def file_to_unbinned_ranges(f, args, _):
file_format = sniff(f)
names = "Chromosome Start End Strand".split()
if file_format == "bed" or "bed.gz":
df = pd.read_csv(f,
sep="\t",
header=None,
usecols=[0, 1, 2, 5],
names=names)
gr = pr.PyRanges(df)
elif file_format == "bedpe":
df = pd.read_csv(f,
sep="\t",
header=None,
usecols=[0, 1, 5, 9],
names=names)
gr = pr.PyRanges(df)
elif file_format == "bampe":
# Shouldn't be reachable at the moment, bampe format is not supported here yet.
raise NotImplementedError(
"bampe format is not supported at the moment")
else:
gr = pr.read_bam(f)
if args["drop_duplicates"]:
gr = gr.drop_duplicate_positions()
gr = pr.gf.genome_bounds(gr, args["chromsizes_"])
# print("args", args)# ["chromsizes"])
return gr #.remove_out_of_genome_bounds_intervals()
def file_to_unbinned_ranges(f, args, _):
file_format = sniff(f)
names = "Chromosome Start End Strand".split()
if file_format == "bed" or "bed.gz":
df = pd.read_csv(f,
sep="\t",
header=None,
usecols=[0, 1, 2, 5],
names=names)
gr = pr.PyRanges(df)
elif file_format == "bedpe":
df = pd.read_csv(f,
sep="\t",
header=None,
usecols=[0, 1, 5, 9],
names=names)
gr = pr.PyRanges(df)
else:
gr = pr.read_bam(f)
if args["drop_duplicates"]:
gr = gr.drop_duplicate_positions()
gr = pr.gf.genome_bounds(gr, args["chromsizes_"])
# print("args", args)# ["chromsizes"])
return gr #.remove_out_of_genome_bounds_intervals()
def file_to_unbinned_ranges(f, args, _):
chromosome_data = {}
file_format = sniff(f)
names = "Chromosome Start End Strand".split()
if file_format == "bed" or "bed.gz":
df = pd.read_csv(f,
sep="\t",
header=None,
usecols=[0, 1, 2, 5],
names=names)
gr = pr.PyRanges(df)
elif file_format == "bedpe":
df = pd.read_csv(f,
sep="\t",
header=None,
usecols=[0, 1, 5, 9],
names=names)
gr = pr.PyRanges(df)
else:
gr = pr.read_bam(f)
return gr
def control_bam():
"""
>>> # +--------------+-----------+-----------+--------------+------------+
>>> # | Chromosome | Start | End | Strand | Flag |
>>> # | (category) | (int32) | (int32) | (category) | (uint16) |
>>> # |--------------+-----------+-----------+--------------+------------|
>>> # | chr1 | 887771 | 887796 | + | 16 |
>>> # | chr1 | 994660 | 994685 | + | 16 |
>>> # | chr1 | 1770383 | 1770408 | + | 16 |
>>> # | chr1 | 1995141 | 1995166 | + | 16 |
>>> # | ... | ... | ... | ... | ... |
>>> # | chrY | 57402214 | 57402239 | + | 16 |
>>> # | chrY | 10643526 | 10643551 | - | 0 |
>>> # | chrY | 11776321 | 11776346 | - | 0 |
>>> # | chrY | 20557165 | 20557190 | - | 0 |
>>> # +--------------+-----------+-----------+--------------+------------+
>>> # Stranded PyRanges object has 10,000 rows and 5 columns from 25 chromosomes.
>>> # For printing, the PyRanges was sorted on Chromosome and Strand.
"""
full_path = get_example_path("control.bam")
return pr.read_bam(full_path)
def read_bam_bin_counts(bins: PyRanges, bams: Dict[str, str], excluded: PyRanges = None, **kwargs) -> AnnData:
""" Count reads in bins from bams
Parameters
----------
bins : pyranges.PyRanges
bins in which to count reads
bams : Dict[Str]
bam filenames with cell ids as keys
excluded: PyRanges
excluded genomic regions to filter reads
Returns
-------
ad.AnnData
binned read counts
"""
bin_data = _convert_pyranges(bins)
bin_data = _add_bin_index(bin_data)
cn_matrix = {}
for cell_id, cell_bam in bams.items():
logging.info(f"reading {cell_bam}")
bam_data = pr.read_bam(cell_bam, **kwargs)
if excluded is not None:
logging.info("excluding reads")
bam_data = bam_data.intersect(excluded, invert=True)
logging.info(f"count overlaps")
bam_data = bam_data.intersect(bins, how='containment')
read_counts = bins.count_overlaps(bam_data, overlap_col='reads')
read_counts = _convert_pyranges(read_counts)
read_counts = _add_bin_index(read_counts)
cn_matrix[cell_id] = read_counts['reads']
cn_matrix = pd.DataFrame(cn_matrix)
cell_data = pd.DataFrame({'cell_id': cn_matrix.columns.values}).set_index('cell_id')
adata = ad.AnnData(
cn_matrix.T,
obs=cell_data,
var=bin_data,
)
return adata
def control_bam():
full_path = get_example_path("control.bam")
return pr.read_bam(full_path)
def test_read_bam():
pr.read_bam("tests/test_data/test_sorted.bam")
def read_bfile(self):
self.check_bfile()
if self.filetype == Type.BED:
return pr.read_bed(self.bfile)
return pr.read_bam(self.bfile)
def get_bam_ranges(self, filter_flag=3844):
logging.info("Extracting BAM ranges")
# note that read_bam by default has a specific set of SAM flags
# this code needs to be updated to allow for selection of +/-, primary
# secondary etc ... - this method is also independent of pysam//samtools
return pr.read_bam(self.bam, filter_flag=3844)
