comparison cwpair2_util.py @ 6:c4b926c9831c draft default tip

planemo upload for repository https://github.com/galaxyproject/tools-iuc/tree/master/tools/cwpair2 commit ca66053a0d915edba3d7f1d4ce014718e7f4f0f2

5:71188f3f4b76

MATCHED_PAIRS = 'MP'
ORPHANS = 'O'
# Data output formats
GFF_EXT = 'gff'
TABULAR_EXT = 'tabular'
# Statistics historgrams output directory.
HISTOGRAM = 'H'
# Statistics outputs
FINAL_PLOTS = 'F'
PREVIEW_PLOTS = 'P'
STATS_GRAPH = 'C'

        self.dist = d or {}
        self.binsize = binsize

    def get_bin(self, x):
        """
        Returns the bin in which a data point falls
        """
        return self.start + (x - self.start) // self.binsize * self.binsize + self.binsize / 2.0

    def add(self, x):
        x = self.get_bin(x)

            x.append(center)
            y.append(self.dist.get(center, 0))
        return x, y

    def mode(self):
        return max(self.dist.items(), key=lambda data: data[1])[0]

    def size(self):
        return sum(self.dist.values())

    sys.stderr.write(msg)
    sys.exit(1)


def distance(peak1, peak2):
    return (peak2[1] + peak2[2]) / 2 - (peak1[1] + peak1[2]) / 2


def gff_row(cname, start, end, score, source, type='.', strand='.', phase='.', attrs={}):
    return (cname, source, type, start, end, score, strand, phase, gff_attrs(attrs))

def parse_chromosomes(reader):
    # This version of cwpair2 accepts only gff format as input.
    chromosomes = {}
    next(reader)
    for line in reader:
        cname, junk, junk, start, end, value, strand, junk, junk = line
        start = int(start)
        end = int(end)
        value = float(value)
        if cname not in chromosomes:
            chromosomes[cname] = []

    values.sort()
    # Get 95% value
    return values[int(len(values) * 0.95)]


def filter(chromosomes, threshold=0.05):
    """
    Filters the peaks to those above a threshold. Threshold < 1.0 is interpreted
    as a proportion of the maximum, >=1.0 as an absolute value.
    """
    if threshold < 1:

    return watson, crick


def all_pair_distribution(chromosomes, up_distance, down_distance, binsize):
    dist = FrequencyDistribution(-up_distance, down_distance, binsize=binsize)
    for cname, data in chromosomes.items():
        watson, crick = split_strands(data)
        crick.sort(key=lambda data: float(data[1]))
        keys = make_keys(crick)
        for peak in watson:
            for cpeak in get_window(crick, peak, up_distance, down_distance, keys):

    output_details = output_files in ["all", "matched_pair_orphan_detail"]
    output_plots = output_files in ["all"]
    output_orphans = output_files in ["all", "matched_pair_orphan", "matched_pair_orphan_detail"]
    # Keep track of statistics for the output file
    statistics = {}
    input = csv.reader(open(dataset_path, 'rt'), delimiter='\t')
    fpath, fname = os.path.split(dataset_path)
    statistics['fname'] = '%s: data %s' % (method, str(galaxy_hid))
    statistics['dir'] = fpath
    if threshold >= 1:
        filter_string = 'fa%d' % threshold

        return os.path.join(output_type, '%s_%s.%s' % (output_type, fname, extension))

    def td_writer(output_type, extension, fname):
        # Returns a tab-delimited writer for a specified output.
        output_file_path = make_path(output_type, extension, fname)
        return csv.writer(open(output_file_path, 'wt'), delimiter='\t')

    try:
        chromosomes = parse_chromosomes(input)
    except Exception:
        stop_err('Unable to parse file "%s".\n%s' % (dataset_path, traceback.format_exc()))
    if output_details:
        # Details
        detailed_output = td_writer('data_%s' % DETAILS, TABULAR_EXT, fname)
        detailed_output.writerow(('chrom', 'start', 'end', 'value', 'strand') * 2 + ('midpoint', 'c-w reads sum', 'c-w distance (bp)'))
    if output_plots:

    statistics['stats_path'] = 'statistics.%s' % TABULAR_EXT
    if output_plots:
        statistics['graph_path'] = make_histogram_path(STATS_GRAPH, fname)
    statistics['perc95'] = perc95(chromosomes)
    if threshold > 0:
        # Apply filter
        filter(chromosomes, threshold)
    if method == 'mode':
        freq = all_pair_distribution(chromosomes, up_distance, down_distance, binsize)
        mode = freq.mode()
        statistics['preview_mode'] = mode
        if output_plots:

6:c4b926c9831c

MATCHED_PAIRS = 'MP'
ORPHANS = 'O'
# Data output formats
GFF_EXT = 'gff'
TABULAR_EXT = 'tabular'
# Statistics histograms output directory.
HISTOGRAM = 'H'
# Statistics outputs
FINAL_PLOTS = 'F'
PREVIEW_PLOTS = 'P'
STATS_GRAPH = 'C'

        self.dist = d or {}
        self.binsize = binsize

    def get_bin(self, x):
        """
        Returns the centre of the bin in which a data point falls
        """
        return self.start + (x - self.start) // self.binsize * self.binsize + self.binsize / 2.0

    def add(self, x):
        x = self.get_bin(x)

            x.append(center)
            y.append(self.dist.get(center, 0))
        return x, y

    def mode(self):
        # There could be more than one mode for a frequency distribution,
        # return the median of the modes to be consistent
        max_frequency = max(self.dist.values())
        modes = sorted(_[0] for _ in self.dist.items() if _[1] == max_frequency)
        median_index = len(modes) // 2
        return modes[median_index]

    def size(self):
        return sum(self.dist.values())

    sys.stderr.write(msg)
    sys.exit(1)


def distance(peak1, peak2):
    return (peak2[1] + peak2[2]) / 2.0 - (peak1[1] + peak1[2]) / 2.0


def gff_row(cname, start, end, score, source, type='.', strand='.', phase='.', attrs={}):
    return (cname, source, type, start, end, score, strand, phase, gff_attrs(attrs))

def parse_chromosomes(reader):
    # This version of cwpair2 accepts only gff format as input.
    chromosomes = {}
    for line in reader:
        line = line.rstrip("\r\n")
        if not line or line.startswith('#'):
            continue
        cname, _, _, start, end, value, strand, _, _ = line.split("\t")
        start = int(start)
        end = int(end)
        value = float(value)
        if cname not in chromosomes:
            chromosomes[cname] = []

    values.sort()
    # Get 95% value
    return values[int(len(values) * 0.95)]


def peak_filter(chromosomes, threshold):
    """
    Filters the peaks to those above a threshold. Threshold < 1.0 is interpreted
    as a proportion of the maximum, >=1.0 as an absolute value.
    """
    if threshold < 1:

    return watson, crick


def all_pair_distribution(chromosomes, up_distance, down_distance, binsize):
    dist = FrequencyDistribution(-up_distance, down_distance, binsize=binsize)
    for data in chromosomes.values():
        watson, crick = split_strands(data)
        crick.sort(key=lambda data: float(data[1]))
        keys = make_keys(crick)
        for peak in watson:
            for cpeak in get_window(crick, peak, up_distance, down_distance, keys):

    output_details = output_files in ["all", "matched_pair_orphan_detail"]
    output_plots = output_files in ["all"]
    output_orphans = output_files in ["all", "matched_pair_orphan", "matched_pair_orphan_detail"]
    # Keep track of statistics for the output file
    statistics = {}
    fpath, fname = os.path.split(dataset_path)
    statistics['fname'] = '%s: data %s' % (method, str(galaxy_hid))
    statistics['dir'] = fpath
    if threshold >= 1:
        filter_string = 'fa%d' % threshold

        return os.path.join(output_type, '%s_%s.%s' % (output_type, fname, extension))

    def td_writer(output_type, extension, fname):
        # Returns a tab-delimited writer for a specified output.
        output_file_path = make_path(output_type, extension, fname)
        return csv.writer(open(output_file_path, 'wt'), delimiter='\t', lineterminator="\n")

    with open(dataset_path, 'rt') as input:
        try:
            chromosomes = parse_chromosomes(input)
        except Exception:
            stop_err('Unable to parse file "%s".\n%s' % (dataset_path, traceback.format_exc()))
    if output_details:
        # Details
        detailed_output = td_writer('data_%s' % DETAILS, TABULAR_EXT, fname)
        detailed_output.writerow(('chrom', 'start', 'end', 'value', 'strand') * 2 + ('midpoint', 'c-w reads sum', 'c-w distance (bp)'))
    if output_plots:

    statistics['stats_path'] = 'statistics.%s' % TABULAR_EXT
    if output_plots:
        statistics['graph_path'] = make_histogram_path(STATS_GRAPH, fname)
    statistics['perc95'] = perc95(chromosomes)
    if threshold > 0:
        # Apply peak_filter
        peak_filter(chromosomes, threshold)
    if method == 'mode':
        freq = all_pair_distribution(chromosomes, up_distance, down_distance, binsize)
        mode = freq.mode()
        statistics['preview_mode'] = mode
        if output_plots: