comparison tools/sample_seqs/sample_seqs.py @ 6:31f5701cd2e9 draft

v0.2.4 Depends on Biopython 1.67 via legacy Tool Shed package or bioconda.

5:6b71ad5d43fb

                  default=False, action="store_true",
                  help="Show version and quit")
options, args = parser.parse_args()

if options.version:
    print("v0.2.3")
    sys.exit(0)

try:
    from Bio import SeqIO
    from Bio.SeqIO.QualityIO import FastqGeneralIterator

        sys.stderr.write("Sampling every %ird sequence\n" % N)
    else:
        sys.stderr.write("Sampling every %ith sequence\n" % N)

    def sampler(iterator):
        global N
        count = 0
        for record in iterator:
            count += 1
            if count % N == 1:

elif options.percent:
    try:
        percent = float(options.percent) / 100.0
    except ValueError:
        sys.exit("Bad -p percent argument %r" % options.percent)
    if percent <= 0.0 or 1.0 <= percent:
        sys.exit("Bad -p percent argument %r" % options.percent)
    sys.stderr.write("Sampling %0.3f%% of sequences\n" % (100.0 * percent))

    def sampler(iterator):
        global percent
        count = 0
        taken = 0
        for record in iterator:
            count += 1

                taken += 1
                yield record
            assert taken == N, "Picked %i, wanted %i" % (taken, N)
    else:
        def sampler(iterator):
            # Mimic the percentage sampler, with double check on final count
            global N, total
            # Do we need a floating point fudge factor epsilon?
            # i.e. What if percentage comes out slighty too low, and
            # we could end up missing last few desired sequences?

    while True:
        if line[0] != ">":
            raise ValueError(
                "Records in Fasta files should start with '>' character")
        try:
            id = line[1:].split(None, 1)[0]
        except IndexError:
            if not no_id_warned:
                sys.stderr.write("WARNING - Malformed FASTA entry with no identifier\n")
        no_id_warned = True
        id = None
        lines = [line]
        line = handle.readline()
        while True:
            if not line:
                break

                count /= 2
            else:
                count = writer.write_file(iterator_filter(SffIterator(in_handle)))
    return count

if seq_format == "sff":
    count = sff_filter(in_file, out_file, sampler, interleaved)
elif seq_format == "fasta":
    count = fasta_filter(in_file, out_file, sampler, interleaved)
elif seq_format.startswith("fastq"):

6:31f5701cd2e9

                  default=False, action="store_true",
                  help="Show version and quit")
options, args = parser.parse_args()

if options.version:
    print("v0.2.4")
    sys.exit(0)

try:
    from Bio import SeqIO
    from Bio.SeqIO.QualityIO import FastqGeneralIterator

        sys.stderr.write("Sampling every %ird sequence\n" % N)
    else:
        sys.stderr.write("Sampling every %ith sequence\n" % N)

    def sampler(iterator):
        """Sample every Nth sequence."""
        global N
        count = 0
        for record in iterator:
            count += 1
            if count % N == 1:

elif options.percent:
    try:
        percent = float(options.percent) / 100.0
    except ValueError:
        sys.exit("Bad -p percent argument %r" % options.percent)
    if not(0.0 <= percent <= 1.0):
        sys.exit("Bad -p percent argument %r" % options.percent)
    sys.stderr.write("Sampling %0.3f%% of sequences\n" % (100.0 * percent))

    def sampler(iterator):
        """Sample given percentage of sequences."""
        global percent
        count = 0
        taken = 0
        for record in iterator:
            count += 1

                taken += 1
                yield record
            assert taken == N, "Picked %i, wanted %i" % (taken, N)
    else:
        def sampler(iterator):
            """Sample given number of sequences."""
            # Mimic the percentage sampler, with double check on final count
            global N, total
            # Do we need a floating point fudge factor epsilon?
            # i.e. What if percentage comes out slighty too low, and
            # we could end up missing last few desired sequences?

    while True:
        if line[0] != ">":
            raise ValueError(
                "Records in Fasta files should start with '>' character")
        try:
            line[1:].split(None, 1)[0]
        except IndexError:
            if not no_id_warned:
                sys.stderr.write("WARNING - Malformed FASTA entry with no identifier\n")
                no_id_warned = True
        lines = [line]
        line = handle.readline()
        while True:
            if not line:
                break

                count /= 2
            else:
                count = writer.write_file(iterator_filter(SffIterator(in_handle)))
    return count


if seq_format == "sff":
    count = sff_filter(in_file, out_file, sampler, interleaved)
elif seq_format == "fasta":
    count = fasta_filter(in_file, out_file, sampler, interleaved)
elif seq_format.startswith("fastq"):