def _func(job, key, WORKDIR, chipseq_targets_peaks):
fig, axs = plt.subplots(2, 2, figsize=[12, 12])
axs = axs.ravel()
with pyext.getPathStack([WORKDIR, key], force=1):
GSIZE = "/home/feng/ref/ATH-TAIR10/genome.sizes"
peakFile = chipseq_targets_peaks['LAST_FILE']
featFile = sutil.bed__leftSummit(
"/home/feng/ref/ATH-TAIR10/annotation/genes.gtf.cds", GSIZE=GSIZE)
pyext.file__link(peakFile, "PEAK_FILE.bed", force=1)
pyext.file__link(featFile, "FEAT_FILE.bed", force=1)
res = synotil.qcplots.qc_summitDist(
# chipseq_targets_genes_peaks,
peakFile,
featFile,
GSIZE=GSIZE,
axs=axs,
CUTOFF=500,
)
df = res[0]
# pyext.readData(peakFile,header=None).set_index(3,drop=0).reindex(df['acc'])
# df = df.set_index('acc',drop=0).reindex(pyext.readData(peakFile,header=None).set_index(3))
df.to_csv("PEAK_DIST.csv", index=0)
lst = df.dropna()['feat_acc'].drop_duplicates().values.tolist()
with pyext.getPathStack([WORKDIR, key], force=1) as stack:
pyext.printlines(lst, "OUT.it")
return dict(LAST_DIR=WORKDIR / key)
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 _func(job, key, WORKDIR):
GSIZE = "/home/feng/ref/ATH-TAIR10/genome.sizes"
with pyext.getPathStack([WORKDIR, key], force=1) as stack:
res = sutil.bed__summit(
'/home/feng/static/lists/1112__ELF3__chipTarg.narrowPeak',
GSIZE=GSIZE,
inplace=False)
res = stack.d / res
return dict(LAST_FILE=res)
def _func(
self,
key,
OUTDIR,
fig_meta,
meta_chip,
chipseq_targets_peaks_file,
):
from pymisca.plotters import plotters
plotters.fig__save
# key = 'figS4E_o'
key = 'figS4E_0905'
df = fig_meta[[key]].dropna().astype(int)
df = df.sort_values(key)
DATA_ACC_LIST = df.index.tolist()
mcurr = meta_chip.reindex(DATA_ACC_LIST)
bwFiles = mcurr['RPKMFile']
# bedFile = '/home/feng/envs/Fig_POLYQ/bedFile.bed'
# bedFile = '/home/feng/static/lists/1112__ELF3__chipTarg.narrowPeak'
res = sutil.extract_bigwig_multiple(
bwFiles=mcurr.RPKMFile,
bedFile=chipseq_targets_peaks_file,
# radius=100,stepSize=10,NCORE=6,
radius=100,
stepSize=10,
NCORE=6,
outIndex=pyext.df__format(mcurr, '{bname}_{index}'),
)
tab = dfc = pyutil.colGroupMean(res, ).apply(sutil.log2p1)
with pyext.getPathStack([OUTDIR], force=1):
plt.figure(figsize=[12, 8])
dfc.boxplot(rot='vertical', )
plt.ylabel('average binding in RPKM')
# plt.gcf().savefig('fig1.png')
# plt.gcf().
plotters.fig__save(plt.gcf(), pyext.f('{key}.png'))
# res.head()
tab2 = res.mean(axis=0)
pile = tab2.to_frame().reset_index().pivot_table(index='bwFile',
columns='pos',
values=0)
pile = pile.T
pyvis.df__heatmap(pile, figsize=[12, 12])
# pyvis.heatmap(pile, xtick=pile.columns,ytick=pile.index)
plotters.fig__save(plt.gcf(), pyext.f('{key}_pileup_heatmap.png'))
def _func(self, key, OUTDIR, signature_targets):
df = pd.DataFrame({"rnaseq_signature_targets": signature_targets})
with pyext.getPathStack([OUTDIR], force=1):
df.to_csv("gene_lists_dataframe.csv", index=0)
# pyext.printlines(signature_targets)
pass
def export(self):
df = pd.DataFrame({"rnaseq_signature_targets": self.signature_targets})
with pyext.getPathStack([self.OUTDIR],force=1):
df.to_csv("gene_lists_dataframe.csv",index=0)
pass