Shell scripts and grep, awk command

Question

[mxlsherry1992]

Dear all,

I have a scripts like the following, but it always reported grep: .pep: No such file or directory when I running the script directly. And I am also confused with the awk command line as well as the paste command line in this script, if anyone could help me to clarify the meaning...

ls *BGI* | while read pipeline
do
grep '>' ${line}.pep | sed 's/>//g' > ${line}.modify.seqid
awk '{if($3 =="gene"){a=index($0,"~~");b=index($0,";");print substr($0,a+2,b-a-2)}}' > ${line}.origin.seqid
paste -d '\t' ${line}.origin.seqid ${line}.modify.seqid > ${line}_coress.seqid
done

Also, here are the related files and what they looks like:

1. Prunella_fulvescens_BGI.pep
2. Prunella_himalayana_BGI.pep

I tried the 'grep command line' like this:

grep '>' Prunella_fulvescens_BGI.pep.fa | sed 's/>//g' > Prunella_fulvescens_BGI.modify.seqid

It will creat a file Prunella_fulvescens_BGI.modify.seqid like this:

PRUFUL_R14685  [mRNA]  locus=scaffold1534:862:1958:- [translate_table: standard]
PRUFUL_R05501  [mRNA]  locus=scaffold10:6068:6277:- [translate_table: standard]
PRUFUL_R10205  [mRNA]  locus=scaffold1726:12292:22891:+ [translate_table: standard]
PRUFUL_R07295  [mRNA]  locus=scaffold2643:33509:34789:+ [translate_table: standard]
PRUFUL_R07296  [mRNA]  locus=scaffold2643:7708:8886:+ [translate_table: standard]
PRUFUL_R10726  [mRNA]  locus=scaffold1079:172399:200680:- [translate_table: standard]
PRUFUL_R13095  [mRNA]  locus=scaffold1079:16827:27665:+ [translate_table: standard]
PRUFUL_R13096  [mRNA]  locus=scaffold1079:111918:114560:+ [translate_table: standard]
PRUFUL_R14411  [mRNA]  locus=scaffold1079:153277:155466:+ [translate_table: standard]
PRUFUL_R07297  [mRNA]  locus=scaffold2912:21167:49513:+ [translate_table: standard]

I will be appreciated if you could help combine the scritps and explain a little bit...:)

[boaty]

it seems like your ${line} is empty. Try echo your ${line} in for loop first

[mxlsherry1992]

Yes, exactly!! When I give up the loop and use grep '>' Prunella_fulvescens_BGI.pep.fa | sed 's/>//g' > Prunella_fulvescens_BGI.modify.seqid directly, this step works well. But if you know how to echo the ${line} to make it work well in the loop....I am very new to the scripts etc....:(

[GenoMax]

grep '>' ${line}.pep

This line in your script is looking for files ending with .pep whereas they seem to have .pep.fa ending. That is first issue. There are likely others.

At a minimum modify that bit to

grep '>' ${line}.pep.fa

and see if that helps.

Note: Please post text sample snippets instead of screen shots. People can't use the data for example purposes when posted as images.

[mxlsherry1992]

Thank you so much for the guidance, I tried to change ".pep" to ".pep.fa" but still the same error....:(