comparison RepEnrich2.py @ 7:61e0404f0d76 draft

planemo upload for repository https://github.com/ARTbio/tools-artbio/tree/main/tools/repenrich2 commit 067aa46ea7482d640369b95d5b3dabec5793396b

6:388a47ca4199

    file = csv.reader(open(filename), delimiter=separator,
                      skipinitialspace=True)
    return [line for line in file if starts_with_numerical(line)]


def run_bowtie(args):
    metagenome, fastqfile = args
    b_opt = "-k 1 -p 1 --quiet --no-hd"
    command = shlex.split(f"bowtie2 {b_opt} -x {metagenome} {fastqfile}")
    bowtie_align = subprocess.run(command, check=True,
                                  capture_output=True, text=True).stdout
    bowtie_align = bowtie_align.rstrip('\r\n').split('\n')
    readlist = [metagenome]
    if paired_end:
        for line in bowtie_align:
            readlist.append(line.split("/")[0])
    else:
        for line in bowtie_align:
            readlist.append(line.split("\t")[0])
    return readlist


# set a reference repeat list for the script
repeat_list = [listline[9].translate(
    str.maketrans(
        '()/', '___')) for listline in import_text(annotation_file, ' ')]
repeat_list = sorted(list(set(repeat_list)))

with open("unique_mapper_counts.tsv", 'w') as fout:
    fout.write("#element\tcount\n")
    for count in sorted(counts):
        fout.write(f"{count}\t{counts[count]}\n")

# multimapper parsing
if not paired_end:
    args_list = [(metagenome, fastqfile_1) for
                 metagenome in repeat_list]
    with ProcessPoolExecutor(max_workers=cpus) as executor:
        results = executor.map(run_bowtie, args_list)
else:
    args_list = [(metagenome, fastqfile_1) for
                 metagenome in repeat_list]
    args_list.extend([(metagenome, fastqfile_2) for
                     metagenome in repeat_list])
    with ProcessPoolExecutor(max_workers=cpus) as executor:
        results = executor.map(run_bowtie, args_list)

# Aggregate results (avoiding race conditions)
metagenome_reads = defaultdict(list)  # repeat_name: list of multimap reads
for result in results:
    metagenome_reads[result[0]] += result[1:]

for name in metagenome_reads:
    #  read are only once in list
    metagenome_reads[name] = list(set(metagenome_reads[name]))
    #  remove "no read" instances
    metagenome_reads[name] = [read for read in metagenome_reads[name]
                              if read != ""]

# implement repeats_by_reads from the inverse dictionnary metagenome_reads
repeats_by_reads = defaultdict(list)  # readids: list of repeats names
for repname in metagenome_reads:
    for read in metagenome_reads[repname]:

7:61e0404f0d76

    file = csv.reader(open(filename), delimiter=separator,
                      skipinitialspace=True)
    return [line for line in file if starts_with_numerical(line)]


# set a reference repeat list for the script
repeat_list = [listline[9].translate(
    str.maketrans(
        '()/', '___')) for listline in import_text(annotation_file, ' ')]
repeat_list = sorted(list(set(repeat_list)))

with open("unique_mapper_counts.tsv", 'w') as fout:
    fout.write("#element\tcount\n")
    for count in sorted(counts):
        fout.write(f"{count}\t{counts[count]}\n")


def run_bowtie(args):
    '''
    write to files to save memory
    '''
    metagenome, fastqfile = args
    b_opt = "-k 1 -p 1 --quiet --no-hd"
    command = shlex.split(f"bowtie2 {b_opt} -x {metagenome} {fastqfile}")
    bowtie_align = subprocess.run(command, check=True,
                                  capture_output=True, text=True).stdout
    bowtie_align = bowtie_align.rstrip('\r\n').split('\n')
    with open(f"{metagenome}.reads", "a+") as readfile:
        for line in bowtie_align:
            read = line.split()[0].split("/")[0]
            if read:
                readfile.write(f"{read}\n")


# multimapper parsing
args_list = [(metagenome, fastqfile_1) for metagenome in repeat_list]
if paired_end:
    args_list.extend([(metagenome, fastqfile_2) for
                     metagenome in repeat_list])
with ProcessPoolExecutor(max_workers=cpus) as executor:
    results = executor.map(run_bowtie, args_list)

# Aggregate results (avoiding race conditions)
metagenome_reads = defaultdict(list)  # metagenome: list of multimap reads

# Now we read .reads file to populate metagnomes_reads
for metagenome in repeat_list:
    with open(f"{metagenome}.reads") as readfile:
        for read in readfile:
            metagenome_reads[metagenome].append(read)
        # read are only once in list
        metagenome_reads[metagenome] = list(set(metagenome_reads[metagenome]))

# implement repeats_by_reads from the inverse dictionnary metagenome_reads
repeats_by_reads = defaultdict(list)  # readids: list of repeats names
for repname in metagenome_reads:
    for read in metagenome_reads[repname]: