print(mod)
ti = time.time()
if settings[mod].mod_category == 'mc':
f_mat = hvftrs_f.format(mod, 'tsv')
gxc_hvftrs[mod] = pd.read_csv(f_mat, sep='\t', header=0, index_col=0)
print(gxc_hvftrs[mod].shape, time.time() - ti)
assert np.all(
gxc_hvftrs[mod].columns.values == metas[mod].index.values
) # make sure cell name is in the sanme order as metas (important if save knn mat)
continue
f_mat = hvftrs_f.format(mod, 'npz')
f_gene = hvftrs_gene.format(mod)
f_cell = hvftrs_cell.format(mod)
_gxc_tmp = snmcseq_utils.load_gc_matrix(f_gene, f_cell, f_mat)
_gene = _gxc_tmp.gene
_cell = _gxc_tmp.cell
_mat = _gxc_tmp.data
gxc_hvftrs[mod] = GC_matrix(_gene, _cell, _mat)
assert np.all(
gxc_hvftrs[mod].cell == metas[mod].index.values
) # make sure cell name is in the sanme order as metas (important if save knn mat)
print(gxc_hvftrs[mod].data.shape, time.time() - ti)
resolutions = [
0.01, 0.02, 0.05, 0.1, 0.2, 0.5, 1, 2, 3, 4, 6, 8, 12, 16, 20, 30, 40, 60,
80, 100, 120
]
# ns = [1000, 2000, 5000, 10000, 20000, 50000, 100000, 200000]
print(mod)
## read data
# read metadata
normalization_option = normalization_options[mod]
f_meta = f_meta_format.format(SRC_DIR, mod) ##
meta = pd.read_csv(f_meta, sep="\t", index_col=0)
metas[mod] = meta
f_data = f_data_format.format(SRC_DIR, mod, '', 'npz')
f_data_gene = f_data_format.format(SRC_DIR, mod, '', 'gene')
f_data_cell = f_data_format.format(SRC_DIR, mod, '', 'cell')
# read counts matrix
print(mod, "Reading in files {}".format(time.time()-ti))
gxc_raw = snmcseq_utils.load_gc_matrix(f_data_gene, f_data_cell, f_data) # checked dimensions in agreement internally
gxc_raws[mod] = gxc_raw
num_cells = len(meta)
num_reads = gxc_raw.data.sum().sum()/num_cells
num_reads_all[mod] = num_reads
print(gxc_raw.data.shape, num_cells, num_reads)
# check meta cells agree with gxc cells
assert np.all(meta.index.values == gxc_raw.cell)
# check genes are uniq
assert len(gxc_raw.gene) == len(np.unique(gxc_raw.gene))
print(mod, "Total time used: {}".format(time.time()-ti))
gxc_raws = collections.OrderedDict()
for mod in mods_selected:
logging.info("Read data {}...".format(mod))
if settings[mod].mod_category == 'mc':
f_gene = raw_f.format(DATA_DIR, mod, '', 'gene')
f_cell = raw_f.format(DATA_DIR, mod, '', 'cell')
f_data_c = raw_f.format(DATA_DIR, mod, 'CH_', 'npz')
f_data_mc = raw_f.format(DATA_DIR, mod, 'mCH_', 'npz')
gxc_raws[mod] = snmcseq_utils.load_gc_matrix_methylation(
f_gene, f_cell, f_data_mc, f_data_c)
else:
f_gene = raw_f.format(DATA_DIR, mod, '', 'gene')
f_cell = raw_f.format(DATA_DIR, mod, '', 'cell')
f_data = raw_f.format(DATA_DIR, mod, '', 'npz')
gxc_raws[mod] = snmcseq_utils.load_gc_matrix(f_gene, f_cell, f_data)
# In[13]:
f = output_clst_and_umap
first_round_cluster_col = 'cluster_joint_r0.1'
df_info = pd.read_csv(
f, sep="\t", index_col='sample')[[first_round_cluster_col, 'modality']]
print(df_info.shape)
df_info.head()
# In[20]:
normalization_options = {
'smarter_nuclei': 'TPM',
'smarter_cells': 'TPM',