-
Notifications
You must be signed in to change notification settings - Fork 6
/
Copy pathdex_test_leave_out_last.py
288 lines (226 loc) · 10.5 KB
/
dex_test_leave_out_last.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
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
import os, sys
def get_immediate_subdirectories(dir):
return [name for name in os.listdir(dir) if os.path.isdir(os.path.join(dir, name))]
sys.path += get_immediate_subdirectories("./")
from datetime import datetime
from DataStore import *
from ReadData import *
from JobManager import *
from Network import *
from Generate_Grid import *
from AnalyzeResults import *
from Helpers import *
from mcz import *
from dfg4grn import *
from ReadConfig import *
from clr import *
from genie3 import *
from tlclr import *
from inferelator import *
def get_golden_genes(filename):
gold_genes = []
f = open(filename, 'r').readlines()
f.pop(0) # Burn first line
for line in f:
gold_genes.append(line.split("\t")[0])
return gold_genes
golden_genes = get_golden_genes("datasets/RootArrayData/NitrogenGenes.txt")
# Instantsiate settings file
settings = {}
settings = ReadConfig(settings)
settings["global"]["working_dir"] = os.getcwd() + '/'
settings["global"]["experiment_name"] = "Dex-" + sys.argv[1]
# Create date string to append to output_dir
t = datetime.now().strftime("%Y-%m-%d_%H.%M.%S")
settings["global"]["output_dir"] = settings["global"]["output_dir"] + "/" + \
settings["global"]["experiment_name"] + "-" + t + "/"
os.mkdir(settings["global"]["output_dir"])
# Read in the gold standard network
# Read in the gold standard network
#goldnet.read_goldstd(settings["global"]["large_network_goldnet_file"])
#ko_file, kd_file, ts_file, wt_file, mf_file, goldnet = get_example_data_files(sys.argv[1], settings)
# Read data into program
# Where the format is "FILENAME" "DATATYPE"
dex_storage = ReadData("datasets/RootArrayData/DexRatios.csv", "dex")
cnlo_storage = ReadData("datasets/RootArrayData/Root_CNLO_Krouk.txt", "dex")
cnlo_no3_storage = ReadData("datasets/RootArrayData/Root_CNLO_Krouk.txt", "dex")
no3_1_storage = ReadData("datasets/RootArrayData/Root_NO3_Wang03.txt", "dex")
no3_2_storage = ReadData("datasets/RootArrayData/Root_NO3_Wang04.txt", "dex")
no3_3_storage = ReadData("datasets/RootArrayData/Root_NO3_Wang07.txt", "dex")
#ts_storage = ReadData("datasets/RootArrayData/Root_WT_Krouk11.txt", "dex")
tfs_file = open("datasets/RootArrayData/tfs.csv", 'r')
line = tfs_file.readlines()[0]
tfs = line.strip().split(',')
tfs = [x.upper() for x in tfs]
kno3_1 = ReadData("datasets/RootArrayData/KNO3norm1.csv", "dex")
kno3_2 = ReadData("datasets/RootArrayData/KNO3norm2.csv", "dex")
kno3_3 = ReadData("datasets/RootArrayData/KNO3norm3.csv", "dex")
kno3_4 = ReadData("datasets/RootArrayData/KNO3norm4.csv", "dex")
dex_storage.filter(kno3_1.gene_list)
cnlo_storage.filter(kno3_1.gene_list)
cnlo_no3_storage.filter(kno3_1.gene_list)
no3_1_storage.filter(kno3_1.gene_list)
no3_2_storage.filter(kno3_1.gene_list)
no3_3_storage.filter(kno3_1.gene_list)
no3_storage = no3_1_storage
no3_storage.combine(no3_2_storage)
no3_storage.combine(no3_3_storage)
cnlo_no3_storage.combine(no3_storage)
#all_storage.combine(cnlo_no3_storage)
#dex_storage.combine(cnlo_storage)
#dex_storage.combine(no3_storage)
#dex_storage.normalize()
no3_storage.normalize()
cnlo_storage.normalize()
cnlo_no3_storage.normalize()
#all_storage.normalize()
# Set delta_t to be without the last time point
settings["global"]["time_series_delta_t"] = "3 3 3 3 3"
# Remove the last time point from each of these
ts_storage = [kno3_1, kno3_2, kno3_3, kno3_4]
print ts_storage[1].experiments
print ts_storage[1].experiments[0].name
print ts_storage[1].experiments[len(ts_storage[1].experiments)-1].ratios
print len(ts_storage[1].experiments)
for i,d in enumerate(ts_storage):
ts_storage[i].experiments = d.experiments[0:len(d.experiments)-2]
print len(ts_storage[1].experiments)
#for s in ts_storage:
#s.normalize()
# Setup job manager
jobman = JobManager(settings)
# Make BANJO jobs
#mczjob = MCZ()
#mczjob.setup(knockout_storage, wildtype_storage, settings, timeseries_storage, knockdown_storage, "MCZ")
#jobman.queueJob(mczjob)
clr_cnlojob = CLR()
clr_cnlojob.setup(cnlo_storage, settings, "clr_cnlo")
jobman.queueJob(clr_cnlojob)
genie3_cnlojob = GENIE3()
genie3_cnlojob.setup(cnlo_storage, settings, "genie3_cnlo")
jobman.queueJob(genie3_cnlojob)
clr_no3job = CLR()
clr_no3job.setup(no3_storage, settings, "clr_no3")
jobman.queueJob(clr_no3job)
genie3_no3job = GENIE3()
genie3_no3job.setup(no3_storage, settings, "genie3_no3")
jobman.queueJob(genie3_no3job)
clr_cnlo_no3job = CLR()
clr_cnlo_no3job.setup(cnlo_no3_storage, settings, "clr_cnlo_no3")
jobman.queueJob(clr_cnlo_no3job)
genie3_cnlo_no3job = GENIE3()
genie3_cnlo_no3job.setup(cnlo_no3_storage, settings, "genie3_cnlo_no3")
jobman.queueJob(genie3_cnlo_no3job)
#tlclrjob = TLCLR()
#dex_storage.experiments = [dex_storage.experiments[0]]
#tlclrjob.setup(None, dex_storage, settings, ts_storage, None, "tlclr_dex_test")
#jobman.queueJob(tlclrjob)
print jobman.queue
jobman.runQueue()
jobman.waitToClear()
algorithms = ["dfg"]
wildtype = ["cnlo", "no3", "cnlo+no3"]
lambda_w = [0.1, 0.5]
use_dex = [True]
randnet = generate_random_network(kno3_1.gene_list)
# Build dex networks
#dex_clrnet = apply_dex("At1g13300", dex_storage, clrjob.network)
#dex_genie3net = apply_dex("At1g13300", dex_storage, genie3job.network)
# Clip our normal networks
#clrjob.network.set_top_edges_percent(int(sys.argv[2]))
#genie3job.network.set_top_edges_percent(int(sys.argv[2]))
for d in use_dex:
for alg in algorithms:
if d:
# Pure DFG, no
dfg = DFG4GRN()
settings["dfg4grn"]["eta_z"] = 0.1
settings["dfg4grn"]["lambda_w"] = 0.1
settings["dfg4grn"]["tau"] = 3
settings["dfg4grn"]["delta_t"] = "3 3 3 3 3"
dfg.setup(ts_storage, TFList(tfs), settings, "DFG-NO_WT_LambdaW-{0}_Dex-{1}".format(0.1, d), 20, None, None, None, False, dex_storage, "At1g13300")
jobman.queueJob(dfg)
else:
dfg = DFG4GRN()
settings["dfg4grn"]["eta_z"] = 0.1
settings["dfg4grn"]["lambda_w"] = 0.1
settings["dfg4grn"]["tau"] = 3
settings["dfg4grn"]["delta_t"] = "3 3 3 3 3"
dfg.setup(ts_storage, TFList(tfs), settings, "DFG-NO_WT_LambdaW-{0}_Dex-{1}".format(0.1, d), 20)
jobman.queueJob(dfg)
for wt in wildtype:
if wt == "cnlo":
genie3_network = genie3_cnlojob.network.copy()
clr_network = clr_cnlojob.network.copy()
elif wt == "no3":
genie3_network = genie3_no3job.network.copy()
clr_network = clr_no3job.network.copy()
else:
genie3_network = genie3_cnlo_no3job.network.copy()
clr_network = clr_cnlo_no3job.network.copy()
clr_network.normalize()
genie3_network.normalize()
if d:
genie3_network = apply_dex("At1g13300", dex_storage, genie3_network, sys.argv[2])
clr_network = apply_dex("At1g13300", dex_storage, clr_network, sys.argv[2])
if alg == "dfg":
dfg = DFG4GRN()
settings["dfg4grn"]["eta_z"] = 0.01
settings["dfg4grn"]["lambda_w"] = 0.5
settings["dfg4grn"]["tau"] = 3
settings["dfg4grn"]["delta_t"] = "3 3 3 3 3"
dfg.setup(ts_storage, TFList(tfs), settings, "DFG-GENIE3_Type-{0}_LambdaW-{1}_Dex-{2}".format(wt, 0.5, d), 20, None, genie3_network, "weights")
jobman.queueJob(dfg)
#dfg = DFG4GRN()
#settings["dfg4grn"]["eta_z"] = 0.01
#settings["dfg4grn"]["lambda_w"] = 0.5
#settings["dfg4grn"]["tau"] = 3
#settings["dfg4grn"]["delta_t"] = "3 3 3 3 3"
#dfg.setup(ts_storage, TFList(tfs), settings, "DFG-CLR_Type-{0}_LambdaW-{1}_Dex-{2}".format(wt, 0.5, d), 20, None, clr_network, "weights")
#jobman.queueJob(dfg)
# Inferelator mode
#dfg = DFG4GRN()
#settings["dfg4grn"]["eta_z"] = 0.1
#settings["dfg4grn"]["lambda_w"] = 0.005
#settings["dfg4grn"]["tau"] = 3
#settings["dfg4grn"]["delta_t"] = "3 3 3 3 3"
#dfg.setup(ts_storage, TFList(tfs), settings, "Inferelator-CLR_Type-{0}_LambdaW-{1}_Dex-{2}".format(wt, 0.005, d), 20, None, clr_network, "weights", True)
#jobman.queueJob(dfg)
## Inferelator mode
#dfg = DFG4GRN()
#settings["dfg4grn"]["eta_z"] = 0.1
#settings["dfg4grn"]["lambda_w"] = 0.005
#settings["dfg4grn"]["tau"] = 3
#settings["dfg4grn"]["delta_t"] = "3 3 3 3 3"
#dfg.setup(ts_storage, TFList(tfs), settings, "Inferelator-GENIE3_Type-{0}_LambdaW-{1}_Dex-{2}".format(wt, 0.005, d), 20, None, genie3_network, "weights", True)
#jobman.queueJob(dfg)
#dfg = DFG4GRN()
#settings["dfg4grn"]["eta_z"] = 0.1
#settings["dfg4grn"]["lambda_w"] = 0.1
#settings["dfg4grn"]["tau"] = 3
#settings["dfg4grn"]["delta_t"] = "3 3 3 3 3"
#dfg.setup(ts_storage, TFList(tfs), settings, "DFG-CLR_DEFAULT_Type-{0}_LambdaW-{1}_Dex-{2}".format(wt, 0.1, d), 20, None, clr_network, "weights")
#jobman.queueJob(dfg)
dfg = DFG4GRN()
settings["dfg4grn"]["eta_z"] = 0.1
settings["dfg4grn"]["lambda_w"] = 0.1
settings["dfg4grn"]["tau"] = 3
settings["dfg4grn"]["delta_t"] = "3 3 3 3 3"
dfg.setup(ts_storage, TFList(tfs), settings, "DFG-GENIE3_DEFAULT_Type-{0}_LambdaW-{1}_Dex-{2}".format(wt, 0.1, d), 20, None, genie3_network, "weights")
jobman.queueJob(dfg)
jobman.runQueue()
jobman.waitToClear()
#tprs, fprs, rocs = GenerateMultiROC(jobman.finished, goldnet, False, settings["global"]["output_dir"] + "/OverallROC.pdf")
#ps, rs, precs = GenerateMultiPR(jobman.finished, goldnet, False, settings["global"]["output_dir"] + "/OverallPR.pdf")
SaveResults(jobman.finished, [], settings)
results = []
for job in jobman.finished:
if job.alg.alg_name == "dfg4grn":
results.append([job.alg.name, job.alg.best_sign, job.alg.best_sign_all, job.alg.avg_sign, job.alg.avg_sign_all])
out = open(settings["global"]["output_dir"] + "/results.txt",'w')
for r in results:
out.write(r[0] + "\t" + str(r[1]) + "\t" + str(r[2]) + "\t"+str(r[3])+"\t"+str(r[4])+"\n")
out = open(settings["global"]["output_dir"] + "/results-sorted.txt",'w')
results.sort(key=lambda x: x[0])
for r in results:
out.write(r[0] + "\t" + str(r[1]) + "\t" + str(r[2]) + "\t"+str(r[3])+"\t"+str(r[4])+"\n")