and it has 192 levels):

SNP
C[T->C]T
C[G->C]A
G[A->C]A
C[T->C]C
C[C->A]G
T[G->A]C
...

I want to pattern match and replace on the following criteria:

gsub("G->T", "C->A", trip)
gsub("G->C", "C->G", trip)
gsub("G->A", "C->T", trip)
gsub("A->T", "T->A", trip)
gsub("A->G", "T->C", trip)
gsub("A->C", "T->G", trip)

but ALSO, if one of the patterns listed above is found, I want the string in which it's contained have additional substitutions applied. Namely:

if ((grep(G->T|G->C|G->C|A->T|A->G|A->C), trip$SNP)==TRUE){
   substr(trip$SNP, 1,1) <- tr /ATCG/TAGC/; #incompatible perl syntax?
   substr(trip$SNP, 8,8) <- tr /ATCG/TAGC/;
   }

As in, if any of these patterns--G->T, G->C, G->C, A->T, A->G, or A->C--is found in a string in trip$SNP, replace the 1st and 8th characters in that string according to this regex: tr /ATCG/TAGC/;

Desired output, with changes highlighted:

SNP
C[T->C]T
C[G->C]A
G[A->C]A
C[T->C]C
C[C->A]G
T[G->A]C

to:

SNP
C[T->C]T
G[C->G]T #<-- changed
C[T->G]T #<-- changed
C[T->C]C
C[C->A]G
A[C->T]G #<-- changed

Is there a more elegant way to do this?

regex r
• 3,220 views • link
updated by Ram Ram • written by soosus • soosus

So, you want to complement (but not reverse) the sequence in SNP if and only if the substitution is G-> something or A-> something, right? You can have a much easier solution, in single line of code then. Is the format for SNP fixed (substitution base +1 base before and after)?

• 0 views • link
updated by Ram Ram • written by Michael Dondrup Michael Dondrup

Yes, it's fixed. It's simply the SNP, expressed as the reference->variant, flanked by its neighbors. And yes, complement but not reverse.

• 0 views • link
written by soosus • soosus

1 answer 

## assuming df is your data frame
SNP <- as.character(df$SNP)

SNP
[1] "C[T->C]T" "C[G->C]A" "G[A->C]A" "C[T->C]C" "C[C->A]G" "T[G->A]C"

i <- grep("(A|G)->", SNP)
SNP[i] <- chartr("ACGT", "TGCA", SNP[i])

SNP
[1] "C[T->C]T" "G[C->G]T" "C[T->G]T" "C[T->C]C" "C[C->A]G" "A[C->T]G"
• 0 views • link
updated by Ram Ram • written by Michael Dondrup Michael Dondrup

Log in to answer this question.

More posts like this