How to parse fasta string using Seq.IO from biopython

Question

I need to use a fasta string instead of fasta file to parse it in Seq.IO in python3

from Bio import SeqIO

fasta_string = '>name\nACCTGTGGCTGCTTGCTTGCTTGGGCT'

rec = SeqIO.parse(fasta_string, "fasta")

print (rec[0])

I read that I need to create a file object using io module like

import io

with io.StringIO() as f:
    f.write('abcdef')
    print('gh', file=f)
    f.seek(0)
    print(f.read())

But Seq.IO needs a path to file not only the object file but I can't write a temporary file.

Any ideas?

Thank you in advance.

score 2 · Answer 1 · answered Apr 03 '21 at 18:16

Ok think I got it from here:

Biopython parse from variable instead of file on SO

from io import StringIO
from Bio import SeqIO

# fasta_string = '>name\nACCTGTGGCTGCTTGCTTGCTTGGGCT'

fasta_string = '>name\nACCTGTGGCTGCTTGCTTGCTTGGGCT\n>name2\nCCCCCCCCCGGGGGGGGGGGTTTTTTTAAA'

fasta_io = StringIO(fasta_string) 

records = SeqIO.parse(fasta_io, "fasta") 

for rec in records:
    print(rec)

fasta_io.close()

output:

ID: name
Name: name
Description: name
Number of features: 0
Seq('ACCTGTGGCTGCTTGCTTGCTTGGGCT')
ID: name2
Name: name2
Description: name2
Number of features: 0
Seq('CCCCCCCCCGGGGGGGGGGGTTTTTTTAAA')

Unfortunately, I did not manage to find this post earlier, sorry for that inconvenience. Thank you so much for your explanations! — Konstantin Kuleshov, Apr 04 '21 at 09:45

score 0 · Answer 2 · answered Apr 03 '21 at 13:24

I quite don't understand your question,

in any case:

from Bio import SeqIO

fasta_string = '>name\nACCTGTGGCTGCTTGCTTGCTTGGGCT'

rec = SeqIO.parse(fasta_string, "fasta")

print (rec[0])

gives ERROR :

FileNotFoundError: [Errno 2] No such file or directory: '>name\nACCTGTGGCTGCTTGCTTGCTTGGGCT'

so to write your sequence out to a fasta file using BioPython

I can use:

from Bio import SeqIO

from Bio.Seq import Seq

from Bio.SeqRecord import SeqRecord
fasta_string = '>name\nACCTGTGGCTGCTTGCTTGCTTGGGCT'

fasta_string_name , fasta_string_seq = fasta_string.strip('>').split('\n')

print(fasta_string_name , fasta_string_seq,'\n\n') # this one just to check the values


# SeqRecord object see https://biopython.org/wiki/SeqRecord
rec = SeqRecord(
    Seq(fasta_string_seq),
    id='',
    name='',
    description=fasta_string_name)


print('SeqRecord object : ',rec, sep='\n') # this one just to check the SeqRecord object 

recordz =[rec]  #this one not necessary but if you have more rec
                #I believe you need them into a list for SeqIO to write them out check it


SeqIO.write(recordz, "example.fasta", "fasta") #this one write out your string as fasta file

example.fasta file written to script path is:

> name
ACCTGTGGCTGCTTGCTTGCTTGGGCT

Again not sure if that is what you were trying to accomplish, just let me know.

Thank you for your response. The main idea is to pass fasta file (with many records) as a huge string to Seq.IO parse. I understand that I can parse this fasta string by myself, however I would like to know if any options to use Seq.IO parse without saving and opening fasta as a file. I am trying to parse fasta files in the stream without writing and reading to files. — Konstantin Kuleshov, Apr 03 '21 at 16:53
ok see here : https://stackoverflow.com/questions/38358191/biopython-parse-from-variable-instead-of-file Biopython parse from variable instead of file see answer number 2. Had to flag your question as duplicate to keep SO in shape — pippo1980, Apr 03 '21 at 17:39
here: fastq_io = StringIO(rec) records = SeqIO.parse(fastq_io, "fastq") fastq_io.close() — pippo1980, Apr 03 '21 at 17:40

How to parse fasta string using Seq.IO from biopython

2 Answers2