Question

This is a very simple question.

I have a lengthy dataset and want to create a subset based on certain entries in a particular column. In this case, I am setting it up like this:

Example data:

> NL

SNP alleles

rs1234 A_T

rs1235 A_G

rs2343 A_T

rs2342 G_C

rs1134 C_G

rs1675 T_A

rs8543 A_T

rs2842 G_A

P <- subset(NL, alleles = "A_T", alleles = "T_A", alleles = "G_C", alleles = "C_G")

This runs without error, but the resulting P is not subset in any way (tail of P still shows same number of entries as original NL).

What am I doing wrong?

Answer 1

The most obvious error is using “=” when you mean”==”. But I’m guessing from context that you really want to “split” this data:

split(NL, NL$alleles)

Which will create a list of dataframes each of which has one of the values for alleles.

But perhaps you do want to use pattern matching:

NL[ grepl("C_G|G_C|A_T|T_A", NL$alleles), ]
     SNP alleles
1 rs1234     A_T
3 rs2343     A_T
4 rs2342     G_C
5 rs1134     C_G
6 rs1675     T_A
7 rs8543     A_T

And illustrating with what I think was your comment-example:

P <- read.table(text="V1 V2 V3 V4 V5 V6 alleles
 15116 25 rsX 0 123412 G A G_A 
15117 25 rsX1 0 23432 A C A_C 
15118 25 rsX2 0 234324 A G A_G 
15119 25 rsX3 0 3423 A G A_G 
15120 25 rsX4 0 2343223 C A C_A 
15121 25 rsX5 0 23523423 A G A_G", header=TRUE)

 P[ grepl("G_A", NL$alleles), ]

#       V1       V2 V3        V4 V5 V6 alleles
# 15116 25 rs306910  0 154613671  G  A     G_A

The subset version:

 subset(P, alleles %in% c("G_A", "A_G") )

      V1   V2 V3       V4 V5 V6 alleles
15116 25  rsX  0   123412  G  A     G_A
15118 25 rsX2  0   234324  A  G     A_G
15119 25 rsX3  0     3423  A  G     A_G
15121 25 rsX5  0 23523423  A  G     A_G