def write_summary(self, summary_fn, isoforms_fa, hq_fa=None, lq_fa=None):
"""Extract number of consensus isoforms predicted, and total
number of bases in all consensuus isoforms from isoforms_fa and write
the two attributes to summary_fn.
if hq_fa (polished high-quality isoforms) is not None, report
the number of polished hq clusters
if lq_fa (polished high-quality isoforms) is not None, report
the number of polished hq clusters
"""
self.add_log("Writing a summary to {f}".format(f=summary_fn),
level=logging.INFO)
try:
summary = ClusterSummary()
with FastaReader(isoforms_fa) as reader:
for r in reader:
summary.numConsensusIsoforms += 1
summary.numTotalBases += len(r.sequence)
if hq_fa is not None:
summary.num_polished_hq_isoforms = 0
with FastaReader(hq_fa) as reader:
for r in reader:
summary.num_polished_hq_isoforms += 1
if lq_fa is not None:
summary.num_polished_lq_isoforms = 0
with FastaReader(lq_fa) as reader:
for r in reader:
summary.num_polished_lq_isoforms += 1
summary.write(summary_fn)
except ZeroDivisionError:
errMsg = "No consensus isoforms predicted."
self.add_log(errMsg, level=logging.ERROR)
raise RuntimeError(errMsg)
def test_write(self):
"""Test ClusterSummary.write."""
outFN = op.join(self.testDir, "out/test_ClusterSummary.txt")
stdoutFN = op.join(self.testDir, "stdout/test_ClusterSummary.txt")
obj = ClusterSummary()
obj.numConsensusIsoforms = 97
obj.numTotalBases = 97 * 3945
obj.write(outFN)
self.assertTrue(filecmp.cmp(outFN, stdoutFN))
def __init__(self,
root_dir,
flnc_fa,
nfl_fa,
bas_fofn,
ccs_fofn,
out_fa,
sge_opts,
ice_opts,
hq_isoforms_fa=None,
hq_isoforms_fq=None,
lq_isoforms_fa=None,
lq_isoforms_fq=None,
report_fn=None,
summary_fn=None):
super(Cluster, self).__init__(prog_name="Cluster",
root_dir=root_dir,
bas_fofn=bas_fofn,
ccs_fofn=ccs_fofn)
self.flnc_fa, self.nfl_fa, self.ccs_fofn = self._validateInputs(
flnc_fa, nfl_fa, ccs_fofn)
self.root_dir, self.out_fa = self._validateOutputs(root_dir, out_fa)
self.hq_isoforms_fa = hq_isoforms_fa
self.hq_isoforms_fq = hq_isoforms_fq
self.lq_isoforms_fa = lq_isoforms_fa
self.lq_isoforms_fq = lq_isoforms_fq
self.sge_opts = sge_opts # SGE, CPU options and etc
self.ice_opts = ice_opts # The ICE clutering algorithm options
self.sanity_check()
self._probqv = None # probability & quality value
self._flnc_splitted_fas = [] # split flnc_fa into smaller files.
self._nflncSplittedFas = [] # split nfl_fa into smaller files.
self._logConfigs() # Log configurations
self.iceinit = None
self.icec = None
self.iceq = None
self.pol = None
self.report_fn = realpath(report_fn) if report_fn is not None \
else op.join(self.root_dir, "cluster_report.csv")
self.summary_fn = realpath(summary_fn) if summary_fn is not None \
else op.join(self.root_dir, "cluster_summary.txt")
self.summary = ClusterSummary()
self.add_log("Finishing create Cluster Object.", level=logging.INFO)
