def umi_visualization(bams, chrom, start, end, output, coor):
doc = genomeview.Document(1000)
global umi_colors
umi_colors = {}
# genome
source = genomeview.FastaGenomeSource(REF_FA)
gv = genomeview.GenomeView(chrom, max(0, start), end, "+", source)
axis = genomeview.Axis()
gv.add_track(axis)
label_track = genomeview.track.TrackLabel('{}:{}-{}'.format(
chrom, start, end))
gv.tracks.insert(0, label_track)
for bam in bams:
# VCF/BAM track
name = bam.split('/')[-1]
if bam[-7:] == '.vcf.gz': # VCF track
variant_track = VCFTrack(bam, name, chrom, start, end)
gv.add_track(variant_track)
else: # bam track
track = genomeview.PairedEndBAMTrack(bam, name=name)
#track = genomeview.SingleEndBAMTrack(bam, name=name)
gv.add_track(track)
track.nuc_colors = {
"A": "blue",
"C": "orange",
"G": "green",
"T": "black",
"N": "gray"
}
track.quick_consensus = False
# format
global count, colors
count, frag_draw, umi_draw, umi_all, any_umi = stats_umi(
bam, chrom, start, end)
if len(coor.split(':')) == 3:
colors = ColorIter(umi_draw) # color generater
umi_ar = list(count.keys())
for umi in umi_ar:
umi_n = count[umi]
if umi_n > 1 and umi not in umi_colors:
umi_colors[umi] = colors.next_color()
track.color_fn = color_by_umi
track.include_read_fn = filter_by_umi # exculde reads out of
else:
colors = ColorIter(len(any_umi)) # color generater
for umi in any_umi:
umi_colors[umi] = colors.next_color()
track.include_read_fn = filter_by_coor # exculde reads out of
track.color_fn = color_by_umi
#track.color_fn = lambda x: "lightgray"
doc.elements.append(gv)
genomeview.save(doc, '{}.svg'.format(output))
def bam_doc():
import genomeview
source = genomeview.FastaGenomeSource(reference_path())
doc = genomeview.Document(900)
view = genomeview.GenomeView("chr4", 96549060, 96549060+250, "+", source)
doc.add_view(view)
bam_track_hg002 = genomeview.SingleEndBAMTrack("data/quick_consensus_test.bam", name="HG002")
bam_track_hg002.min_indel_size = 3
view.add_track(bam_track_hg002)
axis_track = genomeview.Axis()
view.add_track(axis_track)
return doc
def test_view_without_source():
import genomeview
doc = genomeview.Document(900)
view = genomeview.GenomeView("chr4", 96549060, 96549060+1000, "+")
doc.add_view(view)
bam_track_hg002 = genomeview.SingleEndBAMTrack("data/quick_consensus_test.bam", name="HG002")
bam_track_hg002.draw_mismatches = False
view.add_track(bam_track_hg002)
axis_track = genomeview.Axis()
view.add_track(axis_track)
genomeview.save(doc, "results/temp_without_source.svg")
with pytest.raises(AssertionError):
bam_track_hg002.draw_mismatches = True
genomeview.save(doc, "results/temp_without_source_error.svg")
def visualize_data(file_paths,
chrom,
start,
end,
reference_path=None,
width=900,
axis_on_top=False):
"""
Creates a GenomeView document to display the data in the specified
files (eg bam, bed, etc).
Args:
file_paths: this specifies the file paths to be rendered. It must be
either a list/tuple of the paths, or a dictionary mapping
{track_name:path}. (If you are using a python version prior to 3.6,
use collections.ordereddict to ensure the order remains the same.)
Currently supports files ending in .bam, .cram, .bed, .bed.gz,
.bigbed, or .bigwig (or .bw). Most of these file types require a
separate index file to be present (eg a .bam.bai or a .bed.gz.tbi
file must exist).
chrom: chromosome (or contig) to be rendered
start: start coordinate of region to be rendered
end: end coordinate of region to be rendered
reference_path: path to fasta file specifying reference genomic
sequence. This is required in order to display mismatches
in bam tracks.
width: the pixel width of the document
axis_on_top: specifies whether the axis should be added at the bottom
(default) or at the top
"""
if reference_path is not None:
source = genomeview.FastaGenomeSource(reference_path)
else:
source = None
doc = genomeview.Document(width)
view = genomeview.GenomeView(chrom, start, end, "+", source)
doc.add_view(view)
def add_axis():
axis_track = genomeview.Axis("axis")
view.add_track(axis_track)
if axis_on_top:
add_axis()
if isinstance(file_paths, collections.Mapping):
names = file_paths.keys()
file_paths = [file_paths[name] for name in names]
else:
names = [None] * len(file_paths)
file_paths = file_paths
for name, path in zip(names, file_paths):
if path.lower().endswith(".bam") or path.lower().endswith(".cram"):
if utilities.is_paired_end(path):
cur_track = genomeview.PairedEndBAMTrack(path, name=name)
else:
cur_track = genomeview.SingleEndBAMTrack(path, name=name)
if utilities.is_long_frag_dataset(path):
cur_track.min_indel_size = 5
elif path.lower().endswith(".bed") or path.lower().endswith(
".bed.gz") or path.lower().endswith(".bigbed"):
cur_track = genomeview.BEDTrack(path, name=name)
elif path.lower().endswith(".bigwig") or path.lower().endswith(".bw"):
cur_track = genomeview.BigWigTrack(path, name=name)
else:
suffix = os.path.basename(path)
raise ValueError("Unknown file suffix: {}".format(suffix))
view.add_track(cur_track)
if not axis_on_top:
add_axis()
return doc