What tools can convert sequence file from tabular format to fasta format?

[yangjianhunt]

Dear bioinformatics community,

I have got several deepseq files in tab delimited format: for example in this tabular format:
TAGGAACCATTAGCCAACAA 88889
GATTAGGCCCAAATGCAAAG 7799
....

or in this tabular format:
1 1 3233 223322 TAGGGCCTTAGGAAGCCTAA
1 1 3234 222334 AGGTAACCGATAGAGGTCCA
....

I would like to convert these files to fasta format. If I can use one or multiple non-sequence column as the fasta seq title, that will be nice.

What tools or scripts can I use to achieve this?

-these files are pretty big -around 400Mb, so I cannot use excel to do the job.

( I don't have programming skill yet. I google searched and found tab2fasta.pl within HOMER package, and bioscripts.convert (but this doesn't work because it requires 1st column to be name and 2nd to be sequence). I haven't tried HOMER package yet. Thought I would get some insights from your guys first.)


Thanks a lot!

Jian

[dariober]

Hi,

This python script should do it. Say your tab separated file is tabseq.tsv:

A       B       3233    223322  TAGGGCCTTAGGAAGCCTAA
C       D       3234    222334  AGGTAACCGATAGAGGTCCA

Column 5 is the sequence, one or more of the other columns to be used as header.

python tab2fasta.py tabseq.tsv 5 1 2 4  > tabseq.fa

Output (tabseq.fa) will be:

>A_B_223322
TAGGGCCTTAGGAAGCCTAA
>C_D_222334
AGGTAACCGATAGAGGTCCA

Here's the code for tab2fasta.py:

#!/usr/local/bin/python

docstring= """
DESCRIPTION
    Convert tabular to FASTA

USAGE:
    python tab2fasta.py <tab-file> <sequence column> <header column 1> <header column 2> <header column n>  > <outfile>
"""

import sys
if len(sys.argv) < 4:
    sys.exit('\nThree or more arguments required%s' %(docstring))
    
infile= open(sys.argv[1])
seqix= int(sys.argv[2]) - 1 
headerix= sys.argv[3:]
headerix= [(int(x) - 1) for x in headerix]

for line in infile:
    line= line.strip().split('\t')
    header= '>' + '_'.join([line[i] for i in headerix])
    print(header)
    print(line[seqix])

infile.close()

I've done minimal testing so make sure it does what you want!

Good luck
Dario

[essvee]

or if your file is tabseq.tsv:

A       B       3233    223322  TAGGGCCTTAGGAAGCCTAA
C       D       3234    222334  AGGTAACCGATAGAGGTCCA

you can use awk to do this easily:

awk '{print ">"$1"_"$2"_"$3"_"$4"\n"$5}' tabseq.tsv > seqs.fa

The $1, $2, etc are the column numbers, you can change these to whichever order you'd like, for example, for the other format:

TAGGAACCATTAGCCAACAA  88889
GATTAGGCCCAAATGCAAAG  7799

you could do:

awk '{print ">"$2"\n"$1}' tabseq.tsv > seqs.fa

[yangjianhunt]

You are tremendous help!

Hi Dario,

I cannot thank you enough.
I will test the code, modify it if necessary.
Yesterday I was watching the MIT opencourse on beginner programing -they use python as the example language. It's going to take at least a month to learn programming by it. I'd like to learn it. But I want to get the immediate problem solved!

Regards,
Jian

[yangjianhunt]

Thanks, essvee!

Wow, The awk solution is so simple and elegant!
I will try these as well.

I've used a few times of awk-but only through google. I never tried to fully understand the awk language. It's great for parsing!

Thank you thank you thank you!

Jian

[musta1234]

This is awesome.... thanks for the awk and python scripts

Mustapha