This is a test version of Biostars. For the public version, visit https://www.biostars.org.
write multiple record fasta from pyfaidx output

Hello, I'm trying to extract several sequences of DNA from a fasta using the chromosome and coordinates. I then want to write these sequences to a fasta file. I want to do this exclusively in python.

For one sequence, I am able to extract the sequence:

from Bio import SeqIO
from pyfaidx import Fasta

genes = Fasta('Genome.fasta')

f = open('chr02_18s', 'w')

seqFile = genes['chr02'][146062:148216]
f.write(str(seqFile))

This gets the correct sequence, but it only writes the nucleotide sequence to the file, it does not include the fasta header.

If I use:

genes = Fasta('/Users/eflannery/Dropbox/Genomes/PlasmoDB-25_PvivaxSal1_Genome.fasta')
f = open('chr02_18s', 'w')
seqFile = genes['Pv_Sal1_chr02'][146062:148216]
# f.write(str(seqFile))
SeqIO.write(seqFile, f, "fasta")

I get the following error:

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-204-1a62c837614c> in <module>()
      9 seqFile = genes['Pv_Sal1_chr02'][146062:148216]
     10 # f.write(str(seqFile))
---> 11 SeqIO.write(seqFile, f, "fasta")
     12 # slice
     13 # output_handle.write(handle)

//anaconda/lib/python2.7/site-packages/Bio/SeqIO/__init__.pyc in write(sequences, handle, format)
    470         if format in _FormatToWriter:
    471             writer_class = _FormatToWriter[format]
--> 472             count = writer_class(fp).write_file(sequences)
    473         elif format in AlignIO._FormatToWriter:
    474             # Try and turn all the records into a single alignment,

//anaconda/lib/python2.7/site-packages/Bio/SeqIO/Interfaces.pyc in write_file(self, records)
    209         """
    210         self.write_header()
--> 211         count = self.write_records(records)
    212         self.write_footer()
    213         return count

//anaconda/lib/python2.7/site-packages/Bio/SeqIO/Interfaces.pyc in write_records(self, records)
    194         count = 0
    195         for record in records:
--> 196             self.write_record(record)
    197             count += 1
    198         # Mark as true, even if there where no records

//anaconda/lib/python2.7/site-packages/Bio/SeqIO/FastaIO.pyc in write_record(self, record)
    188             title = self.clean(self.record2title(record))
    189         else:
--> 190             id = self.clean(record.id)
    191             description = self.clean(record.description)
    192             if description and description.split(None, 1)[0] == id:

AttributeError: 'Sequence' object has no attribute 'id'

I obviously don't understand what kind of record is created with the fasta function in pyfaidx, but I can't find this info or how to write it to a fasta file anywhere.

Thanks for the help

pyfaidx next-gen fasta
• 4,388 views • link
updated by Ram Ram • written by mosquitoes • mosquitoes

1 answer

You're very close. The 'Sequence' object has these attributes:

Sequence:
  .name
  .seq
  .start
  .end
  .comp

When you "call" the object the Sequence.__rep__ method is called, which looks like this:

def __repr__(self):
    return '\n'.join([''.join(['>', self.longname]), self.seq])

(Sequence.longname gets the full name as represented in the FASTA file - which is not always the representation in the FASTA index file)

When you pass the Sequence object to the str function you call this method:

def __str__(self):
    return self.seq

So, you can see that str(Sequence) only gives back the sequence, and not the name. For what you want to accomplish I would recommend:

• 0 views • link
updated by Ram Ram • written by Matt Shirley Matt Shirley

Log in to answer this question.

More posts like this