def to_reports(subreads, output_dir):
output_files = []
log.info("Loading {f}".format(f=subreads))
ds = SubreadSet(subreads)
ds.loadStats()
for base, module in [("filter_stats_xml", filter_stats_xml),
("adapter_xml", adapter_xml),
("loading_xml", loading_xml),
("control", control)]:
constants = getattr(module, "Constants")
task_id = constants.TOOL_ID
to_report = getattr(module, "to_report_impl")
try:
rpt_output_dir = os.path.join(output_dir, base)
os.mkdir(rpt_output_dir)
file_name = os.path.join(rpt_output_dir, "{b}.json".format(b=base))
report = to_report(ds, rpt_output_dir)
log.info("Writing {f}".format(f=file_name))
report.write_json(file_name)
output_files.append(DataStoreFile(
uuid=report.uuid,
source_id=task_id,
type_id=FileTypes.REPORT.file_type_id,
path=file_name,
is_chunked=False,
name=base))
except InvalidStatsError as e:
log.error("This dataset lacks some required statistics")
log.error("Skipping generation of {b} report".format(b=base))
datastore = DataStore(output_files)
return datastore
def to_report(stats_xml, output_dir, dpi=72):
"""Main point of entry
:type stats_xml: str
:type output_dir: str
:type dpi: int
:rtype: Report
"""
log.info("Analyzing XML {f}".format(f=stats_xml))
# stats_xml should be a dataset:
dset = SubreadSet(stats_xml)
dataset_uuids = [dset.uuid]
# but if it isn't, no problem:
if not dset.metadata.summaryStats:
dset.loadStats(stats_xml)
# an sts file was provided which will generate a new random uuid
dataset_uuids = []
if not dset.metadata.summaryStats.readLenDists:
raise IOError("Pipeline Summary Stats (sts.xml) not found or missing "
"key distributions")
# we want all of the length distributions in this report to look the same,
# so we make the shaper here and pass it around:
alldists = (dset.metadata.summaryStats.readLenDists[:] +
dset.metadata.summaryStats.insertReadLenDists[:])
len_dist_shaper = continuous_dist_shaper(alldists, trim_excess=True)
attr = to_read_stats_attributes(
readLenDists=dset.metadata.summaryStats.readLenDists,
readQualDists=dset.metadata.summaryStats.readQualDists)
attr.extend(
to_insert_stats_attributes(
readLenDists=dset.metadata.summaryStats.insertReadLenDists,
readQualDists=dset.metadata.summaryStats.insertReadQualDists))
plot_groups = to_read_stats_plots(
readLenDists=dset.metadata.summaryStats.readLenDists,
readQualDists=dset.metadata.summaryStats.readQualDists,
output_dir=output_dir,
lenDistShaper=len_dist_shaper)
plot_groups.extend(
to_insert_stats_plots(
readLenDists=dset.metadata.summaryStats.insertReadLenDists,
readQualDists=dset.metadata.summaryStats.insertReadQualDists,
output_dir=output_dir,
lenDistShaper=len_dist_shaper))
# build the report:
report = Report(meta_rpt.id,
title=meta_rpt.title,
attributes=attr,
plotgroups=plot_groups,
dataset_uuids=dataset_uuids)
return meta_rpt.apply_view(report)
def to_report(stats_xml, output_dir):
log.info("Starting {f} v{v}".format(f=os.path.basename(__file__),
v=__version__))
log.info("Analyzing XML {f}".format(f=stats_xml))
dset = SubreadSet(stats_xml)
dset.loadStats()
if stats_xml.endswith(".sts.xml"):
dset.loadStats(stats_xml)
return to_report_impl(dset, output_dir)
def to_report(stats_xml, output_dir, dpi=72):
"""Main point of entry
:type stats_xml: str
:type output_dir: str
:type dpi: int
:rtype: Report
"""
log.info("Analyzing XML {f}".format(f=stats_xml))
# stats_xml should be a dataset:
dset = SubreadSet(stats_xml)
dataset_uuids = [dset.uuid]
# but if it isn't, no problem:
if not dset.metadata.summaryStats:
dset.loadStats(stats_xml)
# an sts file was provided which will generate a new random uuid
dataset_uuids = []
if not dset.metadata.summaryStats.readLenDists:
raise IOError("Pipeline Summary Stats (sts.xml) not found or missing "
"key distributions")
# we want all of the length distributions in this report to look the same,
# so we make the shaper here and pass it around:
alldists = (dset.metadata.summaryStats.readLenDists[:] +
dset.metadata.summaryStats.insertReadLenDists[:])
len_dist_shaper = continuous_dist_shaper(alldists, trim_excess=True)
attr = to_read_stats_attributes(
readLenDists=dset.metadata.summaryStats.readLenDists,
readQualDists=dset.metadata.summaryStats.readQualDists)
attr.extend(to_insert_stats_attributes(
readLenDists=dset.metadata.summaryStats.insertReadLenDists,
readQualDists=dset.metadata.summaryStats.insertReadQualDists))
plot_groups = to_read_stats_plots(
readLenDists=dset.metadata.summaryStats.readLenDists,
readQualDists=dset.metadata.summaryStats.readQualDists,
output_dir=output_dir,
lenDistShaper=len_dist_shaper)
plot_groups.extend(to_insert_stats_plots(
readLenDists=dset.metadata.summaryStats.insertReadLenDists,
readQualDists=dset.metadata.summaryStats.insertReadQualDists,
output_dir=output_dir,
lenDistShaper=len_dist_shaper))
# build the report:
report = Report(meta_rpt.id,
title=meta_rpt.title,
attributes=attr,
plotgroups=plot_groups,
dataset_uuids=dataset_uuids)
return meta_rpt.apply_view(report)
def to_report(stats_xml, output_dir):
"""Main point of entry
:type stats_xml: str
:type output_dir: str
:type dpi: int
:rtype: Report
"""
log.info("Analyzing XML {f}".format(f=stats_xml))
dset = SubreadSet(stats_xml)
if stats_xml.endswith(".sts.xml"):
dset.loadStats(stats_xml)
dset.loadStats()
return to_report_impl(dset, output_dir)
def to_report(stats_xml, output_dir, dpi=DEFAULT_DPI):
# TODO: make dpi matter
"""Main point of entry
:type stats_xml: str
:type output_dir: str
:type dpi: int
:rtype: Report
"""
log.info("Analyzing XML {f}".format(f=stats_xml))
dset = SubreadSet(stats_xml)
if not dset.metadata.summaryStats:
dset.loadStats(stats_xml)
return to_report_impl(dset, output_dir, dpi)
def to_report(stats_xml, output_dir, dpi=72):
"""Main point of entry
:type stats_xml: str
:type output_dir: str
:type dpi: int
:rtype: Report
"""
log.info("Analyzing XML {f}".format(f=stats_xml))
# stats_xml should be a dataset:
dset = SubreadSet(stats_xml)
dataset_uuids = [dset.uuid]
# but if it isn't, no problem:
if not dset.metadata.summaryStats:
dset.loadStats(stats_xml)
# an sts file was provided which will generate a new random uuid
dataset_uuids = []
if not dset.metadata.summaryStats.readLenDists:
raise IOError("Pipeline Summary Stats (sts.xml) not found or missing "
"key distributions")
attr = to_read_stats_attributes(
readLenDists=dset.metadata.summaryStats.readLenDists,
readQualDists=dset.metadata.summaryStats.readQualDists)
attr.extend(to_insert_stats_attributes(
readLenDists=dset.metadata.summaryStats.insertReadLenDists,
readQualDists=dset.metadata.summaryStats.insertReadQualDists))
plot_groups = to_read_stats_plots(
readLenDists=dset.metadata.summaryStats.readLenDists,
readQualDists=dset.metadata.summaryStats.readQualDists,
output_dir=output_dir)
plot_groups.extend(to_insert_stats_plots(
readLenDists=dset.metadata.summaryStats.insertReadLenDists,
readQualDists=dset.metadata.summaryStats.insertReadQualDists,
output_dir=output_dir))
# build the report:
report = Report("raw_data_report",
title="Raw Data Report",
attributes=attr,
plotgroups=plot_groups,
dataset_uuids=dataset_uuids)
return report
def test_reports_with_fixed_bins(self):
# TODO readQualDists are currently unpopulated, turn back on when
# they're repopulated
# for dist_name, nbins in zip(['medianInsertDists', 'readLenDists',
# 'readQualDists'], [200, 200, 50]):
for dist_name, nbins in zip(['medianInsertDists', 'readLenDists'],
[200, 200]):
ss = SubreadSet()
ss.loadStats(get_fixed_bin_sts())
ss2 = SubreadSet()
ss2.loadStats(get_fixed_bin_sts())
# shift ss2
mdist = getattr(ss2.metadata.summaryStats, dist_name)[0].bins
mdist = [0, 0, 0] + mdist[:-3]
getattr(ss2.metadata.summaryStats, dist_name)[0].bins = mdist
ss3 = ss + ss2
ss4 = SubreadSet()
ss4.loadStats(get_fixed_bin_sts())
# shift ss4
mdist = getattr(ss4.metadata.summaryStats, dist_name)[0].bins
mdist = [0 for _ in mdist]
getattr(ss4.metadata.summaryStats, dist_name)[0].bins = mdist
dists = getattr(ss4.metadata.summaryStats, dist_name)
self.assertEqual(len(dists), 1)
for n in [0, 1, 2, 10, 40, 41, 49, 50, 51, 200, 500]:
ds = continuous_dist_shaper(dists, nbins=n)
fixed_dists = [ds(dist) for dist in dists]
self.assertEqual(len(dists[0].bins), nbins)
self.assertEqual(len(fixed_dists[0].bins), nbins)
self.assertEqual(sum(dists[0].bins), sum(fixed_dists[0].bins))
sss = [ss, ss2, ss3]
for sset in sss:
dists = getattr(sset.metadata.summaryStats, dist_name)
self.assertEqual(len(dists), 1)
# 0, requested nbins > numBins fails back to no-op
ops = [1, 2, 3, 4, 7, 10, 40, 41, 49, 50, 51, 200, 500]
no_ops = [0]
for n in no_ops:
ds = continuous_dist_shaper(dists, nbins=n)
fixed_dists = [ds(dist) for dist in dists]
self.assertEqual(len(dists[0].bins), nbins)
self.assertEqual(len(fixed_dists[0].bins), nbins)
self.assertEqual(sum(dists[0].bins),
sum(fixed_dists[0].bins))
for n in ops:
ds = continuous_dist_shaper(dists, nbins=n)
fixed_dists = [ds(dist) for dist in dists]
self.assertEqual(len(dists[0].bins), nbins)
self.assertEqual(len(fixed_dists[0].bins), n)
self.assertEqual(sum(dists[0].bins),
sum(fixed_dists[0].bins))
def test_reports_with_fixed_bins(self):
# TODO readQualDists are currently unpopulated, turn back on when
# they're repopulated
# for dist_name, nbins in zip(['medianInsertDists', 'readLenDists',
# 'readQualDists'], [200, 200, 50]):
for dist_name, nbins in zip(["medianInsertDists", "readLenDists"], [200, 200]):
ss = SubreadSet()
ss.loadStats(get_fixed_bin_sts())
ss2 = SubreadSet()
ss2.loadStats(get_fixed_bin_sts())
# shift ss2
mdist = getattr(ss2.metadata.summaryStats, dist_name)[0].bins
mdist = [0, 0, 0] + mdist[:-3]
getattr(ss2.metadata.summaryStats, dist_name)[0].bins = mdist
ss3 = ss + ss2
ss4 = SubreadSet()
ss4.loadStats(get_fixed_bin_sts())
# shift ss4
mdist = getattr(ss4.metadata.summaryStats, dist_name)[0].bins
mdist = [0 for _ in mdist]
getattr(ss4.metadata.summaryStats, dist_name)[0].bins = mdist
dists = getattr(ss4.metadata.summaryStats, dist_name)
self.assertEqual(len(dists), 1)
for n in [0, 1, 2, 10, 40, 41, 49, 50, 51, 200, 500]:
ds = continuous_dist_shaper(dists, nbins=n)
fixed_dists = [ds(dist) for dist in dists]
self.assertEqual(len(dists[0].bins), nbins)
self.assertEqual(len(fixed_dists[0].bins), nbins)
self.assertEqual(sum(dists[0].bins), sum(fixed_dists[0].bins))
sss = [ss, ss2, ss3]
for sset in sss:
dists = getattr(sset.metadata.summaryStats, dist_name)
self.assertEqual(len(dists), 1)
# 0, requested nbins > numBins fails back to no-op
ops = [1, 2, 3, 4, 7, 10, 40, 41, 49, 50, 51, 200, 500]
no_ops = [0]
for n in no_ops:
ds = continuous_dist_shaper(dists, nbins=n)
fixed_dists = [ds(dist) for dist in dists]
self.assertEqual(len(dists[0].bins), nbins)
self.assertEqual(len(fixed_dists[0].bins), nbins)
self.assertEqual(sum(dists[0].bins), sum(fixed_dists[0].bins))
for n in ops:
ds = continuous_dist_shaper(dists, nbins=n)
fixed_dists = [ds(dist) for dist in dists]
self.assertEqual(len(dists[0].bins), nbins)
self.assertEqual(len(fixed_dists[0].bins), n)
self.assertEqual(sum(dists[0].bins), sum(fixed_dists[0].bins))
def test_stats_metadata(self):
ds = DataSet(data.getBam())
ds.loadStats(data.getStats())
self.assertEqual(ds.metadata.summaryStats.prodDist.numBins, 4)
self.assertEqual(ds.metadata.summaryStats.prodDist.bins,
[1576, 901, 399, 0])
ds1 = DataSet(data.getXml(8))
ds1.loadStats(data.getStats())
ds2 = DataSet(data.getXml(11))
ds2.loadStats(data.getStats())
ds3 = ds1 + ds2
self.assertEqual(ds1.metadata.summaryStats.prodDist.bins,
[1576, 901, 399, 0])
self.assertEqual(ds2.metadata.summaryStats.prodDist.bins,
[1576, 901, 399, 0])
self.assertEqual(ds3.metadata.summaryStats.prodDist.bins,
[3152, 1802, 798, 0])
self.assertEqual(ds1.metadata.summaryStats.readLenDist.bins,
[0, 62, 39, 36, 29, 37, 19, 29, 37, 32, 32, 40, 45,
54, 73, 77, 97, 95, 49, 17, 2, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0])
self.assertEqual(ds2.metadata.summaryStats.readLenDist.bins,
[0, 62, 39, 36, 29, 37, 19, 29, 37, 32, 32, 40, 45,
54, 73, 77, 97, 95, 49, 17, 2, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0])
self.assertEqual(ds3.metadata.summaryStats.readLenDist.bins,
[0, 124, 78, 72, 58, 74, 38, 58, 74, 64, 64, 80, 90,
108, 146, 154, 194, 190, 98, 34, 4, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0])
# Lets check some manual values
ds1 = DataSet(data.getXml(8))
ds1.loadStats(data.getStats())
ds2 = DataSet(data.getXml(11))
ds2.loadStats(data.getStats())
ds1.metadata.summaryStats.readLenDist.bins = (
[0, 10, 9, 8, 7, 6, 4, 2, 1, 0, 0, 1])
self.assertEqual(ds1.metadata.summaryStats.readLenDist.bins,
[0, 10, 9, 8, 7, 6, 4, 2, 1, 0, 0, 1])
ds1.metadata.summaryStats.readLenDist.minBinValue = 10
ds1.metadata.summaryStats.readLenDist.binWidth = 10
ds2.metadata.summaryStats.readLenDist.bins = (
[0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1])
self.assertEqual(ds2.metadata.summaryStats.readLenDist.bins,
[0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1])
ds2.metadata.summaryStats.readLenDist.minBinValue = 20
ds2.metadata.summaryStats.readLenDist.binWidth = 10
ds3 = ds1 + ds2
self.assertEqual(ds3.metadata.summaryStats.readLenDist.bins,
[0, 10, 10, 9, 8, 7, 5, 3, 2, 1, 0, 1, 1])
# now lets swap
ds1 = DataSet(data.getXml(8))
ds1.loadStats(data.getStats())
ds2 = DataSet(data.getXml(11))
ds2.loadStats(data.getStats())
ds1.metadata.summaryStats.readLenDist.bins = (
[0, 10, 9, 8, 7, 6, 4, 2, 1, 0, 0, 1])
self.assertEqual(ds1.metadata.summaryStats.readLenDist.bins,
[0, 10, 9, 8, 7, 6, 4, 2, 1, 0, 0, 1])
ds1.metadata.summaryStats.readLenDist.minBinValue = 20
ds1.metadata.summaryStats.readLenDist.binWidth = 10
ds2.metadata.summaryStats.readLenDist.bins = (
[0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1])
self.assertEqual(ds2.metadata.summaryStats.readLenDist.bins,
[0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1])
ds2.metadata.summaryStats.readLenDist.minBinValue = 10
ds2.metadata.summaryStats.readLenDist.binWidth = 10
ds3 = ds1 + ds2
self.assertEqual(ds3.metadata.summaryStats.readLenDist.bins,
[0, 1, 11, 10, 9, 8, 7, 5, 3, 1, 0, 1, 1])
# now lets do some non-overlapping
ds1 = DataSet(data.getXml(8))
ds1.loadStats(data.getStats())
ds2 = DataSet(data.getXml(11))
ds2.loadStats(data.getStats())
ds1.metadata.summaryStats.readLenDist.bins = (
[1, 1, 1])
self.assertEqual(ds1.metadata.summaryStats.readLenDist.bins,
[1, 1, 1])
ds1.metadata.summaryStats.readLenDist.minBinValue = 10
ds1.metadata.summaryStats.readLenDist.binWidth = 10
ds2.metadata.summaryStats.readLenDist.bins = (
[2, 2, 2])
self.assertEqual(ds2.metadata.summaryStats.readLenDist.bins,
[2, 2, 2])
ds2.metadata.summaryStats.readLenDist.minBinValue = 50
ds2.metadata.summaryStats.readLenDist.binWidth = 10
ds3 = ds1 + ds2
self.assertEqual(ds3.metadata.summaryStats.readLenDist.bins,
[1, 1, 1, 0, 2, 2, 2])
# now lets test the subdataset metadata retention:
ss = SubreadSet(data.getXml(10))
ss.loadStats(data.getStats(0))
ss.loadStats(data.getStats(1))
self.assertEqual(153168.0, ss.metadata.summaryStats.numSequencingZmws)
self.assertEqual(
2876.0, ss.subdatasets[0].metadata.summaryStats.numSequencingZmws)
self.assertEqual(
150292.0, ss.subdatasets[1].metadata.summaryStats.numSequencingZmws)
def get_fixed_bin_dataset():
sfn = get_fixed_bin_sts()
sset = SubreadSet()
sset.loadStats(sfn)
return sset
def to_report(stats_xml, output_dir, dpi=72):
# TODO: make dpi matter
"""Main point of entry
:type stats_xml: str
:type output_dir: str
:type dpi: int
:rtype: Report
"""
log.info("Analyzing XML {f}".format(f=stats_xml))
dset = SubreadSet(stats_xml)
if not dset.metadata.summaryStats:
dset.loadStats(stats_xml)
if not dset.metadata.summaryStats.medianInsertDists:
raise IOError("Pipeline Summary Stats (sts.xml) not found or missing "
"key distributions")
# Pull some stats:
adapter_dimers = np.round(
100.0 * dset.metadata.summaryStats.adapterDimerFraction,
decimals=2)
short_inserts = np.round(
100.0 * dset.metadata.summaryStats.shortInsertFraction,
decimals=2)
plots = []
# Pull some histograms (may have dupes (unmergeable distributions)):
shaper = continuous_dist_shaper(dset.metadata.summaryStats.medianInsertDists)
for i, orig_ins_len_dist in enumerate(
dset.metadata.summaryStats.medianInsertDists):
ins_len_dist = shaper(orig_ins_len_dist)
# make a bar chart:
fig, ax = get_fig_axes_lpr()
ax.bar(map(float, ins_len_dist.labels), ins_len_dist.bins,
color=get_green(0), edgecolor=get_green(0),
width=(ins_len_dist.binWidth * 0.75))
ax.set_xlabel(meta_rpt.get_meta_plotgroup(Constants.PG_ADAPTER).get_meta_plot(Constants.P_ADAPTER).xlabel)
ax.set_ylabel(meta_rpt.get_meta_plotgroup(Constants.PG_ADAPTER).get_meta_plot(Constants.P_ADAPTER).ylabel)
png_fn = os.path.join(output_dir,
"interAdapterDist{i}.png".format(i=i))
png_base, thumbnail_base = save_figure_with_thumbnail(fig, png_fn,
dpi=dpi)
# build the report:
plots.append(Plot("adapter_xml_plot_{i}".format(i=i),
os.path.relpath(png_base, output_dir),
thumbnail=os.path.relpath(thumbnail_base, output_dir)))
plot_groups = [PlotGroup(Constants.PG_ADAPTER,
plots=plots,
thumbnail=os.path.relpath(thumbnail_base, output_dir))]
attributes = [Attribute(i, v) for i,v in
zip([Constants.A_DIMERS, Constants.A_SHORT_INSERTS],
[adapter_dimers, short_inserts])]
tables = []
report = Report(meta_rpt.id,
title=meta_rpt.title,
attributes=attributes,
tables=tables,
)#plotgroups=plot_groups)
return meta_rpt.apply_view(report)
def get_fixed_bin_dataset():
sfn = get_fixed_bin_sts()
sset = SubreadSet()
sset.loadStats(sfn)
return sset
def test_stats_metadata(self):
ds = DataSet(data.getBam())
ds.loadStats(data.getStats())
self.assertEqual(ds.metadata.summaryStats.prodDist.numBins, 4)
self.assertEqual(ds.metadata.summaryStats.prodDist.bins,
[1576, 901, 399, 0])
ds1 = DataSet(data.getXml(8))
ds1.loadStats(data.getStats())
ds2 = DataSet(data.getXml(11))
ds2.loadStats(data.getStats())
ds3 = ds1 + ds2
self.assertEqual(ds1.metadata.summaryStats.prodDist.bins,
[1576, 901, 399, 0])
self.assertEqual(ds2.metadata.summaryStats.prodDist.bins,
[1576, 901, 399, 0])
self.assertEqual(ds3.metadata.summaryStats.prodDist.bins,
[3152, 1802, 798, 0])
self.assertEqual(ds1.metadata.summaryStats.readLenDist.bins, [
0, 62, 39, 36, 29, 37, 19, 29, 37, 32, 32, 40, 45, 54, 73, 77, 97,
95, 49, 17, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0
])
self.assertEqual(ds2.metadata.summaryStats.readLenDist.bins, [
0, 62, 39, 36, 29, 37, 19, 29, 37, 32, 32, 40, 45, 54, 73, 77, 97,
95, 49, 17, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0
])
self.assertEqual(ds3.metadata.summaryStats.readLenDist.bins, [
0, 124, 78, 72, 58, 74, 38, 58, 74, 64, 64, 80, 90, 108, 146, 154,
194, 190, 98, 34, 4, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0
])
# Lets check some manual values
ds1 = DataSet(data.getXml(8))
ds1.loadStats(data.getStats())
ds2 = DataSet(data.getXml(11))
ds2.loadStats(data.getStats())
ds1.metadata.summaryStats.readLenDist.bins = ([
0, 10, 9, 8, 7, 6, 4, 2, 1, 0, 0, 1
])
self.assertEqual(ds1.metadata.summaryStats.readLenDist.bins,
[0, 10, 9, 8, 7, 6, 4, 2, 1, 0, 0, 1])
ds1.metadata.summaryStats.readLenDist.minBinValue = 10
ds1.metadata.summaryStats.readLenDist.binWidth = 10
ds2.metadata.summaryStats.readLenDist.bins = ([
0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1
])
self.assertEqual(ds2.metadata.summaryStats.readLenDist.bins,
[0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1])
ds2.metadata.summaryStats.readLenDist.minBinValue = 20
ds2.metadata.summaryStats.readLenDist.binWidth = 10
ds3 = ds1 + ds2
self.assertEqual(ds3.metadata.summaryStats.readLenDist.bins,
[0, 10, 10, 9, 8, 7, 5, 3, 2, 1, 0, 1, 1])
# now lets swap
ds1 = DataSet(data.getXml(8))
ds1.loadStats(data.getStats())
ds2 = DataSet(data.getXml(11))
ds2.loadStats(data.getStats())
ds1.metadata.summaryStats.readLenDist.bins = ([
0, 10, 9, 8, 7, 6, 4, 2, 1, 0, 0, 1
])
self.assertEqual(ds1.metadata.summaryStats.readLenDist.bins,
[0, 10, 9, 8, 7, 6, 4, 2, 1, 0, 0, 1])
ds1.metadata.summaryStats.readLenDist.minBinValue = 20
ds1.metadata.summaryStats.readLenDist.binWidth = 10
ds2.metadata.summaryStats.readLenDist.bins = ([
0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1
])
self.assertEqual(ds2.metadata.summaryStats.readLenDist.bins,
[0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1])
ds2.metadata.summaryStats.readLenDist.minBinValue = 10
ds2.metadata.summaryStats.readLenDist.binWidth = 10
ds3 = ds1 + ds2
self.assertEqual(ds3.metadata.summaryStats.readLenDist.bins,
[0, 1, 11, 10, 9, 8, 7, 5, 3, 1, 0, 1, 1])
# now lets do some non-overlapping
ds1 = DataSet(data.getXml(8))
ds1.loadStats(data.getStats())
ds2 = DataSet(data.getXml(11))
ds2.loadStats(data.getStats())
ds1.metadata.summaryStats.readLenDist.bins = ([1, 1, 1])
self.assertEqual(ds1.metadata.summaryStats.readLenDist.bins, [1, 1, 1])
ds1.metadata.summaryStats.readLenDist.minBinValue = 10
ds1.metadata.summaryStats.readLenDist.binWidth = 10
ds2.metadata.summaryStats.readLenDist.bins = ([2, 2, 2])
self.assertEqual(ds2.metadata.summaryStats.readLenDist.bins, [2, 2, 2])
ds2.metadata.summaryStats.readLenDist.minBinValue = 50
ds2.metadata.summaryStats.readLenDist.binWidth = 10
ds3 = ds1 + ds2
self.assertEqual(ds3.metadata.summaryStats.readLenDist.bins,
[1, 1, 1, 0, 2, 2, 2])
# now lets test the subdataset metadata retention:
ss = SubreadSet(data.getXml(10))
ss.loadStats(data.getStats(0))
ss.loadStats(data.getStats(1))
self.assertEqual(153168.0, ss.metadata.summaryStats.numSequencingZmws)
self.assertEqual(
2876.0, ss.subdatasets[0].metadata.summaryStats.numSequencingZmws)
self.assertEqual(
150292.0,
ss.subdatasets[1].metadata.summaryStats.numSequencingZmws)