Is Bio.SeqIO.write() sequential for fastq format?

Question

[pyKey]

Hello everyone,

Here I try to run this python code for reading fastq records and their associated barcode from a singlae pandas dataframe and writing them simultaneously in a fastq and a text file, to preserve the order between associations. Somehow the order is lost once both files are filled and re-opened. I have no clue why, my only guess is that SeqIO.write is not writing the records one by one (ok there is also the second guess which my code has a bug ;) ) Any ideas? I also post the related methods.

Thank you very much,

'''Prepare fastq records.'''
def to_IOSeq_rec(row):
    qname = row['NAME']
    seq = row['SEQ']
    qual = row['QUAL']
    xm = row['XM']
    record = SeqRecord(Seq(seq, generic_dna), id=qname, name=qname, description='', dbxrefs=[])
    record.letter_annotations["phred_quality"] = qual
return record, xm

''' Write the separated data in two files.'''
  with open(out_file, 'w') as fq:
    with open(out_umi, 'w') as fu:
        for index, row in df.iterrows():

            record , xm = to_IOSeq_rec(row)

            Bio.SeqIO.write(record, fq, 'fastq')
            fu.write(xm + '\n')

[Joe]

As far as I'm aware the SeqIO.write() call will write things out as it's fed them from your loop - no other reordering should be occurring so I'd say your problem is elsewhere.

[Renesh]

Looks like your code is incomplete. Where is panda dataframe object defined?

Answer 1

[pyKey]

You are right, the order is preserved. I double-checked everything and found my code's bug : I was addressing a wrong dataframe somewhere along the way!

Thank you for the answers,