Python 3.4, BioPython 1.65: Specifying dictionary keys in FASTA file and writing sequences with certain key values to file

Question

[tyleraelliott]

I have a multi-fasta file containing sequences with headers such as:

>AAST01014508.1|(1..6240)|LTR/Pao|ROOA_I-int:ROOA_LTR
>AAST01026747.1|c(2745..6820)|LTR/Pao|ROO_I-int

In the second section of the header where the position is given I want to index all of the sequences using this and print to file all those sequence from the C strand, e.g. >AAST01026747.1|c(2745..6820)|LTR/Pao|ROO_I-int

I was able to create a function easily which did this:

def get_compstrandTE(record):
    parts=record.id.split("|")
    assert len(parts) ==4
    return parts[1]

However, I am now stuck as to how to search through the keys in the dictionary and find only those containing the 'c' and write those to a file. I tried using the example from the BioPython manual but kept running into difficulties.

If anyone has any suggestions I would really appreciate it.

Answer 1

[Peter]

You don't need a dictionary for this task.

Something like this should work, loosely based on of the filtering examples from the Biopython Tutorial using a generator expression:

from Bio import SeqIO
input_file = "big_file.fasta"
output_file = "complements.fasta"

def wanted(record):
    """Returns True if name scheme suggests from complement stand."""
    parts = record.id.split("|")
    assert len(parts) == 4
    return parts[1].startswith("c")

records = (r for r in SeqIO.parse(input_file, "fasta") if wanted(r))
count = SeqIO.write(records, output_file, "fasta")
print("Saved %i records from %s to %s" % (count, input_file, output_file))

[tyleraelliott]

Wonderful, thank again!