def _run(self):
kw = self._data
BIGWIG_PAIRS = kw['BIGWIG_PAIRS']
BIGWIG_FILES = kw['BIGWIG_FILES']
BIGWIG_NAMES = kw['BIGWIG_NAMES']
PARAMS = kw['PARAMS']
FORCE = kw['FORCE']
NCORE = kw['NCORE']
# PARAMS['NCORE'] = kw['NCORE']
# kw['LAST_FILE'] = kw['OFNAME'] = OFNAME = kw['OFNAME'].realpath()
# kw['LAST_DIR'] = OFNAME.dirname()
kw['PEAK_FILE'] = PEAK_FILE = kw['PEAK_FILE'].realpath()
assert kw['OUTDIR']
kw['OUTDIR'] = OUTDIR = kw['OUTDIR'].realpath()
AXIS_DICTS = kw['AXIS_DICTS']
if BIGWIG_PAIRS:
assert not BIGWIG_NAMES and not BIGWIG_FILES, (
'conflict arguments')
BIGWIG_FILES, BIGWIG_NAMES = zip(BIGWIG_PAIRS)
pass
else:
assert BIGWIG_FILES
if not BIGWIG_NAMES:
BIGWIG_NAMES = map(pyext.getBname, BIGWIG_FILES)
with pyext.getPathStack([OUTDIR], force=1):
OFNAME = 'DONE'
# OFNAME =
if not FORCE and pyext.file__notEmpty(OFNAME):
pass
else:
figs, (bwTable, bwAvg) = sjob.figs__peakBW(
peakFile=PEAK_FILE,
bwFiles=BIGWIG_FILES,
outIndex=None if not len(BIGWIG_NAMES) else BIGWIG_NAMES,
NCORE=NCORE,
**PARAMS)
fig = figs.values()[0]
for (axis, d_ax) in zip(fig.axes, AXIS_DICTS):
pyext.obj__dict__call(axis, d_ax)
for OFNAME in ['MAIN.png', 'MAIN.svg']:
plotters.fig__save(fig, OFNAME)
bwTable.to_pickle('bwTable.pk')
bwAvg.to_pickle('bwAvg.pk')
pyext.printlines([pyext.dppJson(kw)], "DATA.json")
pyext.printlines(['DONE'], 'DONE')
# if not OFNAME:
# if not OFNAME
# "outerRadius":500,
# "NCORE":4,
# "peakFile":'/home/feng/static/lists/1112__ELF3__chipTarg.narrowPeak',
# "bwFiles":lambda self,key,DATA_ACC_LIST,chipseq_db:
# map(chipseq_db['get__chipseq__bwfile'],DATA_ACC_LIST)
# "OFNAME":
# return
def dump_peak_quality(sample):
if not pyext.file__notEmpty(sample['npkfile']):
print (sample['data_acc'],sample['npkfile'])
template= '''
sample_accession={sample.data_acc}
number_of_peaks_below_5%FDR=NA
number_of_peaks_above_5fold_enrichment=NA
'''.strip().replace('\t','').replace(' ','')
else:
df = _readData(sample['npkfile'],columns=pyext.columns.bed,header=None)
template= '''
sample_accession={sample.data_acc}
number_of_peaks_below_5%FDR={((df['neglogQval'] > -pyext.np.log10(0.05) )).sum()}
number_of_peaks_above_5fold_enrichment={(df['FC'] > 5 ).sum()}
'''.strip().replace('\t','').replace(' ','')
return pyext.template__format(template,sample)
def sample_get_depth(sample,
OUTPUT_FILE,
FORCE=0):
if not FORCE and pyext.file__notEmpty(OUTPUT_FILE):
pass
else:
d = pyext._DICT_CLASS()
d['DATA_ACC']=sample['data_acc']
CMD = ["gzip","-d<",sample['rawfile_files_orig'][0],'|wc','-l']
res = pyext.shellexec(' '.join(CMD))
d['READ_COUNT_RAW'] = int(res.strip())//4
CMD = ["cat",sample["file_bam_orig"],
"|samtools","view","-F0x4","-F0x100","-c",
]
res = pyext.shellexec(' '.join(CMD))
d['READ_COUNT_UNIQ_MAPPED'] = int(res.strip())
with open(OUTPUT_FILE,'w') as f:
pyext.json.dump(d, f,indent=4)
return pyext.readData(OUTPUT_FILE)
import synotil.util as sutil
import pymisca.util as pyutil
from lazydict import LazyDictionary
template = job = LazyDictionary()
__FILE__ = path.Path(__file__)
DEPENDS = {
'meta_chip': (pyext.readData, '/home/feng/meta/meta_chip.tsv'),
'fig_meta': (pyext.readData, __FILE__.dirname() / '0726-figure-meta.tsv'),
'chipseq_targets_peaks_file':
(None, '/home/feng/static/lists/1112__ELF3__chipTarg.narrowPeak'),
}
for k in DEPENDS:
v = DEPENDS[k]
assert pyext.file__notEmpty(v[1]), (v[1], )
if v[0] is not None:
job[k] = v[0](v[1])
else:
job[k] = v[1]
@pyext.setItem(job, "chipseq_targets_peaks")
def _func(
self,
key,
):
pass