|
410
|
1 import argparse
|
|
|
2 import utils.general_utils as utils
|
|
|
3 from typing import Optional, List
|
|
|
4 import os
|
|
|
5 import numpy as np
|
|
|
6 import pandas as pd
|
|
|
7 import cobra
|
|
|
8 import utils.CBS_backend as CBS_backend
|
|
|
9 from joblib import Parallel, delayed, cpu_count
|
|
|
10 from cobra.sampling import OptGPSampler
|
|
|
11 import sys
|
|
419
|
12 import utils.model_utils as model_utils
|
|
410
|
13
|
|
|
14
|
|
|
15 ################################# process args ###############################
|
|
|
16 def process_args(args :List[str] = None) -> argparse.Namespace:
|
|
|
17 """
|
|
|
18 Processes command-line arguments.
|
|
|
19
|
|
|
20 Args:
|
|
|
21 args (list): List of command-line arguments.
|
|
|
22
|
|
|
23 Returns:
|
|
|
24 Namespace: An object containing parsed arguments.
|
|
|
25 """
|
|
|
26 parser = argparse.ArgumentParser(usage = '%(prog)s [options]',
|
|
|
27 description = 'process some value\'s')
|
|
|
28
|
|
|
29 parser.add_argument("-mo", "--model_upload", type = str,
|
|
|
30 help = "path to input file with custom rules, if provided")
|
|
|
31
|
|
419
|
32 parser.add_argument("-mab", "--model_and_bounds", type = str,
|
|
|
33 choices = ['True', 'False'],
|
|
|
34 required = True,
|
|
|
35 help = "upload mode: True for model+bounds, False for complete models")
|
|
|
36
|
|
|
37
|
|
410
|
38 parser.add_argument('-ol', '--out_log',
|
|
|
39 help = "Output log")
|
|
|
40
|
|
|
41 parser.add_argument('-td', '--tool_dir',
|
|
|
42 type = str,
|
|
|
43 required = True,
|
|
|
44 help = 'your tool directory')
|
|
|
45
|
|
|
46 parser.add_argument('-in', '--input',
|
|
419
|
47 required = True,
|
|
|
48 type=str,
|
|
|
49 help = 'input bounds files or complete model files')
|
|
410
|
50
|
|
419
|
51 parser.add_argument('-ni', '--name',
|
|
410
|
52 required = True,
|
|
|
53 type=str,
|
|
|
54 help = 'cell names')
|
|
|
55
|
|
|
56 parser.add_argument('-a', '--algorithm',
|
|
|
57 type = str,
|
|
|
58 choices = ['OPTGP', 'CBS'],
|
|
|
59 required = True,
|
|
|
60 help = 'choose sampling algorithm')
|
|
|
61
|
|
|
62 parser.add_argument('-th', '--thinning',
|
|
|
63 type = int,
|
|
|
64 default= 100,
|
|
|
65 required=False,
|
|
|
66 help = 'choose thinning')
|
|
|
67
|
|
|
68 parser.add_argument('-ns', '--n_samples',
|
|
|
69 type = int,
|
|
|
70 required = True,
|
|
|
71 help = 'choose how many samples')
|
|
|
72
|
|
|
73 parser.add_argument('-sd', '--seed',
|
|
|
74 type = int,
|
|
|
75 required = True,
|
|
|
76 help = 'seed')
|
|
|
77
|
|
|
78 parser.add_argument('-nb', '--n_batches',
|
|
|
79 type = int,
|
|
|
80 required = True,
|
|
|
81 help = 'choose how many batches')
|
|
|
82
|
|
430
|
83 parser.add_argument('-opt', '--perc_opt',
|
|
|
84 type = float,
|
|
|
85 default=0.9,
|
|
|
86 required = False,
|
|
|
87 help = 'choose the fraction of optimality for FVA (0-1)')
|
|
|
88
|
|
410
|
89 parser.add_argument('-ot', '--output_type',
|
|
|
90 type = str,
|
|
|
91 required = True,
|
|
|
92 help = 'output type')
|
|
|
93
|
|
|
94 parser.add_argument('-ota', '--output_type_analysis',
|
|
|
95 type = str,
|
|
|
96 required = False,
|
|
|
97 help = 'output type analysis')
|
|
|
98
|
|
|
99 parser.add_argument('-idop', '--output_path',
|
|
|
100 type = str,
|
|
|
101 default='flux_simulation',
|
|
|
102 help = 'output path for maps')
|
|
|
103
|
|
|
104 ARGS = parser.parse_args(args)
|
|
|
105 return ARGS
|
|
|
106
|
|
|
107 ########################### warning ###########################################
|
|
|
108 def warning(s :str) -> None:
|
|
|
109 """
|
|
|
110 Log a warning message to an output log file and print it to the console.
|
|
|
111
|
|
|
112 Args:
|
|
|
113 s (str): The warning message to be logged and printed.
|
|
|
114
|
|
|
115 Returns:
|
|
|
116 None
|
|
|
117 """
|
|
|
118 with open(ARGS.out_log, 'a') as log:
|
|
|
119 log.write(s + "\n\n")
|
|
|
120 print(s)
|
|
|
121
|
|
|
122
|
|
|
123 def write_to_file(dataset: pd.DataFrame, name: str, keep_index:bool=False)->None:
|
|
|
124 dataset.index.name = 'Reactions'
|
|
|
125 dataset.to_csv(ARGS.output_path + "/" + name + ".csv", sep = '\t', index = keep_index)
|
|
|
126
|
|
|
127 ############################ dataset input ####################################
|
|
|
128 def read_dataset(data :str, name :str) -> pd.DataFrame:
|
|
|
129 """
|
|
|
130 Read a dataset from a CSV file and return it as a pandas DataFrame.
|
|
|
131
|
|
|
132 Args:
|
|
|
133 data (str): Path to the CSV file containing the dataset.
|
|
|
134 name (str): Name of the dataset, used in error messages.
|
|
|
135
|
|
|
136 Returns:
|
|
|
137 pandas.DataFrame: DataFrame containing the dataset.
|
|
|
138
|
|
|
139 Raises:
|
|
|
140 pd.errors.EmptyDataError: If the CSV file is empty.
|
|
|
141 sys.exit: If the CSV file has the wrong format, the execution is aborted.
|
|
|
142 """
|
|
|
143 try:
|
|
|
144 dataset = pd.read_csv(data, sep = '\t', header = 0, index_col=0, engine='python')
|
|
|
145 except pd.errors.EmptyDataError:
|
|
|
146 sys.exit('Execution aborted: wrong format of ' + name + '\n')
|
|
|
147 if len(dataset.columns) < 2:
|
|
|
148 sys.exit('Execution aborted: wrong format of ' + name + '\n')
|
|
|
149 return dataset
|
|
|
150
|
|
|
151
|
|
|
152
|
|
|
153 def OPTGP_sampler(model:cobra.Model, model_name:str, n_samples:int=1000, thinning:int=100, n_batches:int=1, seed:int=0)-> None:
|
|
|
154 """
|
|
|
155 Samples from the OPTGP (Optimal Global Perturbation) algorithm and saves the results to CSV files.
|
|
|
156
|
|
|
157 Args:
|
|
|
158 model (cobra.Model): The COBRA model to sample from.
|
|
|
159 model_name (str): The name of the model, used in naming output files.
|
|
|
160 n_samples (int, optional): Number of samples per batch. Default is 1000.
|
|
|
161 thinning (int, optional): Thinning parameter for the sampler. Default is 100.
|
|
|
162 n_batches (int, optional): Number of batches to run. Default is 1.
|
|
|
163 seed (int, optional): Random seed for reproducibility. Default is 0.
|
|
|
164
|
|
|
165 Returns:
|
|
|
166 None
|
|
|
167 """
|
|
|
168
|
|
|
169 for i in range(0, n_batches):
|
|
|
170 optgp = OptGPSampler(model, thinning, seed)
|
|
|
171 samples = optgp.sample(n_samples)
|
|
|
172 samples.to_csv(ARGS.output_path + "/" + model_name + '_'+ str(i)+'_OPTGP.csv', index=False)
|
|
|
173 seed+=1
|
|
|
174 samplesTotal = pd.DataFrame()
|
|
|
175 for i in range(0, n_batches):
|
|
|
176 samples_batch = pd.read_csv(ARGS.output_path + "/" + model_name + '_'+ str(i)+'_OPTGP.csv')
|
|
|
177 samplesTotal = pd.concat([samplesTotal, samples_batch], ignore_index = True)
|
|
|
178
|
|
|
179 write_to_file(samplesTotal.T, model_name, True)
|
|
|
180
|
|
|
181 for i in range(0, n_batches):
|
|
|
182 os.remove(ARGS.output_path + "/" + model_name + '_'+ str(i)+'_OPTGP.csv')
|
|
|
183 pass
|
|
|
184
|
|
|
185
|
|
|
186 def CBS_sampler(model:cobra.Model, model_name:str, n_samples:int=1000, n_batches:int=1, seed:int=0)-> None:
|
|
|
187 """
|
|
|
188 Samples using the CBS (Constraint-based Sampling) algorithm and saves the results to CSV files.
|
|
|
189
|
|
|
190 Args:
|
|
|
191 model (cobra.Model): The COBRA model to sample from.
|
|
|
192 model_name (str): The name of the model, used in naming output files.
|
|
|
193 n_samples (int, optional): Number of samples per batch. Default is 1000.
|
|
|
194 n_batches (int, optional): Number of batches to run. Default is 1.
|
|
|
195 seed (int, optional): Random seed for reproducibility. Default is 0.
|
|
|
196
|
|
|
197 Returns:
|
|
|
198 None
|
|
|
199 """
|
|
|
200
|
|
|
201 df_FVA = cobra.flux_analysis.flux_variability_analysis(model,fraction_of_optimum=0).round(6)
|
|
|
202
|
|
|
203 df_coefficients = CBS_backend.randomObjectiveFunction(model, n_samples*n_batches, df_FVA, seed=seed)
|
|
|
204
|
|
|
205 for i in range(0, n_batches):
|
|
|
206 samples = pd.DataFrame(columns =[reaction.id for reaction in model.reactions], index = range(n_samples))
|
|
|
207 try:
|
|
|
208 CBS_backend.randomObjectiveFunctionSampling(model, n_samples, df_coefficients.iloc[:,i*n_samples:(i+1)*n_samples], samples)
|
|
|
209 except Exception as e:
|
|
|
210 utils.logWarning(
|
|
|
211 "Warning: GLPK solver has failed for " + model_name + ". Trying with COBRA interface. Error:" + str(e),
|
|
|
212 ARGS.out_log)
|
|
|
213 CBS_backend.randomObjectiveFunctionSampling_cobrapy(model, n_samples, df_coefficients.iloc[:,i*n_samples:(i+1)*n_samples],
|
|
|
214 samples)
|
|
|
215 utils.logWarning(ARGS.output_path + "/" + model_name + '_'+ str(i)+'_CBS.csv', ARGS.out_log)
|
|
|
216 samples.to_csv(ARGS.output_path + "/" + model_name + '_'+ str(i)+'_CBS.csv', index=False)
|
|
|
217
|
|
|
218 samplesTotal = pd.DataFrame()
|
|
|
219 for i in range(0, n_batches):
|
|
|
220 samples_batch = pd.read_csv(ARGS.output_path + "/" + model_name + '_'+ str(i)+'_CBS.csv')
|
|
|
221 samplesTotal = pd.concat([samplesTotal, samples_batch], ignore_index = True)
|
|
|
222
|
|
|
223 write_to_file(samplesTotal.T, model_name, True)
|
|
|
224
|
|
|
225 for i in range(0, n_batches):
|
|
|
226 os.remove(ARGS.output_path + "/" + model_name + '_'+ str(i)+'_CBS.csv')
|
|
|
227 pass
|
|
|
228
|
|
|
229
|
|
419
|
230
|
|
|
231 def model_sampler_with_bounds(model_input_original: cobra.Model, bounds_path: str, cell_name: str) -> List[pd.DataFrame]:
|
|
410
|
232 """
|
|
419
|
233 MODE 1: Prepares the model with bounds from separate bounds file and performs sampling.
|
|
410
|
234
|
|
|
235 Args:
|
|
|
236 model_input_original (cobra.Model): The original COBRA model.
|
|
|
237 bounds_path (str): Path to the CSV file containing the bounds dataset.
|
|
|
238 cell_name (str): Name of the cell, used to generate filenames for output.
|
|
|
239
|
|
|
240 Returns:
|
|
|
241 List[pd.DataFrame]: A list of DataFrames containing statistics and analysis results.
|
|
|
242 """
|
|
|
243
|
|
|
244 model_input = model_input_original.copy()
|
|
|
245 bounds_df = read_dataset(bounds_path, "bounds dataset")
|
|
419
|
246
|
|
|
247 # Apply bounds to model
|
|
410
|
248 for rxn_index, row in bounds_df.iterrows():
|
|
419
|
249 try:
|
|
|
250 model_input.reactions.get_by_id(rxn_index).lower_bound = row.lower_bound
|
|
|
251 model_input.reactions.get_by_id(rxn_index).upper_bound = row.upper_bound
|
|
|
252 except KeyError:
|
|
|
253 warning(f"Warning: Reaction {rxn_index} not found in model. Skipping.")
|
|
410
|
254
|
|
419
|
255 return perform_sampling_and_analysis(model_input, cell_name)
|
|
|
256
|
|
|
257
|
|
|
258 def perform_sampling_and_analysis(model_input: cobra.Model, cell_name: str) -> List[pd.DataFrame]:
|
|
|
259 """
|
|
|
260 Common function to perform sampling and analysis on a prepared model.
|
|
|
261
|
|
|
262 Args:
|
|
|
263 model_input (cobra.Model): The prepared COBRA model with bounds applied.
|
|
|
264 cell_name (str): Name of the cell, used to generate filenames for output.
|
|
|
265
|
|
|
266 Returns:
|
|
|
267 List[pd.DataFrame]: A list of DataFrames containing statistics and analysis results.
|
|
|
268 """
|
|
410
|
269
|
|
|
270 if ARGS.algorithm == 'OPTGP':
|
|
|
271 OPTGP_sampler(model_input, cell_name, ARGS.n_samples, ARGS.thinning, ARGS.n_batches, ARGS.seed)
|
|
|
272 elif ARGS.algorithm == 'CBS':
|
|
419
|
273 CBS_sampler(model_input, cell_name, ARGS.n_samples, ARGS.n_batches, ARGS.seed)
|
|
410
|
274
|
|
|
275 df_mean, df_median, df_quantiles = fluxes_statistics(cell_name, ARGS.output_types)
|
|
|
276
|
|
|
277 if("fluxes" not in ARGS.output_types):
|
|
419
|
278 os.remove(ARGS.output_path + "/" + cell_name + '.csv')
|
|
410
|
279
|
|
419
|
280 returnList = [df_mean, df_median, df_quantiles]
|
|
410
|
281
|
|
|
282 df_pFBA, df_FVA, df_sensitivity = fluxes_analysis(model_input, cell_name, ARGS.output_type_analysis)
|
|
|
283
|
|
|
284 if("pFBA" in ARGS.output_type_analysis):
|
|
|
285 returnList.append(df_pFBA)
|
|
|
286 if("FVA" in ARGS.output_type_analysis):
|
|
|
287 returnList.append(df_FVA)
|
|
|
288 if("sensitivity" in ARGS.output_type_analysis):
|
|
|
289 returnList.append(df_sensitivity)
|
|
|
290
|
|
|
291 return returnList
|
|
|
292
|
|
|
293 def fluxes_statistics(model_name: str, output_types:List)-> List[pd.DataFrame]:
|
|
|
294 """
|
|
|
295 Computes statistics (mean, median, quantiles) for the fluxes.
|
|
|
296
|
|
|
297 Args:
|
|
|
298 model_name (str): Name of the model, used in filename for input.
|
|
|
299 output_types (List[str]): Types of statistics to compute (mean, median, quantiles).
|
|
|
300
|
|
|
301 Returns:
|
|
|
302 List[pd.DataFrame]: List of DataFrames containing mean, median, and quantiles statistics.
|
|
|
303 """
|
|
|
304
|
|
|
305 df_mean = pd.DataFrame()
|
|
|
306 df_median= pd.DataFrame()
|
|
|
307 df_quantiles= pd.DataFrame()
|
|
|
308
|
|
|
309 df_samples = pd.read_csv(ARGS.output_path + "/" + model_name + '.csv', sep = '\t', index_col = 0).T
|
|
|
310 df_samples = df_samples.round(8)
|
|
|
311
|
|
|
312 for output_type in output_types:
|
|
|
313 if(output_type == "mean"):
|
|
|
314 df_mean = df_samples.mean()
|
|
|
315 df_mean = df_mean.to_frame().T
|
|
|
316 df_mean = df_mean.reset_index(drop=True)
|
|
|
317 df_mean.index = [model_name]
|
|
|
318 elif(output_type == "median"):
|
|
|
319 df_median = df_samples.median()
|
|
|
320 df_median = df_median.to_frame().T
|
|
|
321 df_median = df_median.reset_index(drop=True)
|
|
|
322 df_median.index = [model_name]
|
|
|
323 elif(output_type == "quantiles"):
|
|
|
324 newRow = []
|
|
|
325 cols = []
|
|
|
326 for rxn in df_samples.columns:
|
|
|
327 quantiles = df_samples[rxn].quantile([0.25, 0.50, 0.75])
|
|
|
328 newRow.append(quantiles[0.25])
|
|
|
329 cols.append(rxn + "_q1")
|
|
|
330 newRow.append(quantiles[0.5])
|
|
|
331 cols.append(rxn + "_q2")
|
|
|
332 newRow.append(quantiles[0.75])
|
|
|
333 cols.append(rxn + "_q3")
|
|
|
334 df_quantiles = pd.DataFrame(columns=cols)
|
|
|
335 df_quantiles.loc[0] = newRow
|
|
|
336 df_quantiles = df_quantiles.reset_index(drop=True)
|
|
|
337 df_quantiles.index = [model_name]
|
|
|
338
|
|
|
339 return df_mean, df_median, df_quantiles
|
|
|
340
|
|
|
341 def fluxes_analysis(model:cobra.Model, model_name:str, output_types:List)-> List[pd.DataFrame]:
|
|
|
342 """
|
|
|
343 Performs flux analysis including pFBA, FVA, and sensitivity analysis.
|
|
|
344
|
|
|
345 Args:
|
|
|
346 model (cobra.Model): The COBRA model to analyze.
|
|
|
347 model_name (str): Name of the model, used in filenames for output.
|
|
|
348 output_types (List[str]): Types of analysis to perform (pFBA, FVA, sensitivity).
|
|
|
349
|
|
|
350 Returns:
|
|
|
351 List[pd.DataFrame]: List of DataFrames containing pFBA, FVA, and sensitivity analysis results.
|
|
|
352 """
|
|
|
353
|
|
|
354 df_pFBA = pd.DataFrame()
|
|
|
355 df_FVA= pd.DataFrame()
|
|
|
356 df_sensitivity= pd.DataFrame()
|
|
|
357
|
|
|
358 for output_type in output_types:
|
|
|
359 if(output_type == "pFBA"):
|
|
|
360 model.objective = "Biomass"
|
|
|
361 solution = cobra.flux_analysis.pfba(model)
|
|
|
362 fluxes = solution.fluxes
|
|
419
|
363 df_pFBA.loc[0,[rxn.id for rxn in model.reactions]] = fluxes.tolist()
|
|
410
|
364 df_pFBA = df_pFBA.reset_index(drop=True)
|
|
|
365 df_pFBA.index = [model_name]
|
|
|
366 df_pFBA = df_pFBA.astype(float).round(6)
|
|
|
367 elif(output_type == "FVA"):
|
|
430
|
368 fva = cobra.flux_analysis.flux_variability_analysis(model, fraction_of_optimum=ARGS.perc_opt, processes=1).round(8)
|
|
410
|
369 columns = []
|
|
|
370 for rxn in fva.index.to_list():
|
|
|
371 columns.append(rxn + "_min")
|
|
|
372 columns.append(rxn + "_max")
|
|
|
373 df_FVA= pd.DataFrame(columns = columns)
|
|
|
374 for index_rxn, row in fva.iterrows():
|
|
|
375 df_FVA.loc[0, index_rxn+ "_min"] = fva.loc[index_rxn, "minimum"]
|
|
|
376 df_FVA.loc[0, index_rxn+ "_max"] = fva.loc[index_rxn, "maximum"]
|
|
|
377 df_FVA = df_FVA.reset_index(drop=True)
|
|
|
378 df_FVA.index = [model_name]
|
|
|
379 df_FVA = df_FVA.astype(float).round(6)
|
|
|
380 elif(output_type == "sensitivity"):
|
|
|
381 model.objective = "Biomass"
|
|
|
382 solution_original = model.optimize().objective_value
|
|
|
383 reactions = model.reactions
|
|
|
384 single = cobra.flux_analysis.single_reaction_deletion(model)
|
|
|
385 newRow = []
|
|
|
386 df_sensitivity = pd.DataFrame(columns = [rxn.id for rxn in reactions], index = [model_name])
|
|
|
387 for rxn in reactions:
|
|
|
388 newRow.append(single.knockout[rxn.id].growth.values[0]/solution_original)
|
|
|
389 df_sensitivity.loc[model_name] = newRow
|
|
|
390 df_sensitivity = df_sensitivity.astype(float).round(6)
|
|
|
391 return df_pFBA, df_FVA, df_sensitivity
|
|
|
392
|
|
|
393 ############################# main ###########################################
|
|
|
394 def main(args :List[str] = None) -> None:
|
|
|
395 """
|
|
|
396 Initializes everything and sets the program in motion based on the fronted input arguments.
|
|
|
397
|
|
|
398 Returns:
|
|
|
399 None
|
|
|
400 """
|
|
|
401
|
|
419
|
402 num_processors = max(1, cpu_count() - 1)
|
|
410
|
403
|
|
|
404 global ARGS
|
|
|
405 ARGS = process_args(args)
|
|
|
406
|
|
|
407 if not os.path.exists(ARGS.output_path):
|
|
|
408 os.makedirs(ARGS.output_path)
|
|
419
|
409
|
|
|
410 #ARGS.bounds = ARGS.input.split(",")
|
|
|
411 #ARGS.bounds_name = ARGS.name.split(",")
|
|
|
412 #ARGS.output_types = ARGS.output_type.split(",")
|
|
|
413 #ARGS.output_type_analysis = ARGS.output_type_analysis.split(",")
|
|
|
414
|
|
|
415 # --- Normalize inputs (the tool may pass comma-separated --input and either --name or --names) ---
|
|
421
|
416 ARGS.input_files = ARGS.input.split(",") if ARGS.input else []
|
|
419
|
417 ARGS.file_names = ARGS.name.split(",")
|
|
|
418 # output types (required) -> list
|
|
421
|
419 ARGS.output_types = ARGS.output_type.split(",") if ARGS.output_type else []
|
|
419
|
420 # optional analysis output types -> list or empty
|
|
421
|
421 ARGS.output_type_analysis = ARGS.output_type_analysis.split(",") if ARGS.output_type_analysis else []
|
|
419
|
422
|
|
421
|
423 print("=== INPUT FILES ===")
|
|
422
|
424 print(f"{ARGS.input_files}")
|
|
|
425 print(f"{ARGS.file_names}")
|
|
|
426 print(f"{ARGS.output_type}")
|
|
|
427 print(f"{ARGS.output_types}")
|
|
|
428 print(f"{ARGS.output_type_analysis}")
|
|
410
|
429
|
|
419
|
430 if ARGS.model_and_bounds == "True":
|
|
|
431 # MODE 1: Model + bounds (separate files)
|
|
|
432 print("=== MODE 1: Model + Bounds (separate files) ===")
|
|
|
433
|
|
|
434 # Load base model
|
|
|
435 if not ARGS.model_upload:
|
|
|
436 sys.exit("Error: model_upload is required for Mode 1")
|
|
410
|
437
|
|
419
|
438 base_model = model_utils.build_cobra_model_from_csv(ARGS.model_upload)
|
|
410
|
439
|
|
419
|
440 validation = model_utils.validate_model(base_model)
|
|
|
441
|
|
|
442 print("\n=== VALIDAZIONE MODELLO ===")
|
|
|
443 for key, value in validation.items():
|
|
|
444 print(f"{key}: {value}")
|
|
|
445
|
|
|
446 #Set solver verbosity to 1 to see warning and error messages only.
|
|
|
447 base_model.solver.configuration.verbosity = 1
|
|
410
|
448
|
|
419
|
449 # Process each bounds file with the base model
|
|
|
450 results = Parallel(n_jobs=num_processors)(
|
|
|
451 delayed(model_sampler_with_bounds)(base_model, bounds_file, cell_name)
|
|
|
452 for bounds_file, cell_name in zip(ARGS.input_files, ARGS.file_names)
|
|
|
453 )
|
|
410
|
454
|
|
419
|
455 else:
|
|
|
456 # MODE 2: Multiple complete models
|
|
|
457 print("=== MODE 2: Multiple complete models ===")
|
|
|
458
|
|
|
459 # Process each complete model file
|
|
|
460 results = Parallel(n_jobs=num_processors)(
|
|
|
461 delayed(perform_sampling_and_analysis)(model_utils.build_cobra_model_from_csv(model_file), cell_name)
|
|
|
462 for model_file, cell_name in zip(ARGS.input_files, ARGS.file_names)
|
|
|
463 )
|
|
410
|
464
|
|
|
465
|
|
|
466 all_mean = pd.concat([result[0] for result in results], ignore_index=False)
|
|
|
467 all_median = pd.concat([result[1] for result in results], ignore_index=False)
|
|
|
468 all_quantiles = pd.concat([result[2] for result in results], ignore_index=False)
|
|
|
469
|
|
|
470 if("mean" in ARGS.output_types):
|
|
|
471 all_mean = all_mean.fillna(0.0)
|
|
|
472 all_mean = all_mean.sort_index()
|
|
|
473 write_to_file(all_mean.T, "mean", True)
|
|
|
474
|
|
|
475 if("median" in ARGS.output_types):
|
|
|
476 all_median = all_median.fillna(0.0)
|
|
|
477 all_median = all_median.sort_index()
|
|
|
478 write_to_file(all_median.T, "median", True)
|
|
|
479
|
|
|
480 if("quantiles" in ARGS.output_types):
|
|
|
481 all_quantiles = all_quantiles.fillna(0.0)
|
|
|
482 all_quantiles = all_quantiles.sort_index()
|
|
|
483 write_to_file(all_quantiles.T, "quantiles", True)
|
|
|
484
|
|
|
485 index_result = 3
|
|
|
486 if("pFBA" in ARGS.output_type_analysis):
|
|
|
487 all_pFBA = pd.concat([result[index_result] for result in results], ignore_index=False)
|
|
|
488 all_pFBA = all_pFBA.sort_index()
|
|
|
489 write_to_file(all_pFBA.T, "pFBA", True)
|
|
|
490 index_result+=1
|
|
|
491 if("FVA" in ARGS.output_type_analysis):
|
|
|
492 all_FVA= pd.concat([result[index_result] for result in results], ignore_index=False)
|
|
|
493 all_FVA = all_FVA.sort_index()
|
|
|
494 write_to_file(all_FVA.T, "FVA", True)
|
|
|
495 index_result+=1
|
|
|
496 if("sensitivity" in ARGS.output_type_analysis):
|
|
|
497 all_sensitivity = pd.concat([result[index_result] for result in results], ignore_index=False)
|
|
|
498 all_sensitivity = all_sensitivity.sort_index()
|
|
|
499 write_to_file(all_sensitivity.T, "sensitivity", True)
|
|
|
500
|
|
|
501 pass
|
|
|
502
|
|
|
503 ##############################################################################
|
|
|
504 if __name__ == "__main__":
|
|
|
505 main() |
