How to split a string and swap the result in bash

I have a column which is the output of an annotation command (bcftools query -f '%POS\t%REF\t%ALT\t%BCSQ\n')

The column I'm interested in is the amino acid position and amino acid change for SNPs, like so:

402Y>402H

How can I use bash code to swap these two around (if it contains a ">". Synonymous changes would just be 402Y for example)

So the result would be

402H>402Y

Thanks

text string bash split

• 1,949 views • link

updated 6.3 years ago by wm

• written 6.3 years ago by Luther_Blisset •

2 answers

Could be better to provide some example, but in general, you can do this:

$ echo "402Y>402H" | perl -pe 's/(\d+)(\w)>(\d+)(\w)/$1$4>$3$2/'
402H>402Y

• 0 views • link

written 6.3 years ago by JC

Great, thanks.

I have a couple of questions though -

1) it seems to fail where I have 'stop lost' annotations: 75*>75Q I presume this is because the * appears just before the >.

2) Do you know how I can do this in-place in the original file/table? For example the first two rows look like this:

123520  T   C   missense    Rv0104  Rv0104  protein_coding  +   402Y>402H   123520T>C  
199470  T   G   missense    mce1A   Rv0169  protein_coding  +   313S>313A   199470T>G

I just need to apply this change to the 9th column here. Thanks

• 0 views • link

written 6.3 years ago by Luther_Blisset •

echo "75*>75Q" | perl -pe 's/(\d+)(.)>(\d+)(.)/$1$4>$3$2/'

The "." means anything. This will apply to a new file, is not recommended to do in the origin file.

$ perl  -pe 's/(\d+)(.)>(\d+)(.)/$1$4>$3$2/' < test.txt > out.txt
$ cat out.txt
123520  T   C   missense    Rv0104  Rv0104  protein_coding  +   402H>402Y   123520T>C
199470  T   G   missense    mce1A   Rv0169  protein_coding  +   313A>313S   199470T>G

• 0 views • link

written 6.3 years ago by JC

How about try this command?

$ cat in.txt
123520  T       C       missense        Rv0104  Rv0104  protein_coding  +       402Y>402H       123520T>C
199470  T       G       missense        mce1A   Rv0169  protein_coding  +       313S>313A       199470T>G
199470  T       G       missense        mce1A   Rv0169  protein_coding  +       75*>75Q 199470T>G

$ awk '{OFS="\t"; split($9,a,">"); $9=a[2]">"a[1]; print}' in.txt
123520  T       C       missense        Rv0104  Rv0104  protein_coding  +       402H>402Y       123520T>C
199470  T       G       missense        mce1A   Rv0169  protein_coding  +       313A>313S       199470T>G
199470  T       G       missense        mce1A   Rv0169  protein_coding  +       75Q>75* 199470T>G

• 0 views • link

written 6.3 years ago by wm