filter out low fpkm with unix

[emolinari]

Hi guys!!!

I am terrible in unix scripting...but I have a feeling that is the only way to sort out my problem:

I want to filter out from my .gtf file from cufflinks those transcripts with a fpkm < 1.

I have tried
awk '($14>1)' transcripts.gtf > transcripts.b.gtf

but of course with no success...
$14 is the position of fpkm values
This is the first line of the .gtf file

1 Cufflinks transcript 53049 54936 1 + . gene_id "ENSG00000268020"; transcript_id "ENST00000594647"; FPKM "0.0000000000"; frac "0.000000"; conf_lo "0.000000"; conf_hi "0.000000"; cov "0.000000"; full_read_support "no";


Can anyone help???

thanks!
Manu

[vivek_]

Try removing the quote marks before doing a numerical comparison.

awk '{gsub(/\"/,""); print}' transcripts.gtf  | awk '{if($14 > 1) print}'

[emolinari]

Try removing the quote marks before doing a numerical comparison.

awk '{gsub(/\"/,""); print}' transcripts.gtf  | awk '{if($14 > 1) print}'

Yayyyyy!!!

thanks Vivek! I also imagined it was an issue with quote marks...but I'd never imagined how to do it. That was extremely helpful...thanks indeed!!!


Manu