collecting 50 most frequent proteins from tabular blastX result

Question

[Farbod]

Dear Friends, Hi

I have done a blastX against NCBI nr database (using Diamond and keeping -max_target_seqs = 1) with outfmt 6.

I want to collect 50 proteins with the most frequent occurance in my results.

Is there any command line sccript or program for doing this task?

(I have tried cutting the column of the IDs and then openning it in Microsoft excel and count the duplicates and . . . but opening such file and running the duplicate count in my Windows system computer which is not very powerful is very difficult)

Thank you in advance

[GenoMax]

Perhaps this would help (see @Pierre's answer or python scripts if that is not going to help): Blastp how to find and count duplicates?..

[Farbod]

Dear genomax2, Hi & thank you.

but I could not understand that what is the final correct python script ?

[Asaf]

Simple: cut -f 1 blast_out.tbl | sort | uniq -c | sort -k1gr |head -50

[Farbod]

Dear Asef, Hi

It seems that it is magically working!

Thank you

[Asaf]

No magic, just simple unix command liners

[Farbod]

Dear Asef,

it seems that your script has two sort commands in it, can we reduce it to just one ?

~ Best

[GenoMax]

Probably not. You can start at left and keep running the commands, every-time adding one more term (from the pipes) to see why not.

 cut -f 1 blast_out.tbl | less
 cut -f 1 blast_out.tbl | sort | less
 cut -f 1 blast_out.tbl | sort | uniq -c | less

You get the idea.