-
Notifications
You must be signed in to change notification settings - Fork 1
/
filter_out_cluster_cells.py
192 lines (153 loc) · 6.95 KB
/
filter_out_cluster_cells.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
import pandas as pd
import scanpy as sc
import numpy as np
import stlearn as st
import scipy
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
import os
import sys
from h5py import Dataset, Group
import qnorm
#from sklearn.preprocessing import quantile_transform
import pickle
from scipy import sparse
import pickle
import scipy.linalg
from sklearn.metrics.pairwise import euclidean_distances
import gseapy as gp
import csv
import stlearn as st
from collections import defaultdict
#################### get the whole training dataset
#rootPath = os.path.dirname(sys.path[0])
#os.chdir(rootPath+'/CCST')
print("hello world!")
def read_h5(f, i=0):
print("hello world! read_h5")
for k in f.keys():
if isinstance(f[k], Group):
print('Group', f[k])
print('-'*(10-5*i))
read_h5(f[k], i=i+1)
print('-'*(10-5*i))
elif isinstance(f[k], Dataset):
print('Dataset', f[k])
print(f[k][()])
else:
print('Name', f[k].name)
print("hello world! read_h5_done")
#if __name__ == "__main__":
import argparse
parser = argparse.ArgumentParser()
parser.add_argument( '--data_path', type=str, default='/cluster/home/t116508uhn/64630/spaceranger_output_new/' , help='The path to dataset') #'/cluster/projects/schwartzgroup/fatema/pancreatic_cancer_visium/210827_A00827_0396_BHJLJTDRXY_Notta_Karen/V10M25-61_D1_PDA_64630_Pa_P_Spatial10x_new/outs/'
parser.add_argument( '--data_name', type=str, default='V10M25-61_D1_PDA_64630_Pa_P_Spatial10x_new', help='The name of dataset')
parser.add_argument( '--generated_data_path', type=str, default='generated_data/', help='The folder to store the generated data')
args = parser.parse_args()
# main(args)
def main(args):
print("hello world! main")
toomany_label_file='/cluster/home/t116508uhn/64630/PCA_64embedding_pathologist_label_l1mp5_temp.csv'#'/cluster/home/t116508uhn/64630/TAGConv_test_r4_too-many-cell-clusters_org.csv' #'/cluster/home/t116508uhn/64630/PCA_64embedding_pathologist_label_l1mp5_temp.csv'
toomany_label=[]
with open(toomany_label_file) as file:
csv_file = csv.reader(file, delimiter=",")
for line in csv_file:
toomany_label.append(line)
barcode_label=dict()
cluster_dict=dict()
for i in range (1, len(toomany_label)):
if len(toomany_label[i])>0 :
barcode_label[toomany_label[i][0]] = int(toomany_label[i][1])
cluster_dict[int(toomany_label[i][1])]=1
pathologist_label_file='/cluster/home/t116508uhn/64630/tumor_64630_D1_IX_annotation.csv' #IX_annotation_artifacts.csv' #
pathologist_label=[]
with open(pathologist_label_file) as file:
csv_file = csv.reader(file, delimiter=",")
for line in csv_file:
pathologist_label.append(line)
barcode_tumor=dict()
for i in range (1, len(pathologist_label)):
if pathologist_label[i][1] == 'tumor': #'Tumour':
barcode_tumor[pathologist_label[i][0]] = 1
barcode_file='/cluster/home/t116508uhn/64630/spaceranger_output_new/unzipped/barcodes.tsv' # 1406
barcode_info=[]
#barcode_info.append("")
i=0
with open(barcode_file) as file:
tsv_file = csv.reader(file, delimiter="\t")
for line in tsv_file:
barcode_info.append([line[0],-1,[]])
i=i+1
#cluster_dict[-1]=1
cluster_label=list(cluster_dict.keys())
count=0
for i in range (0, len(barcode_info)):
if (barcode_info[i][0] in barcode_label) and (barcode_info[i][0] in barcode_tumor):
barcode_info[i][1] = barcode_label[barcode_info[i][0]]
else:
count=count+1
data_fold = args.data_path #+args.data_name+'/'
print(data_fold)
adata_h5 = st.Read10X(path=data_fold, count_file='filtered_feature_bc_matrix.h5') #count_file=args.data_name+'_filtered_feature_bc_matrix.h5' )
print(adata_h5)
sc.pp.filter_genes(adata_h5, min_cells=1)
print(adata_h5)
#sc.pp.highly_variable_genes(adata_h5,flavor='seurat_v3', n_top_genes = 500)
#print(adata_h5)
#adata_X =
#sc.pp.normalize_total(adata_h5, target_sum=1) #, exclude_highly_expressed=True)
#print(adata_h5)
#adata_X =
#sc.pp.scale(adata_h5)
#print(adata_h5)
gene_list_all=scipy.sparse.csr_matrix.toarray(adata_h5.X) # row = cells x genes # (1406, 36601)
gene_ids = list(adata_h5.var_names) # 36601
#cell_genes = defaultdict(list)
for cell_index in range (0, gene_list_all.shape[0]):
gene_list_temp=defaultdict(list)
non_zero_index=list(np.where(gene_list_all[cell_index]!=0)[0])
for gene_index in non_zero_index:
if gene_list_all[cell_index][gene_index]>0:
gene_list_temp[gene_ids[gene_index]].append(gene_list_all[cell_index][gene_index])
barcode_info[cell_index][2]=gene_list_temp
target_cluster_id = [11, 12] #[60, 61] #[14, 15] #BB
gene_list_cluster=[]
for i in range (0, len(barcode_info)):
if barcode_info[i][1] in target_cluster_id:
gene_list_cluster = gene_list_cluster + barcode_info[i][2]
gene_list_cluster = list(set(gene_list_cluster))
'''diff_file=['/cluster/home/t116508uhn/64630/differential_TAGConv_test_r4_10_13_org_whitelist.csv',
'/cluster/home/t116508uhn/64630/differential_TAGConv_test_r4_10_59_org_whitelist.csv',
'/cluster/home/t116508uhn/64630/differential_TAGConv_test_r4_10_86_org_whitelist.csv'
]#
gene_list_cluster=[]
for i in range (0, len(diff_file)):
with open(diff_file[i]) as file:
tsv_file = csv.reader(file, delimiter=",")
j=0
for line in tsv_file:
if j==0:
j=j+1
continue
if len(line)>0 and np.float(line[1]) !=-np.inf and np.float(line[1]) < 0 : #and np.float(line[3]) < 0.05:
print(np.float(line[1]))
gene_list_cluster.append(line[0])
j=j+1
gene_list_cluster = list(set(gene_list_cluster)) '''
signature_file='/cluster/home/t116508uhn/64630/GeneList_KF_22Aug10.csv' # 1406
signature_info=defaultdict(list)
#barcode_info.append("")
with open(signature_file) as file:
tsv_file = csv.reader(file, delimiter=",")
for line in tsv_file:
if (line[0].find('Basal') > -1) or (line[0].find('Classical') > -1) :
signature_info[line[0]].append(line[1])
signature_info=dict(signature_info)
enr = gp.enrichr(gene_list=gene_list_cluster,
gene_sets=signature_info,
background=19523,
#organism='human', # don't forget to set organism to the one you desired! e.g. Yeast
outdir=None # don't write to disk
)
enr.results