def extract_closest(
fname=None,
df=None,
):
''' Parse the output of 'bedtools closest'
'''
if df is None:
df = pyutil.readData(fname, header=None, ext='tsv', guess_index=False)
# df = df.dropna()
header = bedHeader + pyutil.paste0([['feature_'], bedHeader]).tolist()
df = df.iloc[:, :18]
df.columns = header[:17] + ['distance']
df['hit'] = df['feature_acc']
return df
print fnames
import sys
vals = map(sptn.getBrachy, fnames)
bnames = map(lambda x: x.split('/')[-1].rsplit('.', 2)[0].replace('_', '-'),
fnames)
for i, d in enumerate(vals):
d['Alias'] = bnames[i]
flats = pyutil.meta2flat([[(x, d.pop(x))
for x in ['RunID', 'SampleID', 'Alias']]
for d in vals])
nonEmptyVals = [{k: v for k, v in d.items() if v is not None} for d in vals]
conds = pyutil.meta2flat(nonEmptyVals)
colNames = pyutil.paste0([flats, conds])
print colNames
#print flats,conds
#print bnames
#print vals
#sys.exit(0)
def callback(df):
df = pyutil.filterMatch(df, key='STRG', negate=1)
return df
df = pyutil.Table2Mat(fnames,
callback=callback,
valCol='TPM',
vals = map(getLight, fnames)
meta['light'] = vals
vals = map(getGenotype, fnames)
vals = ['Bd21' if x == 'BdWT' else x for x in vals]
meta['gtype'] = vals
meta = pd.DataFrame(meta)
meta.loc[meta['RunID'] == '143R', 'light'] = 'LD'
# meta[meta['RunI']]
meta['Age_int'] = map(Rid2Age.get, meta['RunID'])
meta['Age'] = ['Wk%d' % x for x in meta['Age_int']]
#### Discussed with Mingjun on June 6th about 143R
# meta['ZTime'][[4,5]] = 'ZT0'
meta.loc[(meta['RunID'] == '143R') & (meta['sampleID'].isin(['S5', 'S6'])),
'ZTime'] = 'ZT0'
# meta['ZTime'][[4,5]] = 'ZT0'
vals = [int(re.sub('[^-\d]', '', x)) for x in meta['ZTime']]
vals = [24 + x if x < 0 else x for x in vals]
meta['ZTime_int'] = vals
meta['ZTime'] = pyutil.paste0([['ZT'] * len(meta), meta['ZTime_int']])
meta['fname'] = list(fnames)
meta.to_csv('meta.csv')
meta
# sys.exit(0)
def meta2name(meta, keys=['gtype', 'light', 'Age', 'ZTime']):
res = pyutil.paste0([meta[k] for k in keys], '_')
return res