def test_coverage_module(tmpdir):
bed = bedtools.GenomeCov(sequana_data("JB409847.cov.csv"),
sequana_data("JB409847.gbk"))
directory = tmpdir.mkdir('test_coverage_module')
config.output_dir = str(directory)
config.sample_name = "JB409847"
CoverageModule(bed)
def __init__(self, data):
""".. rubric:: constructor
:param data: it can be a csv filename created by sequana_coverage or a
:class:`bedtools.GenomeCov` object.
"""
super().__init__()
try:
self.bed = bedtools.GenomeCov(data)
except FileNotFoundError:
msg = ("The csv file is not present. Please, check if your"
" file is present.")
raise FileNotFoundError(msg)
except TypeError:
self.bed = data
try:
html_list = self.create_reports()
except TypeError:
msg = ("Data must be either a csv file or a :class:`GenomeCov` "
"instance where zscore is computed.")
raise TypeError(msg)
self.title = "Coverage Report of {0}".format(config.sample_name)
self.intro = ("<p>Report the coverage of your sample {0} to check the "
"quality of your mapping and to highlight regions of "
"interest (under and over covered).</p>".format(
config.sample_name))
self.create_report_content(html_list)
self.create_html("sequana_coverage.html")
def test_binining():
filename = sequana_data('JB409847.bed')
# using chunksize of 7000, we test odd number
bed = bedtools.GenomeCov(filename,
sequana_data('JB409847.gbk'),
chunksize=7000)
chrom = bed.chr_list[0]
chrom.run(501, k=2, circular=True, binning=2, cnv_delta=100)
def test_ChromosomeCovMultiChunk():
filename = sequana_data('JB409847.bed')
# using chunksize of 7000, we test odd number
bed = bedtools.GenomeCov(filename,
sequana_data('JB409847.gbk'),
chunksize=7000)
chrom = bed.chr_list[0]
res = chrom.run(501, k=2, circular=True)
res.get_summary()
res.get_rois()
def test_chromosome():
filename = sequana_data('JB409847.bed')
# using chunksize of 7000, we test odd number
bed = bedtools.GenomeCov(filename,
sequana_data('JB409847.gbk'),
chunksize=7000)
chrom = bed.chr_list[0]
chrom.run(501, k=2, circular=True)
print(chrom)
# using chunksize of 7000, we test even number
bed = bedtools.GenomeCov(filename,
sequana_data('JB409847.gbk'),
chunksize=7000)
chrom = bed.chr_list[0]
chrom.run(501, k=2, circular=True)
# no chunksize
bed = bedtools.GenomeCov(filename, sequana_data('JB409847.gbk'))
chrom = bed.chr_list[0]
chrom.run(501, k=2, circular=True)
print(chrom)
# no chunksize
bed = bedtools.GenomeCov(filename, sequana_data('JB409847.gbk'))
chrom = bed.chr_list[0]
try:
chrom._coverage_scaling()
assert False
except KeyError:
assert True
except:
assert False
# zscore not computed yet, so error
try:
chrom.plot_rois(3000, 8000)
assert False
except:
assert True
chrom.run(4001)
chrom.plot_rois(3000, 8000)
def test_coverage_module(tmpdir):
bed = bedtools.GenomeCov(sequana_data("JB409847.bed"))
fasta = sequana_data("JB409847.fasta")
bed.compute_gc_content(fasta)
c = bed.chr_list[0]
c.run(4001)
directory = tmpdir.mkdir('test_coverage_module')
config.output_dir = str(directory)
config.sample_name = "JB409847"
CoverageModule(bed)
def test_canvasjs_linegraph():
bed = bedtools.GenomeCov(sequana_data("JB409847.bed"))
fasta = sequana_data("JB409847.fasta")
bed.compute_gc_content(fasta)
c = bed.chr_list[0]
c.run(4001)
df = bed[0].df
csv = df.to_csv(columns=['pos', 'cov', 'gc'], index=False,
float_format='%.3g')
# create CanvasJS stuff
cjs = CanvasJSLineGraph(csv, 'cov', 'pos', ['cov', 'gc'])
# set options
cjs.set_options({'zoomEnabled': 'true',
'zoomType': 'x',
'exportEnabled': 'true'})
# set title
cjs.set_title("Genome Coverage")
# set legend
cjs.set_legend({'verticalAlign': 'bottom',
'horizontalAlign': 'center',
'cursor':'pointer'},
hide_on_click=True)
# set axis
cjs.set_axis_x({'title': "Position (bp)",
'labelAngle': 30,
'minimum': 0,
'maximum': len(df)})
cjs.set_axis_y({'title': "Coverage (Count)"})
cjs.set_axis_y2({'title': "GC content (ratio)",
'minimum':0,
'maximum': 1,
'lineColor': '#FFC425',
'titleFontColor': '#FFC425',
'labelFontColor': '#FFC425'})
# set datas
cjs.set_data(index=0, data_dict={'type': 'line',
'name': "Coverage",
'showInLegend': 'true',
'color': '#5BC0DE',
'lineColor': '#5BC0DE'})
cjs.set_data(index=1, data_dict={'type': 'line',
'axisYType': 'secondary',
'name': "GC content",
'showInLegend': 'true',
'color': '#FFC425',
'lineColor': '#FFC425'})
# create canvasJS
cjs.create_canvasjs()
def test_datatables():
bed = bedtools.GenomeCov(sequana_data("JB409847.cov.csv"),
sequana_data("JB409847.gbk"))
rois = bed[0].get_roi()
rois.df['link'] = 'test'
datatable_js = DataTableFunction(rois.df, 'roi')
datatable_js.set_links_to_column('link', 'start')
datatable_js.datatable_options = {
'scrollX': 'true',
'pageLength': 15,
'scrollCollapse': 'true',
'dom': 'Bfrtip',
'buttons': ['copy', 'csv']
}
datatable = DataTable(rois.df, 'rois', datatable_js)
html_table = datatable.create_datatable(float_format='%.3g')
def test_gc_content():
bed = sequana_data('JB409847.bed')
fasta = sequana_data('JB409847.fasta')
cov = bedtools.GenomeCov(bed)
cov.compute_gc_content(fasta)
cov.get_stats()
ch = cov[0]
ch.moving_average(4001, circular=True)
ch.running_median(4001,circular=True)
ch.compute_zscore()
ch.get_evenness()
ch.get_cv()
assert ch.get_centralness() > 0.84 and ch.get_centralness()<0.85
with TempFile(suffix=".png") as fh:
ch.plot_gc_vs_coverage(filename=fh.name)
with TempFile() as fh:
ch.to_csv(fh.name)
ch.get_max_gc_correlation(fasta)
def test_datatables():
bed = bedtools.GenomeCov(sequana_data("JB409847.bed"),
sequana_data("JB409847.gbk"))
fasta = sequana_data("JB409847.fasta")
bed.compute_gc_content(fasta)
c = bed.chr_list[0]
c.run(4001)
rois = c.get_rois()
rois.df['link'] = 'test'
datatable_js = DataTableFunction(rois.df, 'roi')
datatable_js.set_links_to_column('link', 'start')
datatable_js.datatable_options = {
'scrollX': 'true',
'pageLength': 15,
'scrollCollapse': 'true',
'dom': 'Bfrtip',
'buttons': ['copy', 'csv']
}
datatable = DataTable(rois.df, 'rois', datatable_js)
html_table = datatable.create_datatable(float_format='%.3g')
def test_genomecov():
filename = sequana_data('JB409847.bed')
try:
bed = bedtools.GenomeCov("dummy.csv")
assert False
except:
assert True
try:
bed = bedtools.GenomeCov(filename, "dummy.gbk")
assert False
except:
assert True
# !now let us read the good data sets
bed = bedtools.GenomeCov(filename, sequana_data('JB409847.gbk'))
bed.compute_coverage(4001)
bed = bedtools.GenomeCov(filename, sequana_data('JB409847.gbk'))
bed2 = bedtools.GenomeCov(filename, sequana_data('JB409847.gbk'))
assert bed == bed
# test equality for same chromosome but different data
bed2.chr_list[0].df["cov"] += 100
assert bed != bed2
# test equality for same chromosome but different data
bed2.chr_list[0].df["cov"] -= 100
bed2.chr_list.append("dummy")
assert bed != bed2
# setter must be bool
try:
bed.circular = 1
assert False
except:
assert True
# cant use setter
try:
bed.feature_dict = {}
assert False
except:
assert True
assert len(bed) == 1
# a getter for the first chromosome
bed[0]
# setter available but not sure this is useful
bed.window_size = 4001
bed.hist()
# This requires to call other method before
for chrom in bed:
chrom.moving_average(n=501)
chrom.running_median(n=501, circular=True)
chrom.running_median(n=501, circular=False)
chrom.compute_zscore()
roi = chrom.get_roi()
with TempFile(suffix='.png') as fh:
chrom.plot_coverage(filename=fh.name)
with TempFile(suffix='.png') as fh:
chrom.plot_hist_zscore(filename=fh.name)
with TempFile(suffix='.png') as fh:
chrom.plot_hist_normalized_coverage(filename=fh.name)
len(chrom)
print(chrom)
chrom.get_size()
chrom.get_mean_cov()
chrom.get_var_coef()
with TempFile(suffix='.csv') as fh:
bed.to_csv(fh.name)
bed2 = bedtools.GenomeCov(fh.name, sequana_data('JB409847.gbk'))
# plotting
bed.chr_list[0].plot_hist_coverage()
bed.chr_list[0].plot_hist_coverage(logx=False,logy=True)
bed.chr_list[0].plot_hist_coverage(logx=True,logy=False)
with TempFile(suffix=".png") as fh:
bed.chr_list[0].plot_hist_coverage(logx=False,logy=False,
filename=fh.name)
def test_genomecov():
filename = sequana_data('JB409847.bed')
# wrong file
try:
bed = bedtools.GenomeCov("dummy.csv")
assert False
except:
assert True
# wrong threshold
try:
bed = bedtools.GenomeCov(filename, high_threshold=2)
assert False
except:
assert True
# wrong threshold
try:
bed = bedtools.GenomeCov(filename, low_threshold=-2)
assert False
except:
assert True
# wrong genbank
try:
bed = bedtools.GenomeCov(filename, "dummy.gbk")
assert False
except:
assert True
# !now let us read the good data sets by chunkd
bed = bedtools.GenomeCov(filename,
sequana_data('JB409847.gbk'),
chunksize=5000)
for c in bed.chr_list:
c.run(1001, k=2)
# setter must be bool
try:
bed.circular = 1
assert False
except:
assert True
# cant use setter
try:
bed.feature_dict = {}
assert False
except:
assert True
assert len(bed) == 1
# a getter for the first chromosome
bed[0]
# setter available but not sure this is useful
bed.window_size = 4000
bed.window_size = 4001
bed.hist()
# This requires to call other method before
for chrom in bed:
chrom.moving_average(n=501)
chrom.running_median(n=501, circular=True)
chrom.running_median(n=501, circular=False)
chrom.compute_zscore()
roi = chrom.get_rois()
with TempFile(suffix='.png') as fh:
chrom.plot_coverage(filename=fh.name)
with TempFile(suffix='.png') as fh:
chrom.plot_hist_zscore(filename=fh.name)
with TempFile(suffix='.png') as fh:
chrom.plot_hist_normalized_coverage(filename=fh.name)
len(chrom)
print(chrom)
chrom.get_size()
chrom.DOC
chrom.CV
with TempFile(suffix='.csv') as fh:
bed.gc_window_size = 100
bed.to_csv(fh.name)
# plotting
bed.chr_list[0].plot_hist_coverage()
bed.chr_list[0].plot_hist_coverage(logx=False, logy=True)
bed.chr_list[0].plot_hist_coverage(logx=True, logy=False)
with TempFile(suffix=".png") as fh:
bed.chr_list[0].plot_hist_coverage(logx=False,
logy=False,
filename=fh.name)