Пример #1
0
def correct():
    answers = request.form.getlist('question[]')
    inputfile = request.files.get('file')

    inputfile.save('input/' + inputfile.filename)

    answers = [question for question in answers if question]

    corrector.main(inputfile.filename, answers)

    df = pd.read_csv('static/output.csv')
    df.index += 1
    return df.to_html()
Пример #2
0
def main(args):
    os.environ["LC_ALL"] = "C"

    if len(args) == 1:
        options_storage.usage(spades_version)
        sys.exit(0)

    log = logging.getLogger('spades')
    log.setLevel(logging.DEBUG)

    console = logging.StreamHandler(sys.stdout)
    console.setFormatter(logging.Formatter('%(message)s'))
    console.setLevel(logging.DEBUG)
    log.addHandler(console)

    support.check_binaries(bin_home, log)

    # parse options and safe all parameters to cfg
    cfg, dataset_data = fill_cfg(args, log)

    if options_storage.continue_mode:
        cmd_line, options = get_options_from_params(os.path.join(options_storage.output_dir, "params.txt"))
        if not options:
            support.error("failed to parse command line of the previous run! Please restart from the beginning or specify another output directory.")
        cfg, dataset_data = fill_cfg(options, log)
        if options_storage.restart_from:
            check_cfg_for_restart_from(cfg)
        options_storage.continue_mode = True

    log_filename = os.path.join(cfg["common"].output_dir, "spades.log")
    if options_storage.continue_mode:
        log_handler = logging.FileHandler(log_filename, mode='a')
    else:
        log_handler = logging.FileHandler(log_filename, mode='w')
    log.addHandler(log_handler)

    if options_storage.continue_mode:
        log.info("\n======= SPAdes pipeline continued. Log can be found here: " + log_filename + "\n")
        log.info("Restored from " + cmd_line)
        if options_storage.restart_from:
            updated_params = ""
            flag = False
            for v in args[1:]:
                if v == '-o' or v == '--restart-from':
                    flag = True
                    continue
                if flag:
                    flag = False
                    continue
                updated_params += " " + v
            updated_params = updated_params.strip()
            log.info("with updated parameters: " + updated_params)
            cmd_line += " " + updated_params
        log.info("")

    params_filename = os.path.join(cfg["common"].output_dir, "params.txt")
    params_handler = logging.FileHandler(params_filename, mode='w')
    log.addHandler(params_handler)

    if options_storage.continue_mode:
        log.info(cmd_line)
    else:
        command = "Command line:"
        for v in args:
            command += " " + v
        log.info(command)

    # special case
    if "mismatch_corrector" in cfg and not support.get_lib_ids_by_type(dataset_data, 'paired-end'):
        support.warning('cannot perform mismatch correction without at least one paired-end library! Skipping this step.', log)
        del cfg["mismatch_corrector"]

    print_used_values(cfg, log)
    log.removeHandler(params_handler)

    if not options_storage.continue_mode:
        log.info("\n======= SPAdes pipeline started. Log can be found here: " + log_filename + "\n")

    # splitting interlaced reads and processing Ns in additional contigs if needed
    if support.dataset_has_interlaced_reads(dataset_data) or support.dataset_has_additional_contigs(dataset_data):
        dir_for_split_reads = os.path.join(options_storage.output_dir, 'split_input')
        if support.dataset_has_interlaced_reads(dataset_data):
            if not os.path.isdir(dir_for_split_reads):
                os.makedirs(dir_for_split_reads)
            dataset_data = support.split_interlaced_reads(dataset_data, dir_for_split_reads, log)
        if support.dataset_has_additional_contigs(dataset_data):
            dataset_data = support.process_Ns_in_additional_contigs(dataset_data, dir_for_split_reads, log)
        options_storage.dataset_yaml_filename = os.path.join(options_storage.output_dir, "input_dataset.yaml")
        pyyaml.dump(dataset_data, open(options_storage.dataset_yaml_filename, 'w'))
        cfg["dataset"].yaml_filename = options_storage.dataset_yaml_filename

    try:
        # copying configs before all computations (to prevent its changing at run time)
        tmp_configs_dir = os.path.join(cfg["common"].output_dir, "configs")
        if os.path.isdir(tmp_configs_dir) and not options_storage.continue_mode:
            shutil.rmtree(tmp_configs_dir)
        if not os.path.isdir(tmp_configs_dir):
            dir_util.copy_tree(os.path.join(spades_home, "configs"), tmp_configs_dir, preserve_times=False)

        corrected_dataset_yaml_filename = ''
        if "error_correction" in cfg:
            STAGE_NAME = "Read error correction"
            bh_cfg = merge_configs(cfg["error_correction"], cfg["common"])
            corrected_dataset_yaml_filename = os.path.join(bh_cfg.output_dir, "corrected.yaml")
            if os.path.isfile(corrected_dataset_yaml_filename) and options_storage.continue_mode \
                and not options_storage.restart_from == "ec":
                log.info("\n===== Skipping %s (already processed). \n" % STAGE_NAME)
            else:
                support.continue_from_here(log)

                if "HEAPCHECK" in os.environ:
                    del os.environ["HEAPCHECK"]
                if "heap_check" in bh_cfg.__dict__:
                    os.environ["HEAPCHECK"] = bh_cfg.heap_check

                if os.path.exists(bh_cfg.output_dir):
                    shutil.rmtree(bh_cfg.output_dir)
                os.makedirs(bh_cfg.output_dir)

                if support.get_lib_ids_by_type(dataset_data, options_storage.LONG_READS_TYPES):
                    not_used_dataset_data = support.get_libs_by_type(dataset_data, options_storage.LONG_READS_TYPES)
                    to_correct_dataset_data = support.rm_libs_by_type(dataset_data, options_storage.LONG_READS_TYPES)
                    to_correct_dataset_yaml_filename = os.path.join(bh_cfg.output_dir, "to_correct.yaml")
                    pyyaml.dump(to_correct_dataset_data, open(to_correct_dataset_yaml_filename, 'w'))
                    bh_cfg.__dict__["dataset_yaml_filename"] = to_correct_dataset_yaml_filename
                else:
                    not_used_dataset_data = None
                    bh_cfg.__dict__["dataset_yaml_filename"] = cfg["dataset"].yaml_filename

                log.info("\n===== %s started. \n" % STAGE_NAME)
                hammer_logic.run_hammer(corrected_dataset_yaml_filename, tmp_configs_dir, bin_home, bh_cfg, not_used_dataset_data,
                    ext_python_modules_home, log)
                log.info("\n===== %s finished. \n" % STAGE_NAME)

        result_contigs_filename = os.path.join(cfg["common"].output_dir, "contigs.fasta")
        result_scaffolds_filename = os.path.join(cfg["common"].output_dir, "scaffolds.fasta")
        misc_dir = os.path.join(cfg["common"].output_dir, "misc")
        ### if mismatch correction is enabled then result contigs are copied to misc directory
        assembled_contigs_filename = os.path.join(misc_dir, "assembled_contigs.fasta")
        assembled_scaffolds_filename = os.path.join(misc_dir, "assembled_scaffolds.fasta")
        if "assembly" in cfg:
            STAGE_NAME = "Assembling"
            spades_cfg = merge_configs(cfg["assembly"], cfg["common"])
            spades_cfg.__dict__["result_contigs"] = result_contigs_filename
            spades_cfg.__dict__["result_scaffolds"] = result_scaffolds_filename

            if options_storage.continue_mode and (os.path.isfile(spades_cfg.result_contigs)
                                                  or ("mismatch_corrector" in cfg and
                                                      os.path.isfile(assembled_contigs_filename)))\
                and not options_storage.restart_from == 'as' \
                and not (options_storage.restart_from and options_storage.restart_from.startswith('k')):

                log.info("\n===== Skipping %s (already processed). \n" % STAGE_NAME)
                # calculating latest_dir for the next stages
                latest_dir = support.get_latest_dir(os.path.join(spades_cfg.output_dir, "K*"))
                if not latest_dir:
                    support.error("failed to continue the previous run! Please restart from previous stages or from the beginning.", log)
            else:
                old_result_files = [result_contigs_filename, result_scaffolds_filename,
                                    assembled_contigs_filename, assembled_scaffolds_filename]
                for format in [".fasta", ".fastg"]:
                    for old_result_file in old_result_files:
                        if os.path.isfile(old_result_file[:-6] + format):
                            os.remove(old_result_file[:-6] + format)

                if options_storage.restart_from == 'as':
                    support.continue_from_here(log)

                if os.path.isfile(corrected_dataset_yaml_filename):
                    dataset_data = pyyaml.load(open(corrected_dataset_yaml_filename, 'r'))
                    dataset_data = support.relative2abs_paths(dataset_data, os.path.dirname(corrected_dataset_yaml_filename))
                if spades_cfg.disable_rr:
                    spades_cfg.__dict__["rr_enable"] = False
                else:
                    spades_cfg.__dict__["rr_enable"] = True

                if "HEAPCHECK" in os.environ:
                    del os.environ["HEAPCHECK"]
                if "heap_check" in spades_cfg.__dict__:
                    os.environ["HEAPCHECK"] = spades_cfg.heap_check

                log.info("\n===== %s started.\n" % STAGE_NAME)

                # creating dataset
                dataset_filename = os.path.join(spades_cfg.output_dir, "dataset.info")
                if not os.path.isfile(dataset_filename) or not options_storage.continue_mode:
                    dataset_file = open(dataset_filename, 'w')
                    import process_cfg
                    dataset_file.write("single_cell" + '\t' + process_cfg.bool_to_str(cfg["dataset"].single_cell) + '\n')
                    if os.path.isfile(corrected_dataset_yaml_filename):
                        dataset_file.write("reads" + '\t' + process_cfg.process_spaces(corrected_dataset_yaml_filename) + '\n')
                    else:
                        dataset_file.write("reads" + '\t' + process_cfg.process_spaces(cfg["dataset"].yaml_filename) + '\n')
                    if spades_cfg.developer_mode and "reference" in cfg["dataset"].__dict__:
                        dataset_file.write("reference_genome" + '\t')
                        dataset_file.write(process_cfg.process_spaces(cfg["dataset"].reference) + '\n')
                    dataset_file.close()
                spades_cfg.__dict__["dataset"] = dataset_filename

                latest_dir = spades_logic.run_spades(tmp_configs_dir, bin_home, spades_cfg, dataset_data, ext_python_modules_home, log)

                if os.path.isdir(misc_dir) and not options_storage.continue_mode:
                    shutil.rmtree(misc_dir)
                if not os.path.isdir(misc_dir):
                    os.makedirs(misc_dir)

                if options_storage.continue_mode and options_storage.restart_from and options_storage.restart_from.startswith('k'):
                    k_str = options_storage.restart_from[1:]
                    if k_str.find(":") != -1:
                        k_str = k_str[:k_str.find(":")]
                    support.error("failed to continue from K=%s because this K was not processed in the original run!" % k_str, log)
                log.info("\n===== %s finished. \n" % STAGE_NAME)

            #corrector
            if "mismatch_corrector" in cfg and (os.path.isfile(result_contigs_filename) or
                                                (options_storage.continue_mode and os.path.isfile(assembled_contigs_filename))):
                STAGE_NAME = "Mismatch correction"
                to_correct = dict()
                to_correct["contigs"] = (result_contigs_filename, assembled_contigs_filename)
                if os.path.isfile(result_scaffolds_filename) or (options_storage.continue_mode and
                                                                 os.path.isfile(assembled_scaffolds_filename)):
                    to_correct["scaffolds"] = (result_scaffolds_filename, assembled_scaffolds_filename)

                # moving assembled contigs (scaffolds) to misc dir
                for assembly_type, (old, new) in to_correct.items():
                    if options_storage.continue_mode and os.path.isfile(new):
                        continue
                    for format in [".fasta", ".fastg"]:
                        if os.path.isfile(old[:-6] + format):
                            shutil.move(old[:-6] + format, new[:-6] + format)

                if options_storage.continue_mode and os.path.isfile(result_contigs_filename) and \
                    (os.path.isfile(result_scaffolds_filename) or not os.path.isfile(assembled_scaffolds_filename)) \
                    and not options_storage.restart_from == 'mc':
                    log.info("\n===== Skipping %s (already processed). \n" % STAGE_NAME)
                else:
                    if options_storage.restart_from == 'mc':
                        support.continue_from_here(log)

                    log.info("\n===== %s started." % STAGE_NAME)
                    # detecting paired-end library with the largest insert size
                    est_params_data = pyyaml.load(open(os.path.join(latest_dir, "final.lib_data"), 'r'))
                    max_IS_library = None
                    for reads_library in est_params_data:
                        if reads_library['type'] == 'paired-end':
                            if not max_IS_library or float(reads_library["insert size mean"]) > float(max_IS_library["insert size mean"]):
                                max_IS_library = reads_library
                    if not max_IS_library:
                        support.error('Mismatch correction cannot be performed without at least one paired-end library!', log)
                    if not max_IS_library["insert size mean"]:
                        support.warning('Failed to estimate insert size for all paired-end libraries. Starting Mismatch correction'
                                        ' based on the first paired-end library and with default insert size.', log)
                    else:
                        cfg["mismatch_corrector"].__dict__["insert-size"] = round(max_IS_library["insert size mean"])
                    yaml_dirname = os.path.dirname(options_storage.dataset_yaml_filename)
                    cfg["mismatch_corrector"].__dict__["1"] = list(map(lambda x: os.path.join(yaml_dirname, x),
                        max_IS_library['left reads']))
                    cfg["mismatch_corrector"].__dict__["2"] = list(map(lambda x: os.path.join(yaml_dirname, x),
                        max_IS_library['right reads']))
                    #TODO: add reads orientation

                    import corrector
                    corrector_cfg = cfg["mismatch_corrector"]
                    args = []
                    for key, values in corrector_cfg.__dict__.items():
                        if key == "output-dir":
                            continue

                        # for processing list of reads
                        if not isinstance(values, list):
                            values = [values]
                        for value in values:
                            if len(key) == 1:
                                args.append('-' + key)
                            else:
                                args.append('--' + key)
                            if value is not None:
                                args.append(value)

                    # processing contigs and scaffolds (or only contigs)
                    for assembly_type, (corrected, assembled) in to_correct.items():
                        if options_storage.continue_mode and os.path.isfile(corrected):
                            log.info("\n== Skipping processing of " + assembly_type + " (already processed)\n")
                            continue

                        support.continue_from_here(log)
                        log.info("\n== Processing of " + assembly_type + "\n")

                        cur_args = args[:]
                        cur_args += ['-c', assembled]
                        tmp_dir_for_corrector = support.get_tmp_dir(prefix="mis_cor_%s_" % assembly_type)
                        cur_args += ['--output-dir', tmp_dir_for_corrector]

                        # correcting
                        corrector.main(cur_args, ext_python_modules_home, log)

                        result_corrected_filename = os.path.join(tmp_dir_for_corrector, "corrected_contigs.fasta")
                        # moving corrected contigs (scaffolds) to SPAdes output dir
                        if os.path.isfile(result_corrected_filename):
                            shutil.move(result_corrected_filename, corrected)

                        if os.path.isdir(tmp_dir_for_corrector):
                            shutil.rmtree(tmp_dir_for_corrector)

                        assembled_fastg = assembled[:-6] + ".fastg"
                        if os.path.isfile(assembled_fastg):
                            support.create_fastg_from_fasta(corrected, assembled_fastg, log)
                    log.info("\n===== %s finished.\n" % STAGE_NAME)

        if not cfg["common"].developer_mode and os.path.isdir(tmp_configs_dir):
            shutil.rmtree(tmp_configs_dir)

        #log.info("")
        if "error_correction" in cfg and os.path.isdir(os.path.dirname(corrected_dataset_yaml_filename)):
            log.info(" * Corrected reads are in " + support.process_spaces(os.path.dirname(corrected_dataset_yaml_filename) + "/"))
        if "assembly" in cfg and os.path.isfile(result_contigs_filename):
            message = " * Assembled contigs are in " + support.process_spaces(result_contigs_filename)
            if os.path.isfile(result_contigs_filename[:-6] + ".fastg"):
                message += " (" + os.path.basename(result_contigs_filename[:-6] + ".fastg") + ")"
            log.info(message)
        if "assembly" in cfg and os.path.isfile(result_scaffolds_filename):
            message = " * Assembled scaffolds are in " + support.process_spaces(result_scaffolds_filename)
            if os.path.isfile(result_scaffolds_filename[:-6] + ".fastg"):
                message += " (" + os.path.basename(result_scaffolds_filename[:-6] + ".fastg") + ")"
            log.info(message)
        #log.info("")

        #breaking scaffolds
        if os.path.isfile(result_scaffolds_filename):
            if not os.path.isdir(misc_dir):
                os.makedirs(misc_dir)
            result_broken_scaffolds = os.path.join(misc_dir, "broken_scaffolds.fasta")
            if not os.path.isfile(result_broken_scaffolds) or not options_storage.continue_mode:
                modified, broken_scaffolds = support.break_scaffolds(result_scaffolds_filename,
                    options_storage.THRESHOLD_FOR_BREAKING_SCAFFOLDS)
                if modified:
                    support.write_fasta(result_broken_scaffolds, broken_scaffolds)
                    #log.info(" * Scaffolds broken by " + str(options_storage.THRESHOLD_FOR_BREAKING_SCAFFOLDS) +
                    # " Ns are in " + result_broken_scaffolds)

        ### printing WARNINGS SUMMARY
        if not support.log_warnings(log):
            log.info("\n======= SPAdes pipeline finished.")  # otherwise it finished WITH WARNINGS

        if options_storage.test_mode:
            for result_filename in [result_contigs_filename, result_scaffolds_filename]:
                if os.path.isfile(result_filename):
                    result_fasta = list(support.read_fasta(result_filename))
                    # correctness check: should be one contig of length 1000 bp
                    correct_number = 1
                    correct_length = 1000
                    if not len(result_fasta):
                        support.error("TEST FAILED: %s does not contain contigs!" % result_filename)
                    elif len(result_fasta) > correct_number:
                        support.error("TEST FAILED: %s contains more than %d contig (%d)!" %
                                      (result_filename, correct_number, len(result_fasta)))
                    elif len(result_fasta[0][1]) != correct_length:
                        if len(result_fasta[0][1]) > correct_length:
                            relation = "more"
                        else:
                            relation = "less"
                        support.error("TEST FAILED: %s contains %s than %d bp (%d bp)!" %
                                      (result_filename, relation, correct_length, len(result_fasta[0][1])))
                else:
                    support.error("TEST FAILED: " + result_filename + " does not exist!")
            log.info("\n========= TEST PASSED CORRECTLY.")


        log.info("\nSPAdes log can be found here: " + log_filename)
        log.info("")
        log.info("Thank you for using SPAdes!")
        log.removeHandler(log_handler)

    except Exception:
        exc_type, exc_value, _ = sys.exc_info()
        if exc_type == SystemExit:
            sys.exit(exc_value)
        else:
            if exc_type == OSError and exc_value.errno == errno.ENOEXEC: # Exec format error
                support.error("It looks like you are using SPAdes binaries for another platform.\n" +
                              support.get_spades_binaries_info_message())
            else:
                log.exception(exc_value)
                support.error("exception caught: %s" % exc_type, log)
    except BaseException: # since python 2.5 system-exiting exceptions (e.g. KeyboardInterrupt) are derived from BaseException
        exc_type, exc_value, _ = sys.exc_info()
        if exc_type == SystemExit:
            sys.exit(exc_value)
        else:
            log.exception(exc_value)
            support.error("exception caught: %s" % exc_type, log)
Пример #3
0
def main():
    os.environ["LC_ALL"] = "C"

    if len(sys.argv) == 1:
        options_storage.usage(spades_version)
        sys.exit(0)

    log = logging.getLogger('spades')
    log.setLevel(logging.DEBUG)

    console = logging.StreamHandler(sys.stdout)
    console.setFormatter(logging.Formatter('%(message)s'))
    console.setLevel(logging.DEBUG)
    log.addHandler(console)

    check_binaries(bin_home, log)

    # parse options and safe all parameters to cfg
    cfg, dataset_data = fill_cfg(sys.argv, log)

    if options_storage.continue_mode:
        cmd_line, options = get_options_from_params(
            os.path.join(options_storage.output_dir, "params.txt"))
        if not options:
            support.error(
                "failed to parse command line of the previous run! Please restart from the beginning."
            )
        cfg, dataset_data = fill_cfg(options, log)
        options_storage.continue_mode = True

    log_filename = os.path.join(cfg["common"].output_dir, "spades.log")
    if options_storage.continue_mode:
        log_handler = logging.FileHandler(log_filename, mode='a')
    else:
        log_handler = logging.FileHandler(log_filename, mode='w')
    log.addHandler(log_handler)

    if options_storage.continue_mode:
        log.info(
            "\n======= SPAdes pipeline continued. Log can be found here: " +
            log_filename + "\n")
        log.info("Restored from " + cmd_line)
    else:
        params_filename = os.path.join(cfg["common"].output_dir, "params.txt")
        params_handler = logging.FileHandler(params_filename, mode='w')
        log.addHandler(params_handler)

        command = "Command line:"
        for v in sys.argv:
            command += " " + v
        log.info(command)

        print_used_values(cfg, log)
        log.removeHandler(params_handler)

        log.info("\n======= SPAdes pipeline started. Log can be found here: " +
                 log_filename + "\n")

    # splitting interlaced reads if needed
    if support.dataset_has_interlaced_reads(dataset_data):
        dir_for_split_reads = os.path.join(
            os.path.abspath(options_storage.output_dir), 'split_reads')
        if not os.path.isdir(dir_for_split_reads):
            os.makedirs(dir_for_split_reads)
        dataset_data = support.split_interlaced_reads(dataset_data,
                                                      dir_for_split_reads, log)
        options_storage.dataset_yaml_filename = os.path.join(
            options_storage.output_dir, "input_dataset.yaml")
        pyyaml.dump(dataset_data,
                    open(options_storage.dataset_yaml_filename, 'w'))
        cfg["dataset"].yaml_filename = os.path.abspath(
            options_storage.dataset_yaml_filename)

    try:
        # copying configs before all computations (to prevent its changing at run time)
        tmp_configs_dir = os.path.join(cfg["common"].output_dir, "configs")
        if os.path.isdir(
                tmp_configs_dir) and not options_storage.continue_mode:
            shutil.rmtree(tmp_configs_dir)
        if not os.path.isdir(tmp_configs_dir):
            shutil.copytree(os.path.join(spades_home, "configs"),
                            tmp_configs_dir)

        corrected_dataset_yaml_filename = ''
        if "error_correction" in cfg:
            bh_cfg = merge_configs(cfg["error_correction"], cfg["common"])
            bh_cfg.__dict__["dataset_yaml_filename"] = cfg[
                "dataset"].yaml_filename
            corrected_dataset_yaml_filename = os.path.join(
                bh_cfg.output_dir, "corrected.yaml")
            if os.path.isfile(corrected_dataset_yaml_filename
                              ) and options_storage.continue_mode:
                log.info(
                    "\n===== Skipping read error correction (already processed). \n"
                )
            else:
                options_storage.continue_mode = False  # continue from here

                if "HEAPCHECK" in os.environ:
                    del os.environ["HEAPCHECK"]
                if "heap_check" in bh_cfg.__dict__:
                    os.environ["HEAPCHECK"] = bh_cfg.heap_check

                if os.path.exists(bh_cfg.output_dir):
                    shutil.rmtree(bh_cfg.output_dir)

                os.makedirs(bh_cfg.output_dir)
                if not os.path.exists(bh_cfg.tmp_dir):
                    os.makedirs(bh_cfg.tmp_dir)

                log.info("\n===== Read error correction started. \n")
                bh_logic.run_bh(corrected_dataset_yaml_filename,
                                tmp_configs_dir, bin_home, bh_cfg,
                                ext_python_modules_home, log)
                log.info("\n===== Read error correction finished. \n")

        result_contigs_filename = os.path.join(cfg["common"].output_dir,
                                               "contigs.fasta")
        result_scaffolds_filename = os.path.join(cfg["common"].output_dir,
                                                 "scaffolds.fasta")
        misc_dir = os.path.join(cfg["common"].output_dir, "misc")
        ### if mismatch correction is enabled then result contigs are copied to misc directory
        assembled_contigs_filename = os.path.join(misc_dir,
                                                  "assembled_contigs.fasta")
        assembled_scaffolds_filename = os.path.join(
            misc_dir, "assembled_scaffolds.fasta")
        if "assembly" in cfg:
            spades_cfg = merge_configs(cfg["assembly"], cfg["common"])
            spades_cfg.__dict__["result_contigs"] = result_contigs_filename
            spades_cfg.__dict__["result_scaffolds"] = result_scaffolds_filename
            spades_cfg.__dict__["additional_contigs"] = os.path.join(
                spades_cfg.output_dir, "simplified_contigs.fasta")

            if options_storage.continue_mode and (
                    os.path.isfile(spades_cfg.result_contigs) or
                ("mismatch_corrector" in cfg
                 and os.path.isfile(assembled_contigs_filename))):
                log.info("\n===== Skipping assembling (already processed). \n")
                # calculating latest_dir for the next stages
                latest_dir = support.get_latest_dir(
                    os.path.join(spades_cfg.output_dir, "K*"))
                if not latest_dir:
                    support.error(
                        "failed to continue the previous run! Please restart from the beginning."
                    )
            else:
                if os.path.isfile(corrected_dataset_yaml_filename):
                    dataset_data = pyyaml.load(
                        open(corrected_dataset_yaml_filename, 'r'))
                    dataset_data = support.relative2abs_paths(
                        dataset_data,
                        os.path.dirname(corrected_dataset_yaml_filename))
                if support.dataset_has_paired_reads(dataset_data):
                    spades_cfg.__dict__["paired_mode"] = True
                else:
                    spades_cfg.__dict__["paired_mode"] = False

                if options_storage.rectangles:
                    spades_cfg.__dict__["resolving_mode"] = "rectangles"

                if "HEAPCHECK" in os.environ:
                    del os.environ["HEAPCHECK"]
                if "heap_check" in spades_cfg.__dict__:
                    os.environ["HEAPCHECK"] = spades_cfg.heap_check

                log.info("\n===== Assembling started.\n")

                # creating dataset
                dataset_filename = os.path.join(spades_cfg.output_dir,
                                                "dataset.info")
                if not os.path.isfile(
                        dataset_filename) or not options_storage.continue_mode:
                    dataset_file = open(dataset_filename, 'w')
                    import process_cfg
                    dataset_file.write(
                        "single_cell" + '\t' +
                        process_cfg.bool_to_str(cfg["dataset"].single_cell) +
                        '\n')
                    if os.path.isfile(corrected_dataset_yaml_filename):
                        dataset_file.write(
                            "reads" + '\t' + process_cfg.process_spaces(
                                corrected_dataset_yaml_filename) + '\n')
                    else:
                        dataset_file.write("reads" + '\t' +
                                           process_cfg.process_spaces(
                                               cfg["dataset"].yaml_filename) +
                                           '\n')
                    if spades_cfg.developer_mode and "reference" in cfg[
                            "dataset"].__dict__:
                        dataset_file.write("reference_genome" + '\t')
                        dataset_file.write(
                            process_cfg.process_spaces(
                                os.path.abspath(cfg["dataset"].reference)) +
                            '\n')
                    dataset_file.close()
                spades_cfg.__dict__["dataset"] = dataset_filename

                latest_dir = spades_logic.run_spades(tmp_configs_dir, bin_home,
                                                     spades_cfg, log)

                #rectangles
                if spades_cfg.paired_mode and options_storage.rectangles:
                    if options_storage.continue_mode:  # TODO: continue mode
                        support.warning(
                            "sorry, --continue doesn't work with --rectangles yet. Skipping repeat resolving."
                        )
                    else:
                        sys.path.append(
                            os.path.join(python_modules_home, "rectangles"))
                        import rrr

                        rrr_input_dir = os.path.join(latest_dir, "saves")
                        rrr_outpath = os.path.join(spades_cfg.output_dir,
                                                   "rectangles")
                        if not os.path.exists(rrr_outpath):
                            os.mkdir(rrr_outpath)

                        rrr_reference_information_file = os.path.join(
                            rrr_input_dir,
                            "late_pair_info_counted_etalon_distance.txt")
                        rrr_test_util = rrr.TestUtils(
                            rrr_reference_information_file,
                            os.path.join(rrr_outpath, "rectangles.log"))
                        rrr.resolve(rrr_input_dir, rrr_outpath, rrr_test_util,
                                    "", cfg["dataset"].single_cell,
                                    spades_cfg.careful)

                        shutil.copyfile(
                            os.path.join(
                                rrr_outpath,
                                "rectangles_extend_before_scaffold.fasta"),
                            spades_cfg.result_contigs)
                        shutil.copyfile(
                            os.path.join(rrr_outpath,
                                         "rectangles_extend.fasta"),
                            spades_cfg.result_scaffolds)

                        if not spades_cfg.developer_mode:
                            if os.path.exists(rrr_input_dir):
                                shutil.rmtree(rrr_input_dir)
                            if os.path.exists(rrr_outpath):
                                shutil.rmtree(rrr_outpath, True)
                            if os.path.exists(rrr_outpath):
                                os.system('rm -r ' + rrr_outpath)
                                #EOR

                if os.path.isdir(
                        misc_dir) and not options_storage.continue_mode:
                    shutil.rmtree(misc_dir)
                if not os.path.isdir(misc_dir):
                    os.makedirs(misc_dir)
                    if os.path.isfile(spades_cfg.additional_contigs):
                        shutil.move(spades_cfg.additional_contigs, misc_dir)

                log.info("\n===== Assembling finished. \n")

            #corrector
            if "mismatch_corrector" in cfg and (
                    os.path.isfile(result_contigs_filename) or
                (options_storage.continue_mode
                 and os.path.isfile(assembled_contigs_filename))):
                to_correct = dict()
                to_correct["contigs"] = (result_contigs_filename,
                                         assembled_contigs_filename)
                if os.path.isfile(result_scaffolds_filename) or (
                        options_storage.continue_mode
                        and os.path.isfile(assembled_scaffolds_filename)):
                    to_correct["scaffolds"] = (result_scaffolds_filename,
                                               assembled_scaffolds_filename)

                # moving assembled contigs (scaffolds) to misc dir
                for k, (old, new) in to_correct.items():
                    if options_storage.continue_mode and os.path.isfile(new):
                        continue
                    shutil.move(old, new)

                if options_storage.continue_mode and os.path.isfile(result_contigs_filename) and \
                    (os.path.isfile(result_scaffolds_filename) or not os.path.isfile(assembled_scaffolds_filename)):
                    log.info(
                        "\n===== Skipping mismatch correction (already processed). \n"
                    )
                else:
                    log.info("\n===== Mismatch correction started.")

                    # detecting paired-end library with the largest insert size
                    dataset_data = pyyaml.load(
                        open(options_storage.dataset_yaml_filename, 'r')
                    )  ### initial dataset, i.e. before error correction
                    dataset_data = support.relative2abs_paths(
                        dataset_data,
                        os.path.dirname(options_storage.dataset_yaml_filename))
                    paired_end_libraries_ids = []
                    for id, reads_library in enumerate(dataset_data):
                        if reads_library['type'] == 'paired-end':
                            paired_end_libraries_ids.append(id)
                    if not len(paired_end_libraries_ids):
                        support.error(
                            'Mismatch correction cannot be performed without at least one paired-end library!'
                        )
                    estimated_params = load_config_from_file(
                        os.path.join(latest_dir, "_est_params.info"))
                    max_insert_size = -1
                    target_paired_end_library_id = -1
                    for id in paired_end_libraries_ids:
                        if float(estimated_params.__dict__[
                                "insert_size_" + str(id)]) > max_insert_size:
                            max_insert_size = float(
                                estimated_params.__dict__["insert_size_" +
                                                          str(id)])
                            target_paired_end_library_id = id
                    yaml_dirname = os.path.dirname(
                        options_storage.dataset_yaml_filename)
                    cfg["mismatch_corrector"].__dict__["1"] = list(
                        map(
                            lambda x: os.path.join(yaml_dirname, x),
                            dataset_data[target_paired_end_library_id]
                            ['left reads']))
                    cfg["mismatch_corrector"].__dict__["2"] = list(
                        map(
                            lambda x: os.path.join(yaml_dirname, x),
                            dataset_data[target_paired_end_library_id]
                            ['right reads']))
                    cfg["mismatch_corrector"].__dict__["insert-size"] = round(
                        max_insert_size)
                    #TODO: add reads orientation

                    import corrector
                    corrector_cfg = cfg["mismatch_corrector"]
                    args = []
                    for key, values in corrector_cfg.__dict__.items():
                        if key == "output-dir":
                            continue

                        # for processing list of reads
                        if not isinstance(values, list):
                            values = [values]
                        for value in values:
                            if len(key) == 1:
                                args.append('-' + key)
                            else:
                                args.append('--' + key)
                            if value:
                                args.append(value)

                    # processing contigs and scaffolds (or only contigs)
                    for k, (corrected, assembled) in to_correct.items():
                        if options_storage.continue_mode and os.path.isfile(
                                corrected):
                            log.info("\n== Skipping processing of " + k +
                                     " (already processed)\n")
                            continue

                        options_storage.continue_mode = False
                        log.info("\n== Processing of " + k + "\n")

                        cur_args = args[:]
                        cur_args += ['-c', assembled]
                        tmp_dir_for_corrector = os.path.join(
                            corrector_cfg.__dict__["output-dir"],
                            "mismatch_corrector_" + k)
                        cur_args += ['--output-dir', tmp_dir_for_corrector]

                        # correcting
                        corrector.main(cur_args, ext_python_modules_home, log)

                        result_corrected_filename = os.path.abspath(
                            os.path.join(tmp_dir_for_corrector,
                                         "corrected_contigs.fasta"))
                        # moving corrected contigs (scaffolds) to SPAdes output dir
                        if os.path.isfile(result_corrected_filename):
                            shutil.move(result_corrected_filename, corrected)

                        if os.path.isdir(tmp_dir_for_corrector):
                            shutil.rmtree(tmp_dir_for_corrector)

                    log.info("\n===== Mismatch correction finished.\n")

        if not cfg["common"].developer_mode and os.path.isdir(tmp_configs_dir):
            shutil.rmtree(tmp_configs_dir)

        #log.info("")
        if os.path.isdir(os.path.dirname(corrected_dataset_yaml_filename)):
            log.info(" * Corrected reads are in " +
                     os.path.dirname(corrected_dataset_yaml_filename) + "/")
        if os.path.isfile(result_contigs_filename):
            log.info(" * Assembled contigs are in " + result_contigs_filename)
        if os.path.isfile(result_scaffolds_filename):
            log.info(" * Assembled scaffolds are in " +
                     result_scaffolds_filename)
        #log.info("")

        #breaking scaffolds
        if os.path.isfile(result_scaffolds_filename):
            if not os.path.isdir(misc_dir):
                os.makedirs(misc_dir)
            result_broken_scaffolds = os.path.join(misc_dir,
                                                   "broken_scaffolds.fasta")
            threshold = 3
            if not os.path.isfile(result_broken_scaffolds
                                  ) or not options_storage.continue_mode:
                support.break_scaffolds(result_scaffolds_filename, threshold,
                                        result_broken_scaffolds)
                #log.info(" * Scaffolds broken by " + str(threshold) + " Ns are in " + result_broken_scaffolds)

        ### printing WARNINGS SUMMARY
        if not support.log_warnings(log):
            log.info("\n======= SPAdes pipeline finished."
                     )  # otherwise it finished WITH WARNINGS

        log.info("\nSPAdes log can be found here: " + log_filename)
        log.info("")
        log.info("Thank you for using SPAdes!")
        log.removeHandler(log_handler)

    except Exception:
        _, exc, _ = sys.exc_info()
        log.exception(exc)
        support.error("exception caught", log)
Пример #4
0
def main(args):
    os.environ["LC_ALL"] = "C"

    if len(args) == 1:
        options_storage.usage(spades_version)
        sys.exit(0)

    log = logging.getLogger('spades')
    log.setLevel(logging.DEBUG)

    console = logging.StreamHandler(sys.stdout)
    console.setFormatter(logging.Formatter('%(message)s'))
    console.setLevel(logging.DEBUG)
    log.addHandler(console)

    support.check_binaries(bin_home, log)

    # parse options and safe all parameters to cfg
    options = args
    cfg, dataset_data = fill_cfg(options, log)

    if options_storage.continue_mode:
        cmd_line, options = get_options_from_params(
            os.path.join(options_storage.output_dir, "params.txt"), args[0])
        if not options:
            support.error(
                "failed to parse command line of the previous run! Please restart from the beginning or specify another output directory."
            )
        cfg, dataset_data = fill_cfg(options, log)
        if options_storage.restart_from:
            check_cfg_for_restart_from(cfg)
        options_storage.continue_mode = True

    log_filename = os.path.join(cfg["common"].output_dir, "spades.log")
    if options_storage.continue_mode:
        log_handler = logging.FileHandler(log_filename, mode='a')
    else:
        log_handler = logging.FileHandler(log_filename, mode='w')
    log.addHandler(log_handler)

    if options_storage.continue_mode:
        log.info(
            "\n======= SPAdes pipeline continued. Log can be found here: " +
            log_filename + "\n")
        log.info("Restored from " + cmd_line)
        if options_storage.restart_from:
            updated_params = ""
            flag = False
            for v in args[1:]:
                if v == '-o' or v == '--restart-from':
                    flag = True
                    continue
                if flag:
                    flag = False
                    continue
                updated_params += " " + v
            updated_params = updated_params.strip()
            log.info("with updated parameters: " + updated_params)
            cmd_line += " " + updated_params
        log.info("")

    params_filename = os.path.join(cfg["common"].output_dir, "params.txt")
    params_handler = logging.FileHandler(params_filename, mode='w')
    log.addHandler(params_handler)

    if options_storage.continue_mode:
        log.info(cmd_line)
    else:
        command = "Command line:"
        for v in args:
            command += " " + v
        log.info(command)

    # special case
    if "mismatch_corrector" in cfg and not support.get_lib_ids_by_type(
            dataset_data, 'paired-end'):
        support.warning(
            'cannot perform mismatch correction without at least one paired-end library! Skipping this step.',
            log)
        del cfg["mismatch_corrector"]

    print_used_values(cfg, log)
    log.removeHandler(params_handler)

    support.check_single_reads_in_options(options, log)

    if not options_storage.continue_mode:
        log.info("\n======= SPAdes pipeline started. Log can be found here: " +
                 log_filename + "\n")

    # splitting interlaced reads and processing Ns in additional contigs if needed
    if support.dataset_has_interlaced_reads(
            dataset_data) or support.dataset_has_additional_contigs(
                dataset_data):
        dir_for_split_reads = os.path.join(options_storage.output_dir,
                                           'split_input')
        if support.dataset_has_interlaced_reads(dataset_data):
            if not os.path.isdir(dir_for_split_reads):
                os.makedirs(dir_for_split_reads)
            dataset_data = support.split_interlaced_reads(
                dataset_data, dir_for_split_reads, log)
        if support.dataset_has_additional_contigs(dataset_data):
            dataset_data = support.process_Ns_in_additional_contigs(
                dataset_data, dir_for_split_reads, log)
        options_storage.dataset_yaml_filename = os.path.join(
            options_storage.output_dir, "input_dataset.yaml")
        pyyaml.dump(dataset_data,
                    open(options_storage.dataset_yaml_filename, 'w'))
        cfg["dataset"].yaml_filename = options_storage.dataset_yaml_filename

    try:
        # copying configs before all computations (to prevent its changing at run time)
        tmp_configs_dir = os.path.join(cfg["common"].output_dir, "configs")
        if os.path.isdir(
                tmp_configs_dir) and not options_storage.continue_mode:
            shutil.rmtree(tmp_configs_dir)
        if not os.path.isdir(tmp_configs_dir):
            dir_util.copy_tree(os.path.join(spades_home, "configs"),
                               tmp_configs_dir,
                               preserve_times=False)

        corrected_dataset_yaml_filename = ''
        if "error_correction" in cfg:
            STAGE_NAME = "Read error correction"
            bh_cfg = merge_configs(cfg["error_correction"], cfg["common"])
            corrected_dataset_yaml_filename = os.path.join(
                bh_cfg.output_dir, "corrected.yaml")
            if os.path.isfile(corrected_dataset_yaml_filename) and options_storage.continue_mode \
                and not options_storage.restart_from == "ec":
                log.info("\n===== Skipping %s (already processed). \n" %
                         STAGE_NAME)
            else:
                support.continue_from_here(log)

                if "HEAPCHECK" in os.environ:
                    del os.environ["HEAPCHECK"]
                if "heap_check" in bh_cfg.__dict__:
                    os.environ["HEAPCHECK"] = bh_cfg.heap_check

                if os.path.exists(bh_cfg.output_dir):
                    shutil.rmtree(bh_cfg.output_dir)
                os.makedirs(bh_cfg.output_dir)

                if support.get_lib_ids_by_type(
                        dataset_data, options_storage.LONG_READS_TYPES):
                    not_used_dataset_data = support.get_libs_by_type(
                        dataset_data, options_storage.LONG_READS_TYPES)
                    to_correct_dataset_data = support.rm_libs_by_type(
                        dataset_data, options_storage.LONG_READS_TYPES)
                    to_correct_dataset_yaml_filename = os.path.join(
                        bh_cfg.output_dir, "to_correct.yaml")
                    pyyaml.dump(to_correct_dataset_data,
                                open(to_correct_dataset_yaml_filename, 'w'))
                    bh_cfg.__dict__[
                        "dataset_yaml_filename"] = to_correct_dataset_yaml_filename
                else:
                    not_used_dataset_data = None
                    bh_cfg.__dict__["dataset_yaml_filename"] = cfg[
                        "dataset"].yaml_filename

                log.info("\n===== %s started. \n" % STAGE_NAME)
                hammer_logic.run_hammer(corrected_dataset_yaml_filename,
                                        tmp_configs_dir, bin_home, bh_cfg,
                                        not_used_dataset_data,
                                        ext_python_modules_home, log)
                log.info("\n===== %s finished. \n" % STAGE_NAME)

        result_contigs_filename = os.path.join(cfg["common"].output_dir,
                                               "contigs.fasta")
        result_scaffolds_filename = os.path.join(cfg["common"].output_dir,
                                                 "scaffolds.fasta")
        misc_dir = os.path.join(cfg["common"].output_dir, "misc")
        ### if mismatch correction is enabled then result contigs are copied to misc directory
        assembled_contigs_filename = os.path.join(misc_dir,
                                                  "assembled_contigs.fasta")
        assembled_scaffolds_filename = os.path.join(
            misc_dir, "assembled_scaffolds.fasta")
        if "assembly" in cfg:
            STAGE_NAME = "Assembling"
            spades_cfg = merge_configs(cfg["assembly"], cfg["common"])
            spades_cfg.__dict__["result_contigs"] = result_contigs_filename
            spades_cfg.__dict__["result_scaffolds"] = result_scaffolds_filename

            if options_storage.continue_mode and (os.path.isfile(spades_cfg.result_contigs)
                                                  or ("mismatch_corrector" in cfg and
                                                      os.path.isfile(assembled_contigs_filename)))\
                and not options_storage.restart_from == 'as' \
                and not (options_storage.restart_from and options_storage.restart_from.startswith('k')):

                log.info("\n===== Skipping %s (already processed). \n" %
                         STAGE_NAME)
                # calculating latest_dir for the next stages
                latest_dir = support.get_latest_dir(
                    os.path.join(spades_cfg.output_dir, "K*"))
                if not latest_dir:
                    support.error(
                        "failed to continue the previous run! Please restart from previous stages or from the beginning.",
                        log)
            else:
                old_result_files = [
                    result_contigs_filename, result_scaffolds_filename,
                    assembled_contigs_filename, assembled_scaffolds_filename
                ]
                for format in [".fasta", ".fastg"]:
                    for old_result_file in old_result_files:
                        if os.path.isfile(old_result_file[:-6] + format):
                            os.remove(old_result_file[:-6] + format)

                if options_storage.restart_from == 'as':
                    support.continue_from_here(log)

                if os.path.isfile(corrected_dataset_yaml_filename):
                    dataset_data = pyyaml.load(
                        open(corrected_dataset_yaml_filename, 'r'))
                    dataset_data = support.relative2abs_paths(
                        dataset_data,
                        os.path.dirname(corrected_dataset_yaml_filename))
                if spades_cfg.disable_rr:
                    spades_cfg.__dict__["rr_enable"] = False
                else:
                    spades_cfg.__dict__["rr_enable"] = True

                if "HEAPCHECK" in os.environ:
                    del os.environ["HEAPCHECK"]
                if "heap_check" in spades_cfg.__dict__:
                    os.environ["HEAPCHECK"] = spades_cfg.heap_check

                log.info("\n===== %s started.\n" % STAGE_NAME)

                # creating dataset
                dataset_filename = os.path.join(spades_cfg.output_dir,
                                                "dataset.info")
                if not os.path.isfile(
                        dataset_filename) or not options_storage.continue_mode:
                    dataset_file = open(dataset_filename, 'w')
                    import process_cfg
                    dataset_file.write(
                        "single_cell" + '\t' +
                        process_cfg.bool_to_str(cfg["dataset"].single_cell) +
                        '\n')
                    if os.path.isfile(corrected_dataset_yaml_filename):
                        dataset_file.write(
                            "reads" + '\t' + process_cfg.process_spaces(
                                corrected_dataset_yaml_filename) + '\n')
                    else:
                        dataset_file.write("reads" + '\t' +
                                           process_cfg.process_spaces(
                                               cfg["dataset"].yaml_filename) +
                                           '\n')
                    if spades_cfg.developer_mode and "reference" in cfg[
                            "dataset"].__dict__:
                        dataset_file.write("reference_genome" + '\t')
                        dataset_file.write(
                            process_cfg.process_spaces(
                                cfg["dataset"].reference) + '\n')
                    dataset_file.close()
                spades_cfg.__dict__["dataset"] = dataset_filename

                latest_dir = spades_logic.run_spades(tmp_configs_dir, bin_home,
                                                     spades_cfg, dataset_data,
                                                     ext_python_modules_home,
                                                     log)

                if os.path.isdir(
                        misc_dir) and not options_storage.continue_mode:
                    shutil.rmtree(misc_dir)
                if not os.path.isdir(misc_dir):
                    os.makedirs(misc_dir)

                if options_storage.continue_mode and options_storage.restart_from and options_storage.restart_from.startswith(
                        'k'):
                    k_str = options_storage.restart_from[1:]
                    if k_str.find(":") != -1:
                        k_str = k_str[:k_str.find(":")]
                    support.error(
                        "failed to continue from K=%s because this K was not processed in the original run!"
                        % k_str, log)
                log.info("\n===== %s finished. \n" % STAGE_NAME)

            #corrector
            if "mismatch_corrector" in cfg and (
                    os.path.isfile(result_contigs_filename) or
                (options_storage.continue_mode
                 and os.path.isfile(assembled_contigs_filename))):
                STAGE_NAME = "Mismatch correction"
                to_correct = dict()
                to_correct["contigs"] = (result_contigs_filename,
                                         assembled_contigs_filename)
                if os.path.isfile(result_scaffolds_filename) or (
                        options_storage.continue_mode
                        and os.path.isfile(assembled_scaffolds_filename)):
                    to_correct["scaffolds"] = (result_scaffolds_filename,
                                               assembled_scaffolds_filename)

                # moving assembled contigs (scaffolds) to misc dir
                for assembly_type, (old, new) in to_correct.items():
                    if options_storage.continue_mode and os.path.isfile(new):
                        continue
                    for format in [".fasta", ".fastg"]:
                        if os.path.isfile(old[:-6] + format):
                            shutil.move(old[:-6] + format, new[:-6] + format)

                if options_storage.continue_mode and os.path.isfile(result_contigs_filename) and \
                    (os.path.isfile(result_scaffolds_filename) or not os.path.isfile(assembled_scaffolds_filename)) \
                    and not options_storage.restart_from == 'mc':
                    log.info("\n===== Skipping %s (already processed). \n" %
                             STAGE_NAME)
                else:
                    if options_storage.restart_from == 'mc':
                        support.continue_from_here(log)

                    log.info("\n===== %s started." % STAGE_NAME)
                    # detecting paired-end library with the largest insert size
                    est_params_data = pyyaml.load(
                        open(os.path.join(latest_dir, "final.lib_data"), 'r'))
                    max_IS_library = None
                    for reads_library in est_params_data:
                        if reads_library['type'] == 'paired-end':
                            if not max_IS_library or float(
                                    reads_library["insert size mean"]) > float(
                                        max_IS_library["insert size mean"]):
                                max_IS_library = reads_library
                    if not max_IS_library:
                        support.error(
                            'Mismatch correction cannot be performed without at least one paired-end library!',
                            log)
                    if not max_IS_library["insert size mean"]:
                        support.warning(
                            'Failed to estimate insert size for all paired-end libraries. Starting Mismatch correction'
                            ' based on the first paired-end library and with default insert size.',
                            log)
                    else:
                        cfg["mismatch_corrector"].__dict__[
                            "insert-size"] = round(
                                max_IS_library["insert size mean"])
                    yaml_dirname = os.path.dirname(
                        options_storage.dataset_yaml_filename)
                    cfg["mismatch_corrector"].__dict__["1"] = list(
                        map(lambda x: os.path.join(yaml_dirname, x),
                            max_IS_library['left reads']))
                    cfg["mismatch_corrector"].__dict__["2"] = list(
                        map(lambda x: os.path.join(yaml_dirname, x),
                            max_IS_library['right reads']))
                    #TODO: add reads orientation

                    import corrector
                    corrector_cfg = cfg["mismatch_corrector"]
                    args = []
                    for key, values in corrector_cfg.__dict__.items():
                        if key == "output-dir":
                            continue

                        # for processing list of reads
                        if not isinstance(values, list):
                            values = [values]
                        for value in values:
                            if len(key) == 1:
                                args.append('-' + key)
                            else:
                                args.append('--' + key)
                            if value is not None:
                                args.append(value)

                    # processing contigs and scaffolds (or only contigs)
                    for assembly_type, (corrected,
                                        assembled) in to_correct.items():
                        if options_storage.continue_mode and os.path.isfile(
                                corrected):
                            log.info("\n== Skipping processing of " +
                                     assembly_type + " (already processed)\n")
                            continue

                        support.continue_from_here(log)
                        log.info("\n== Processing of " + assembly_type + "\n")

                        cur_args = args[:]
                        cur_args += ['-c', assembled]
                        tmp_dir_for_corrector = support.get_tmp_dir(
                            prefix="mis_cor_%s_" % assembly_type)
                        cur_args += ['--output-dir', tmp_dir_for_corrector]

                        # correcting
                        corrector.main(cur_args, ext_python_modules_home, log)

                        result_corrected_filename = os.path.join(
                            tmp_dir_for_corrector, "corrected_contigs.fasta")
                        # moving corrected contigs (scaffolds) to SPAdes output dir
                        if os.path.isfile(result_corrected_filename):
                            shutil.move(result_corrected_filename, corrected)

                        if os.path.isdir(tmp_dir_for_corrector):
                            shutil.rmtree(tmp_dir_for_corrector)

                        assembled_fastg = assembled[:-6] + ".fastg"
                        if os.path.isfile(assembled_fastg):
                            support.create_fastg_from_fasta(
                                corrected, assembled_fastg, log)
                    log.info("\n===== %s finished.\n" % STAGE_NAME)

        if not cfg["common"].developer_mode and os.path.isdir(tmp_configs_dir):
            shutil.rmtree(tmp_configs_dir)

        #log.info("")
        if "error_correction" in cfg and os.path.isdir(
                os.path.dirname(corrected_dataset_yaml_filename)):
            log.info(" * Corrected reads are in " + support.process_spaces(
                os.path.dirname(corrected_dataset_yaml_filename) + "/"))
        if "assembly" in cfg and os.path.isfile(result_contigs_filename):
            message = " * Assembled contigs are in " + support.process_spaces(
                result_contigs_filename)
            if os.path.isfile(result_contigs_filename[:-6] + ".fastg"):
                message += " (" + os.path.basename(
                    result_contigs_filename[:-6] + ".fastg") + ")"
            log.info(message)
        if "assembly" in cfg and os.path.isfile(result_scaffolds_filename):
            message = " * Assembled scaffolds are in " + support.process_spaces(
                result_scaffolds_filename)
            if os.path.isfile(result_scaffolds_filename[:-6] + ".fastg"):
                message += " (" + os.path.basename(
                    result_scaffolds_filename[:-6] + ".fastg") + ")"
            log.info(message)
        #log.info("")

        #breaking scaffolds
        if os.path.isfile(result_scaffolds_filename):
            if not os.path.isdir(misc_dir):
                os.makedirs(misc_dir)
            result_broken_scaffolds = os.path.join(misc_dir,
                                                   "broken_scaffolds.fasta")
            if not os.path.isfile(result_broken_scaffolds
                                  ) or not options_storage.continue_mode:
                modified, broken_scaffolds = support.break_scaffolds(
                    result_scaffolds_filename,
                    options_storage.THRESHOLD_FOR_BREAKING_SCAFFOLDS)
                if modified:
                    support.write_fasta(result_broken_scaffolds,
                                        broken_scaffolds)
                    #log.info(" * Scaffolds broken by " + str(options_storage.THRESHOLD_FOR_BREAKING_SCAFFOLDS) +
                    # " Ns are in " + result_broken_scaffolds)

        ### printing WARNINGS SUMMARY
        if not support.log_warnings(log):
            log.info("\n======= SPAdes pipeline finished."
                     )  # otherwise it finished WITH WARNINGS

        if options_storage.test_mode:
            for result_filename in [
                    result_contigs_filename, result_scaffolds_filename
            ]:
                if os.path.isfile(result_filename):
                    result_fasta = list(support.read_fasta(result_filename))
                    # correctness check: should be one contig of length 1000 bp
                    correct_number = 1
                    correct_length = 1000
                    if not len(result_fasta):
                        support.error(
                            "TEST FAILED: %s does not contain contigs!" %
                            result_filename)
                    elif len(result_fasta) > correct_number:
                        support.error(
                            "TEST FAILED: %s contains more than %d contig (%d)!"
                            % (result_filename, correct_number,
                               len(result_fasta)))
                    elif len(result_fasta[0][1]) != correct_length:
                        if len(result_fasta[0][1]) > correct_length:
                            relation = "more"
                        else:
                            relation = "less"
                        support.error(
                            "TEST FAILED: %s contains %s than %d bp (%d bp)!" %
                            (result_filename, relation, correct_length,
                             len(result_fasta[0][1])))
                else:
                    support.error("TEST FAILED: " + result_filename +
                                  " does not exist!")
            log.info("\n========= TEST PASSED CORRECTLY.")

        log.info("\nSPAdes log can be found here: " + log_filename)
        log.info("")
        log.info("Thank you for using SPAdes!")
        log.removeHandler(log_handler)

    except Exception:
        exc_type, exc_value, _ = sys.exc_info()
        if exc_type == SystemExit:
            sys.exit(exc_value)
        else:
            if exc_type == OSError and exc_value.errno == errno.ENOEXEC:  # Exec format error
                support.error(
                    "It looks like you are using SPAdes binaries for another platform.\n"
                    + support.get_spades_binaries_info_message())
            else:
                log.exception(exc_value)
                support.error("exception caught: %s" % exc_type, log)
    except BaseException:  # since python 2.5 system-exiting exceptions (e.g. KeyboardInterrupt) are derived from BaseException
        exc_type, exc_value, _ = sys.exc_info()
        if exc_type == SystemExit:
            sys.exit(exc_value)
        else:
            log.exception(exc_value)
            support.error("exception caught: %s" % exc_type, log)
Пример #5
0
                        if value:
                            args.append(value)

                # processing contigs and scaffolds (or only contigs)
                for k, (corrected, assembled) in to_correct.items():
                    log.info("\n== Processing " + k + "\n")

                    cur_args = args[:]
                    cur_args += ["-c", assembled]
                    tmp_dir_for_corrector = os.path.join(
                        corrector_cfg.__dict__["output-dir"], "mismatch_corrector_" + k
                    )
                    cur_args += ["--output-dir", tmp_dir_for_corrector]

                    # correcting
                    corrector.main(cur_args, ext_python_modules_home, log)

                    result_corrected_filename = os.path.abspath(
                        os.path.join(tmp_dir_for_corrector, "corrected_contigs.fasta")
                    )
                    # moving corrected contigs (scaffolds) to SPAdes output dir
                    if os.path.isfile(result_corrected_filename):
                        shutil.move(result_corrected_filename, corrected)

                    if os.path.isdir(tmp_dir_for_corrector):
                        shutil.rmtree(tmp_dir_for_corrector)

                log.info("\n===== Mismatch correction finished.\n")

        if not cfg["common"].developer_mode and os.path.isdir(tmp_configs_dir):
            shutil.rmtree(tmp_configs_dir)
Пример #6
0
def main():
    os.environ["LC_ALL"] = "C"

    if len(sys.argv) == 1:
        options_storage.usage(spades_version)
        sys.exit(0)

    log = logging.getLogger('spades')
    log.setLevel(logging.DEBUG)

    console = logging.StreamHandler(sys.stdout)
    console.setFormatter(logging.Formatter('%(message)s'))
    console.setLevel(logging.DEBUG)
    log.addHandler(console)

    check_binaries(bin_home, log)

    # parse options and safe all parameters to cfg
    cfg, dataset_data = fill_cfg(sys.argv, log)

    if options_storage.continue_mode:
        cmd_line, options = get_options_from_params(os.path.join(options_storage.output_dir, "params.txt"))
        if not options:
            support.error("failed to parse command line of the previous run! Please restart from the beginning.")
        cfg, dataset_data = fill_cfg(options, log)
        options_storage.continue_mode = True

    log_filename = os.path.join(cfg["common"].output_dir, "spades.log")
    if options_storage.continue_mode:
        log_handler = logging.FileHandler(log_filename, mode='a')
    else:
        log_handler = logging.FileHandler(log_filename, mode='w')
    log.addHandler(log_handler)

    if options_storage.continue_mode:
        log.info("\n======= SPAdes pipeline continued. Log can be found here: " + log_filename + "\n")
        log.info("Restored from " + cmd_line)
    else:
        params_filename = os.path.join(cfg["common"].output_dir, "params.txt")
        params_handler = logging.FileHandler(params_filename, mode='w')
        log.addHandler(params_handler)

        command = "Command line:"
        for v in sys.argv:
            command += " " + v
        log.info(command)

        print_used_values(cfg, log)
        log.removeHandler(params_handler)

        log.info("\n======= SPAdes pipeline started. Log can be found here: " + log_filename + "\n")

    # splitting interlaced reads if needed
    if support.dataset_has_interlaced_reads(dataset_data):
        dir_for_split_reads = os.path.join(os.path.abspath(options_storage.output_dir), 'split_reads')
        if not os.path.isdir(dir_for_split_reads):
            os.makedirs(dir_for_split_reads)
        dataset_data = support.split_interlaced_reads(dataset_data, dir_for_split_reads, log)
        options_storage.dataset_yaml_filename = os.path.join(options_storage.output_dir, "input_dataset.yaml")
        pyyaml.dump(dataset_data, open(options_storage.dataset_yaml_filename, 'w'))
        cfg["dataset"].yaml_filename = os.path.abspath(options_storage.dataset_yaml_filename)

    try:
        # copying configs before all computations (to prevent its changing at run time)
        tmp_configs_dir = os.path.join(cfg["common"].output_dir, "configs")
        if os.path.isdir(tmp_configs_dir) and not options_storage.continue_mode:
            shutil.rmtree(tmp_configs_dir)
        if not os.path.isdir(tmp_configs_dir):
            shutil.copytree(os.path.join(spades_home, "configs"), tmp_configs_dir)

        corrected_dataset_yaml_filename = ''
        if "error_correction" in cfg:
            bh_cfg = merge_configs(cfg["error_correction"], cfg["common"])
            bh_cfg.__dict__["dataset_yaml_filename"] = cfg["dataset"].yaml_filename
            corrected_dataset_yaml_filename = os.path.join(bh_cfg.output_dir, "corrected.yaml")
            if os.path.isfile(corrected_dataset_yaml_filename) and options_storage.continue_mode:
                log.info("\n===== Skipping read error correction (already processed). \n")
            else:
                options_storage.continue_mode = False # continue from here

                if "HEAPCHECK" in os.environ:
                    del os.environ["HEAPCHECK"]
                if "heap_check" in bh_cfg.__dict__:
                    os.environ["HEAPCHECK"] = bh_cfg.heap_check

                if os.path.exists(bh_cfg.output_dir):
                    shutil.rmtree(bh_cfg.output_dir)

                os.makedirs(bh_cfg.output_dir)
                if not os.path.exists(bh_cfg.tmp_dir):
                    os.makedirs(bh_cfg.tmp_dir)

                log.info("\n===== Read error correction started. \n")
                bh_logic.run_bh(corrected_dataset_yaml_filename, tmp_configs_dir, bin_home, bh_cfg,
                    ext_python_modules_home, log)
                log.info("\n===== Read error correction finished. \n")

        result_contigs_filename = os.path.join(cfg["common"].output_dir, "contigs.fasta")
        result_scaffolds_filename = os.path.join(cfg["common"].output_dir, "scaffolds.fasta")
        misc_dir = os.path.join(cfg["common"].output_dir, "misc")
        ### if mismatch correction is enabled then result contigs are copied to misc directory
        assembled_contigs_filename = os.path.join(misc_dir, "assembled_contigs.fasta")
        assembled_scaffolds_filename = os.path.join(misc_dir, "assembled_scaffolds.fasta")
        if "assembly" in cfg:
            spades_cfg = merge_configs(cfg["assembly"], cfg["common"])
            spades_cfg.__dict__["result_contigs"] = result_contigs_filename
            spades_cfg.__dict__["result_scaffolds"] = result_scaffolds_filename
            spades_cfg.__dict__["additional_contigs"] = os.path.join(spades_cfg.output_dir, "simplified_contigs.fasta")

            if options_storage.continue_mode and (os.path.isfile(spades_cfg.result_contigs)
                                                  or ("mismatch_corrector" in cfg and
                                                      os.path.isfile(assembled_contigs_filename))):
                log.info("\n===== Skipping assembling (already processed). \n")
                # calculating latest_dir for the next stages
                latest_dir = support.get_latest_dir(os.path.join(spades_cfg.output_dir, "K*"))
                if not latest_dir:
                    support.error("failed to continue the previous run! Please restart from the beginning.")
            else:
                if os.path.isfile(corrected_dataset_yaml_filename):
                    dataset_data = pyyaml.load(open(corrected_dataset_yaml_filename, 'r'))
                    dataset_data = support.relative2abs_paths(dataset_data, os.path.dirname(corrected_dataset_yaml_filename))
                if support.dataset_has_paired_reads(dataset_data):
                    spades_cfg.__dict__["paired_mode"] = True
                else:
                    spades_cfg.__dict__["paired_mode"] = False

                if options_storage.rectangles:
                    spades_cfg.__dict__["resolving_mode"] = "rectangles"

                if "HEAPCHECK" in os.environ:
                    del os.environ["HEAPCHECK"]
                if "heap_check" in spades_cfg.__dict__:
                    os.environ["HEAPCHECK"] = spades_cfg.heap_check

                log.info("\n===== Assembling started.\n")

                # creating dataset
                dataset_filename = os.path.join(spades_cfg.output_dir, "dataset.info")
                if not os.path.isfile(dataset_filename) or not options_storage.continue_mode:
                    dataset_file = open(dataset_filename, 'w')
                    import process_cfg
                    dataset_file.write("single_cell" + '\t' + process_cfg.bool_to_str(cfg["dataset"].single_cell) + '\n')
                    if os.path.isfile(corrected_dataset_yaml_filename):
                        dataset_file.write("reads" + '\t' + process_cfg.process_spaces(corrected_dataset_yaml_filename) + '\n')
                    else:
                        dataset_file.write("reads" + '\t' + process_cfg.process_spaces(cfg["dataset"].yaml_filename) + '\n')
                    if spades_cfg.developer_mode and "reference" in cfg["dataset"].__dict__:
                        dataset_file.write("reference_genome" + '\t')
                        dataset_file.write(process_cfg.process_spaces(os.path.abspath(cfg["dataset"].reference)) + '\n')
                    dataset_file.close()
                spades_cfg.__dict__["dataset"] = dataset_filename

                latest_dir = spades_logic.run_spades(tmp_configs_dir, bin_home, spades_cfg, log)

                #rectangles
                if spades_cfg.paired_mode and options_storage.rectangles:
                    if options_storage.continue_mode: # TODO: continue mode
                        support.warning("sorry, --continue doesn't work with --rectangles yet. Skipping repeat resolving.")
                    else:
                        sys.path.append(os.path.join(python_modules_home, "rectangles"))
                        import rrr

                        rrr_input_dir = os.path.join(latest_dir, "saves")
                        rrr_outpath = os.path.join(spades_cfg.output_dir, "rectangles")
                        if not os.path.exists(rrr_outpath):
                            os.mkdir(rrr_outpath)

                        rrr_reference_information_file = os.path.join(rrr_input_dir,
                            "late_pair_info_counted_etalon_distance.txt")
                        rrr_test_util = rrr.TestUtils(rrr_reference_information_file,
                            os.path.join(rrr_outpath, "rectangles.log"))
                        rrr.resolve(rrr_input_dir, rrr_outpath, rrr_test_util, "", cfg["dataset"].single_cell, spades_cfg.careful)

                        shutil.copyfile(os.path.join(rrr_outpath, "rectangles_extend_before_scaffold.fasta"), spades_cfg.result_contigs)
                        shutil.copyfile(os.path.join(rrr_outpath, "rectangles_extend.fasta"), spades_cfg.result_scaffolds)

                        if not spades_cfg.developer_mode:
                            if os.path.exists(rrr_input_dir):
                                shutil.rmtree(rrr_input_dir)
                            if os.path.exists(rrr_outpath):
                                shutil.rmtree(rrr_outpath, True)
                            if os.path.exists(rrr_outpath):
                                os.system('rm -r ' + rrr_outpath)
                                #EOR

                if os.path.isdir(misc_dir) and not options_storage.continue_mode:
                    shutil.rmtree(misc_dir)
                if not os.path.isdir(misc_dir):
                    os.makedirs(misc_dir)
                    if os.path.isfile(spades_cfg.additional_contigs):
                        shutil.move(spades_cfg.additional_contigs, misc_dir)

                log.info("\n===== Assembling finished. \n")

            #corrector
            if "mismatch_corrector" in cfg and (os.path.isfile(result_contigs_filename) or
                                                (options_storage.continue_mode and os.path.isfile(assembled_contigs_filename))):
                to_correct = dict()
                to_correct["contigs"] = (result_contigs_filename, assembled_contigs_filename)
                if os.path.isfile(result_scaffolds_filename) or (options_storage.continue_mode and
                                                                 os.path.isfile(assembled_scaffolds_filename)):
                    to_correct["scaffolds"] = (result_scaffolds_filename, assembled_scaffolds_filename)

                # moving assembled contigs (scaffolds) to misc dir
                for k, (old, new) in to_correct.items():
                    if options_storage.continue_mode and os.path.isfile(new):
                        continue
                    shutil.move(old, new)

                if options_storage.continue_mode and os.path.isfile(result_contigs_filename) and \
                    (os.path.isfile(result_scaffolds_filename) or not os.path.isfile(assembled_scaffolds_filename)):
                    log.info("\n===== Skipping mismatch correction (already processed). \n")
                else:
                    log.info("\n===== Mismatch correction started.")

                    # detecting paired-end library with the largest insert size
                    dataset_data = pyyaml.load(open(options_storage.dataset_yaml_filename, 'r')) ### initial dataset, i.e. before error correction
                    dataset_data = support.relative2abs_paths(dataset_data, os.path.dirname(options_storage.dataset_yaml_filename))
                    paired_end_libraries_ids = []
                    for id, reads_library in enumerate(dataset_data):
                        if reads_library['type'] == 'paired-end':
                            paired_end_libraries_ids.append(id)
                    if not len(paired_end_libraries_ids):
                        support.error('Mismatch correction cannot be performed without at least one paired-end library!')
                    estimated_params = load_config_from_file(os.path.join(latest_dir, "_est_params.info"))
                    max_insert_size = -1
                    target_paired_end_library_id = -1
                    for id in paired_end_libraries_ids:
                        if float(estimated_params.__dict__["insert_size_" + str(id)]) > max_insert_size:
                            max_insert_size = float(estimated_params.__dict__["insert_size_" + str(id)])
                            target_paired_end_library_id = id
                    yaml_dirname = os.path.dirname(options_storage.dataset_yaml_filename)
                    cfg["mismatch_corrector"].__dict__["1"] = list(map(lambda x: os.path.join(yaml_dirname, x),
                        dataset_data[target_paired_end_library_id]['left reads']))
                    cfg["mismatch_corrector"].__dict__["2"] = list(map(lambda x: os.path.join(yaml_dirname, x),
                        dataset_data[target_paired_end_library_id]['right reads']))
                    cfg["mismatch_corrector"].__dict__["insert-size"] = round(max_insert_size)
                    #TODO: add reads orientation

                    import corrector
                    corrector_cfg = cfg["mismatch_corrector"]
                    args = []
                    for key, values in corrector_cfg.__dict__.items():
                        if key == "output-dir":
                            continue

                        # for processing list of reads
                        if not isinstance(values, list):
                            values = [values]
                        for value in values:
                            if len(key) == 1:
                                args.append('-' + key)
                            else:
                                args.append('--' + key)
                            if value:
                                args.append(value)

                    # processing contigs and scaffolds (or only contigs)
                    for k, (corrected, assembled) in to_correct.items():
                        if options_storage.continue_mode and os.path.isfile(corrected):
                            log.info("\n== Skipping processing of " + k + " (already processed)\n")
                            continue

                        options_storage.continue_mode = False
                        log.info("\n== Processing of " + k + "\n")

                        cur_args = args[:]
                        cur_args += ['-c', assembled]
                        tmp_dir_for_corrector = os.path.join(corrector_cfg.__dict__["output-dir"], "mismatch_corrector_" + k)
                        cur_args += ['--output-dir', tmp_dir_for_corrector]

                        # correcting
                        corrector.main(cur_args, ext_python_modules_home, log)

                        result_corrected_filename = os.path.abspath(os.path.join(tmp_dir_for_corrector, "corrected_contigs.fasta"))
                        # moving corrected contigs (scaffolds) to SPAdes output dir
                        if os.path.isfile(result_corrected_filename):
                            shutil.move(result_corrected_filename, corrected)

                        if os.path.isdir(tmp_dir_for_corrector):
                            shutil.rmtree(tmp_dir_for_corrector)

                    log.info("\n===== Mismatch correction finished.\n")

        if not cfg["common"].developer_mode and os.path.isdir(tmp_configs_dir):
            shutil.rmtree(tmp_configs_dir)

        #log.info("")
        if os.path.isdir(os.path.dirname(corrected_dataset_yaml_filename)):
            log.info(" * Corrected reads are in " + os.path.dirname(corrected_dataset_yaml_filename) + "/")
        if os.path.isfile(result_contigs_filename):
            log.info(" * Assembled contigs are in " + result_contigs_filename)
        if os.path.isfile(result_scaffolds_filename):
            log.info(" * Assembled scaffolds are in " + result_scaffolds_filename)
        #log.info("")

        #breaking scaffolds
        if os.path.isfile(result_scaffolds_filename):
            if not os.path.isdir(misc_dir):
                os.makedirs(misc_dir)
            result_broken_scaffolds = os.path.join(misc_dir, "broken_scaffolds.fasta")
            threshold = 3
            if not os.path.isfile(result_broken_scaffolds) or not options_storage.continue_mode:
                support.break_scaffolds(result_scaffolds_filename, threshold, result_broken_scaffolds)
                #log.info(" * Scaffolds broken by " + str(threshold) + " Ns are in " + result_broken_scaffolds)

        ### printing WARNINGS SUMMARY
        if not support.log_warnings(log):
            log.info("\n======= SPAdes pipeline finished.")  # otherwise it finished WITH WARNINGS

        log.info("\nSPAdes log can be found here: " + log_filename)
        log.info("")
        log.info("Thank you for using SPAdes!")
        log.removeHandler(log_handler)

    except Exception:
        _, exc, _ = sys.exc_info()
        log.exception(exc)
        support.error("exception caught", log)