from clustergrammer2 import net
df = {}
>> clustergrammer2 backend version 0.5.1
import pandas as pd
from ast import literal_eval as make_tuple
import warnings
warnings.filterwarnings('ignore')
df_ds = pd.read_parquet('../data/cao_2million-cell_2019_61-embryo_downsample/cao_embryo_cell-type_downsample.parquet')
df_ds.columns = [make_tuple(x) for x in df_ds.columns.tolist()]
df_ds.shape
(5000, 2229)
ser_cell_types = pd.Series([x[1].split(': ')[1] for x in df_ds.columns.tolist()])
cell_type_dist = pd.read_csv('../data/cao_2million-cell_2019_61-embryo_downsample/cell_type_dist.txt', sep='\t', index_col=0, header=None)
print(cell_type_dist)
1 0 Chondrocytes & osteoblasts 104698 Connective tissue progenitors 98964 Intermediate Mesoderm 89518 Jaw and tooth progenitors 82289 Early mesenchyme 71949 Excitatory neurons 68567 Epithelial cells 66209 Radial glia 65428 Neural progenitor cells 58332 Postmitotic premature neurons 56033 Oligodendrocyte Progenitors 54606 Isthmic organizer cells 48498 Neural Tube 45985 Inhibitory neurons 44658 Myocytes 43197 Definitive erythroid lineage 34205 Chondroctye progenitors 33539 Inhibitory neuron progenitors 31214 Premature oligodendrocyte 29538 Limb mesenchyme 26559 Sensory neurons 26477 Endothelial cells 26431 Stromal cells 23259 Osteoblasts 23223 Schwann cell precursor 23145 Granule neurons 16131 Notochord cells 15481 Primitive erythroid lineage 15138 Inhibitory interneurons 13533 Hepatocytes 11229 White blood cells 9202 Ependymal cell 8566 Cholinergic neurons 7060 Cardiac muscle lineages 4867 Megakaryocytes 3572 Melanocytes 2827 Lens 1954 Neutrophils 506
cell_type_dist.sum().get_values()[0]
1386587
net.load_df(df_ds)
net.filter_N_top(inst_rc='row', N_top=250, rank_type='var')
net.normalize(axis='row', norm_type='zscore')
net.clip(-5,5)
net.widget()
ExampleWidget(network='{"row_nodes": [{"name": "Gm42418", "ini": 250, "clust": 5, "rank": 79, "rankvar": 87, "…
inst_cell_type = 'Sensory neurons'
net.load_df(df_ds)
net.filter_cat(axis='col', cat_index=1, cat_name='Cell Type: ' + inst_cell_type)
net.filter_N_top(inst_rc='row', N_top=250, rank_type='var')
net.normalize(axis='row', norm_type='zscore')
net.clip(-5,5)
net.widget()
ExampleWidget(network='{"row_nodes": [{"name": "mt-Rnr2", "ini": 250, "clust": 9, "rank": 135, "rankvar": 190,…