Perl Help

[Wei-HD]

Hi All,

When I upload the junction.bed file to UCSC, there was an error happen: "the chromosome name is case sensitive..." because name of chromosome is "1, 2, 3, 4,..." in my bowtie index.

I wanted to change the name "1, 2, 3..." to "chr1, chr2, chr3...", then I use this code "perl -pi -e 's/^/chr/' filename" But I met an issue, since I run this code two times by accident , then the first string of each line of bed file became this embarrassing condition: "chrchr1,chrchr2,chrchr3..."

Does anyone can help me to change it back to chr1.....? I am not familiar with Perl.

Many thanks in advance!

[RockChalkJayhawk]

perl -ane 's/chrchr/chr/g; print;' filename

[Wei-HD]

Thank you very much, RockChalkJayhawk, that works very well.

[Wei-HD]

Sorry, I am sort of silly at that time. I can just use text editor and then replace all.

[pmiguel]

perl -ane 's/chrchr/chr/g; print;' filename

Or,

perl -pe 's/chrchr/chr/g'

accomplishes the same thing. -a isn't used in this context. and -p is like -n but prints every line.

If your fingers are tired, might make a difference...

--
Phillip

[RockChalkJayhawk]

Or,

perl -pe 's/chrchr/chr/g'

accomplishes the same thing. -a isn't used in this context. and -p is like -n but prints every line.

If your fingers are tired, might make a difference...

--
Phillip

You can learn something new everyday. Thanks Phillip!

[Wei-HD]

Thanks RockChalkJayhawk and Phillip!

[krobison]

Your original line probably should have been

perl -pi -e 's/^([^c])/chr$1/'

[Joann]

Thanks Keith

Your original line probably should have been

perl -pi -e 's/^([^c])/chr$1/'

Was just wondering.

[pmiguel]

Your original line probably should have been

perl -pi -e 's/^([^c])/chr$1/'

or create a back up file in case the in place edit (-i) goes awry:

perl -i'.bak' -pe 's/^/chr/' file.bed

because if you find out you hosed file.bed, you can

mv file.bed.bak file.bed

to recover.

--
Phillip