# WTCHG ALL samples
objwtchg_all = rnaseq_data.all_hgic_loader(annotate_by='Ensembl Gene ID',
include_derived=True)
to_keep_wtchg = (
'GIBCO_NSC_P4',
# 'DURA018_NSC_N2_P6',
# 'DURA018_NSC_N4_P4',
'DURA019_NSC_N8C_P2',
'DURA030_NSC_N16B6_P1',
'DURA031_NSC_N44B_P2')
# rRNA gene IDs
rrna_ensg = set(gtf_reader.get_rrna())
# MT gene_ids
mt_ensg = set(gtf_reader.get_mitochondrial())
# combine the data
data = pd.concat((obj73721.data.loc[:, to_keep73721], obj61794.data,
objwtchg_all.data.loc[:, to_keep_wtchg]),
axis=1)
# combine the metadata
meta = pd.concat((
obj73721.meta.loc[to_keep73721],
obj61794.meta,
objwtchg_all.meta.loc[to_keep_wtchg, :],
),
axis=0)
data = data.loc[data.index.str.contains('ENSG')]
eps = .01
elif units == 'estimated_counts':
eps = .01
elif units == 'cpm':
eps = .01
if source == 'star' and units == 'tpm':
raise NotImplementedError("Unable to convert STAR counts into TPM.")
if transform in {'vst', 'rlog'
} and units not in {'counts', 'estimated_counts'}:
raise AttributeError(
"vst and rlog transformations require counts as input")
if remove_mt:
mt_ensg = set(gtf_reader.get_mitochondrial('GRCm38r88'))
loc = loader.RnaSeqFileLocations(root_dir=os.path.join(
loader.RNASEQ_DIR, 'wtchg_p170390'),
alignment_subdir='mouse',
batch_id='wtchg_p170390')
kwargs = loc.loader_kwargs(source)
kwargs.update(
dict(tax_id=10090,
samples=[
u'mDura3N1human', u'mDura5N24Ahuman', u'mDura6N6human',
u'mDura3N1mouse', u'mDura5N24Amouse', u'mDura6N6mouse',
u'eNSC3mouse', u'eNSC5mouse', u'eNSC6mouse'
]))
kwargs.update(load_kwargs)