Parsing AME tsv file

Question

[rbronste]

I am trying to find a quick and easy way to parse an AME generated true positive sequences.tsv file to pull out just a 3 column BED, the format look as follows, any ideas would be awesome thanks!

motif_DB  motif_ID  seq_ID   FASTA_score PWM_score  class

Jaspar  MA0004.1    chr5:144788829-144789179_shuf_2     2183    12.7135 fp
Jaspar  MA0004.1    chr5:112339537-112339887_shuf_1     1713    12.7131  tp
Jaspar  MA0004.1    chr16:94739915-94740265_shuf_1      1668    12.712   tp

Answer 1

[Pierre Lindenbaum]

grep -v ^motif in.tsv | cut -f 3 | cut -d '_' -f 1 | tr ":-" "\t"

[rbronste]

Thanks very helpful! Is there additionally a way to include only the true-positive sequences (tp in final column) in the output bed?

[WouterDeCoster]

That will be another grep or awk in the command :)
I think you can figure out how to do that?

[rbronste]

Maybe a hint? :) Not as familiar with awk, though trying to learn.

[WouterDeCoster]

I would add another grep to get lines with tp, prior to cut.

[rbronste]

Ok figured it out seems to work like this for true positive intervals with specific motifs IDs:

grep -v ^motif sequences.tsv | grep -w tp | grep -w MA0258.2 | cut -f 2,3,6 | cut -d '_' -f 1 | tr ":-" "\t" | head

MA0258.2    chr12   15566967    15567317    tp
MA0258.2    chr11   88155633    88155983    tp
MA0258.2    chr15   51071410    51071760    tp
MA0258.2    chr14   22151488    22151838    tp

Thanks for your help.

[rbronste]

Thanks very helpful!

Used this to get the following:

grep -v ^motif sequences.tsv | cut -f 2,3  | cut -d '_' -f 1 | tr ":-" "\t" 

    MA0258.2    chr12   15566967    15567317    tp
    MA0258.2    chr11   88155633    88155983    tp
    MA0258.2    chr15   51071410    51071760    tp
    MA0258.2    chr14   22151488    22151838    tp

However can't quite figure out how to select only specific motif_IDs in the .tsv file as well as only tp (true positive) values for those specific motif IDs.