def fetch_array(self, genome_range, genome_range2=None, balance=None, resolution='auto'):
"""
Parameters
----------
genome_range : {str, GenomeRange}
Intervals within input chromosome range.
genome_range2 : {str, GenomeRange}, optional.
balance : bool, optional
balance matrix or not,
default `self.is_balance`.
resolution : {'auto', int}
resolution of the data. for example 5000.
'auto' for calculate resolution automatically.
default 'auto'
Return
------
arr : numpy.ndarray
"""
from coolbox.utilities.hic.wrap import CoolerWrap
path = self.properties['file']
if balance is None:
balance = self.is_balance
wrap = CoolerWrap(path, balance=balance, binsize=resolution)
arr = wrap.fetch(genome_range, genome_range2)
return arr
def infer_binsize(self, gr: GenomeRange, **kwargs) -> int:
from coolbox.utilities.hic.wrap import CoolerWrap
path = self.properties['file']
wrap = CoolerWrap(path,
balance=self.balance,
binsize=kwargs.get('resolution', 'auto'))
return wrap.infer_binsize(gr)
def fetch_data(self, gr: GenomeRange, gr2=None, **kwargs) -> np.ndarray:
from coolbox.utilities.hic.wrap import CoolerWrap
path = self.properties['file']
binsize = kwargs.get('resolution',
self.properties.get('resolution', 'auto'))
wrap = CoolerWrap(path, balance=self.balance, binsize=binsize)
arr = wrap.fetch(gr, gr2)
self.fetched_binsize = wrap.fetched_binsize # expose fetched binsize
return self.fill_zero_nan(arr)
def fetch_pixels(self,
genome_range,
genome_range2=None,
balance=None,
resolution='auto',
join=True):
"""
Parameters
----------
genome_range : {str, GenomeRange}
Intervals within input chromosome range.
genome_range2 : {str, GenomeRange}, optional.
balance : bool, optional
balance matrix or not,
default `self.is_balance`.
resolution : {'auto', int}
resolution of the data. for example 5000.
'auto' for calculate resolution automatically.
default 'auto'
join : bool
whether to expand the bin ID columns
into (chrom, start, end).
default True
Return
------
pixels : pandas.core.frame.DataFrame
Hi-C pixels table.
The pixel table contains the non-zero upper triangle entries of the contact map.
"""
from coolbox.utilities.hic.wrap import CoolerWrap
genome_range = to_gr(genome_range)
if genome_range2 is not None:
genome_range2 = to_gr(genome_range2)
path = self.properties['file']
if balance is None:
balance = self.is_balance
wrap = CoolerWrap(path, balance=balance, binsize=resolution)
pixels = wrap.fetch_pixels(genome_range, genome_range2, join=join)
return pixels
def fetch_matrix(self,
genome_range,
genome_range2=None,
resolution='auto') -> np.ndarray:
"""
${fetch_matrix}
"""
from coolbox.utilities.hic.wrap import CoolerWrap
path = self.properties['file']
wrap = CoolerWrap(path, balance=self.balance, binsize=resolution)
arr = wrap.fetch(genome_range, genome_range2)
self.fetched_binsize = wrap.fetched_binsize # expose fetched binsize
return self.fill_zero_nan(arr)
def fetch_pixels(self, gr: GenomeRange, **kwargs):
"""
Fetch the pixels table of upper triangle of the original contact matrix(not processed).
Parameters
----------
gr2 : GenomeRange, optional.
balance : bool, optional
balance matrix or not,
default `self.is_balance`.
resolution : {'auto', int}
resolution of the data. for example 5000.
'auto' for calculate resolution automatically.
default 'auto'
join : bool
whether to expand the bin ID columns
into (chrom, start, end).
default True
Returns
-------
pixels : pandas.core.frame.DataFrame
Hi-C pixels table.
The pixel table contains the non-zero upper triangle entries of the contact map.
"""
from coolbox.utilities.hic.wrap import CoolerWrap
path = self.properties['file']
balance = kwargs.get('balance', self.is_balance)
wrap = CoolerWrap(path,
balance=balance,
binsize=kwargs.get('resolution', 'auto'))
pixels = wrap.fetch_pixels(gr,
kwargs.get('gr2'),
join=kwargs.get('join', True))
return pixels
def __fetch_matrix(self, genome_range, resolution='auto'):
"""
Fetch the matrix.
Parameters
----------
genome_range : coolbox.utilities.GenomeRange
The genome range to fetch.
resolution : {'auto', int}
The matrix resolution, for multi-resolution(.hic or multi-cool) file.
Use 'auto' to infer the resolution automatically.
default 'auto'
"""
from coolbox.utilities.hic.wrap import StrawWrap, CoolerWrap
path = self.properties['file']
if self.file_type == '.hic':
wrap = StrawWrap(path,
normalization=self.balance,
binsize=resolution)
else:
wrap = CoolerWrap(path, balance=self.balance, binsize=resolution)
arr = wrap.fetch(genome_range)
self.fetched_binsize = wrap.fetched_binsize # expose fetched binsize
# fill zero and nan with small value
small = self.small_value
arr[arr == 0] = small
arr[np.isnan(arr)] = small
if 'transform' in self.properties and self.properties[
'transform'] != 'no':
arr = self.__transform_matrix(arr)
return arr