CODEWITHSUNDEEP
rsorting
Johnathan
Johnathan

Reputation: 1907

How can I make my sort more elegant?

I have the following data frame that with genome coordinates that I wish to sort by the first and second columns both in an increasing manner:

chr4 085976379 111570775
chr1  004336501  141626155
chr10 003051921 081538660
...

My code:

dat[order(dat[,1], dat[,2]), ]

I get:

chr1  004336501  141626155
chr10 003051921 081538660
chr4 085976379 111570775

However, I would like to have:

chr1  004336501  141626155
chr4 085976379 111570775
chr10 003051921 081538660

I could remove "chr", resort, then add "chr", but I was wondering if there were a more "elegant" one liner?

Thank you!

Upvotes: 2

Views: 51

Answers (2)

A5C1D2H2I1M1N2O1R2T1
A5C1D2H2I1M1N2O1R2T1

Reputation: 193517

You can also try mixedorder from the "gtools" package:

library(gtools)
mixedorder(mydf$V1)
# [1] 2 1 3

mydf[order(mixedorder(mydf$V1), mydf$V2), ]
#      V1       V2        V3
# 2  chr1  4336501 141626155
# 1  chr4 85976379 111570775
# 3 chr10  3051921  81538660

Upvotes: 0

Carlos Cinelli
Carlos Cinelli

Reputation: 11597

You could try:

dat[order(as.numeric(gsub("chr", "",dat[,1])), dat[,2]), ] 
     V1       V2        V3
2  chr1  4336501 141626155
1  chr4 85976379 111570775
3 chr10  3051921  81538660

Upvotes: 1

Related Questions