|
93
|
1 import os
|
|
|
2 import csv
|
|
|
3 import cobra
|
|
|
4 import pickle
|
|
|
5 import argparse
|
|
|
6 import pandas as pd
|
|
|
7 import utils.general_utils as utils
|
|
|
8 import utils.rule_parsing as rulesUtils
|
|
147
|
9 from typing import Optional, Tuple, Union, List, Dict
|
|
93
|
10 import utils.reaction_parsing as reactionUtils
|
|
370
|
11 import openpyxl
|
|
93
|
12
|
|
|
13 ARGS : argparse.Namespace
|
|
343
|
14 def process_args(args: List[str] = None) -> argparse.Namespace:
|
|
|
15 """
|
|
|
16 Parse command-line arguments for CustomDataGenerator.
|
|
93
|
17 """
|
|
343
|
18
|
|
|
19 parser = argparse.ArgumentParser(
|
|
|
20 usage="%(prog)s [options]",
|
|
|
21 description="Generate custom data from a given model"
|
|
|
22 )
|
|
93
|
23
|
|
343
|
24 parser.add_argument("--out_log", type=str, required=True,
|
|
|
25 help="Output log file")
|
|
93
|
26
|
|
343
|
27 parser.add_argument("--model", type=str,
|
|
|
28 help="Built-in model identifier (e.g., ENGRO2, Recon, HMRcore)")
|
|
|
29 parser.add_argument("--input", type=str,
|
|
|
30 help="Custom model file (JSON or XML)")
|
|
|
31 parser.add_argument("--name", type=str, required=True,
|
|
|
32 help="Model name (default or custom)")
|
|
93
|
33
|
|
343
|
34 parser.add_argument("--medium_selector", type=str, required=True,
|
|
|
35 help="Medium selection option (default/custom)")
|
|
|
36 parser.add_argument("--medium", type=str,
|
|
|
37 help="Custom medium file if medium_selector=Custom")
|
|
|
38
|
|
|
39 parser.add_argument("--output_format", type=str, choices=["tabular", "xlsx"], required=True,
|
|
|
40 help="Output format: CSV (tabular) or Excel (xlsx)")
|
|
|
41
|
|
375
|
42 parser.add_argument("--out_tabular", type=str,
|
|
|
43 help="Output file for the merged dataset (CSV or XLSX)")
|
|
|
44
|
|
|
45 parser.add_argument("--out_xlsx", type=str,
|
|
365
|
46 help="Output file for the merged dataset (CSV or XLSX)")
|
|
343
|
47
|
|
353
|
48 parser.add_argument("--tool_dir", type=str, default=os.path.dirname(__file__),
|
|
363
|
49 help="Tool directory (passed from Galaxy as $__tool_directory__)")
|
|
353
|
50
|
|
93
|
51
|
|
343
|
52 return parser.parse_args(args)
|
|
93
|
53
|
|
|
54 ################################- INPUT DATA LOADING -################################
|
|
|
55 def load_custom_model(file_path :utils.FilePath, ext :Optional[utils.FileFormat] = None) -> cobra.Model:
|
|
|
56 """
|
|
|
57 Loads a custom model from a file, either in JSON or XML format.
|
|
|
58
|
|
|
59 Args:
|
|
|
60 file_path : The path to the file containing the custom model.
|
|
|
61 ext : explicit file extension. Necessary for standard use in galaxy because of its weird behaviour.
|
|
|
62
|
|
|
63 Raises:
|
|
|
64 DataErr : if the file is in an invalid format or cannot be opened for whatever reason.
|
|
|
65
|
|
|
66 Returns:
|
|
|
67 cobra.Model : the model, if successfully opened.
|
|
|
68 """
|
|
|
69 ext = ext if ext else file_path.ext
|
|
|
70 try:
|
|
|
71 if ext is utils.FileFormat.XML:
|
|
|
72 return cobra.io.read_sbml_model(file_path.show())
|
|
|
73
|
|
|
74 if ext is utils.FileFormat.JSON:
|
|
|
75 return cobra.io.load_json_model(file_path.show())
|
|
|
76
|
|
|
77 except Exception as e: raise utils.DataErr(file_path, e.__str__())
|
|
|
78 raise utils.DataErr(file_path,
|
|
|
79 f"Formato \"{file_path.ext}\" non riconosciuto, sono supportati solo file JSON e XML")
|
|
|
80
|
|
|
81 ################################- DATA GENERATION -################################
|
|
|
82 ReactionId = str
|
|
|
83 def generate_rules(model: cobra.Model, *, asParsed = True) -> Union[Dict[ReactionId, rulesUtils.OpList], Dict[ReactionId, str]]:
|
|
|
84 """
|
|
|
85 Generates a dictionary mapping reaction ids to rules from the model.
|
|
|
86
|
|
|
87 Args:
|
|
|
88 model : the model to derive data from.
|
|
|
89 asParsed : if True parses the rules to an optimized runtime format, otherwise leaves them as strings.
|
|
|
90
|
|
|
91 Returns:
|
|
|
92 Dict[ReactionId, rulesUtils.OpList] : the generated dictionary of parsed rules.
|
|
|
93 Dict[ReactionId, str] : the generated dictionary of raw rules.
|
|
|
94 """
|
|
|
95 # Is the below approach convoluted? yes
|
|
|
96 # Ok but is it inefficient? probably
|
|
|
97 # Ok but at least I don't have to repeat the check at every rule (I'm clinically insane)
|
|
|
98 _ruleGetter = lambda reaction : reaction.gene_reaction_rule
|
|
|
99 ruleExtractor = (lambda reaction :
|
|
|
100 rulesUtils.parseRuleToNestedList(_ruleGetter(reaction))) if asParsed else _ruleGetter
|
|
|
101
|
|
|
102 return {
|
|
|
103 reaction.id : ruleExtractor(reaction)
|
|
|
104 for reaction in model.reactions
|
|
|
105 if reaction.gene_reaction_rule }
|
|
|
106
|
|
|
107 def generate_reactions(model :cobra.Model, *, asParsed = True) -> Dict[ReactionId, str]:
|
|
|
108 """
|
|
|
109 Generates a dictionary mapping reaction ids to reaction formulas from the model.
|
|
|
110
|
|
|
111 Args:
|
|
|
112 model : the model to derive data from.
|
|
|
113 asParsed : if True parses the reactions to an optimized runtime format, otherwise leaves them as they are.
|
|
|
114
|
|
|
115 Returns:
|
|
|
116 Dict[ReactionId, str] : the generated dictionary.
|
|
|
117 """
|
|
|
118
|
|
|
119 unparsedReactions = {
|
|
|
120 reaction.id : reaction.reaction
|
|
|
121 for reaction in model.reactions
|
|
|
122 if reaction.reaction
|
|
|
123 }
|
|
|
124
|
|
|
125 if not asParsed: return unparsedReactions
|
|
|
126
|
|
|
127 return reactionUtils.create_reaction_dict(unparsedReactions)
|
|
|
128
|
|
|
129 def get_medium(model:cobra.Model) -> pd.DataFrame:
|
|
|
130 trueMedium=[]
|
|
|
131 for r in model.reactions:
|
|
|
132 positiveCoeff=0
|
|
|
133 for m in r.metabolites:
|
|
|
134 if r.get_coefficient(m.id)>0:
|
|
|
135 positiveCoeff=1;
|
|
|
136 if (positiveCoeff==0 and r.lower_bound<0):
|
|
|
137 trueMedium.append(r.id)
|
|
|
138
|
|
|
139 df_medium = pd.DataFrame()
|
|
|
140 df_medium["reaction"] = trueMedium
|
|
|
141 return df_medium
|
|
|
142
|
|
|
143 def generate_bounds(model:cobra.Model) -> pd.DataFrame:
|
|
|
144
|
|
|
145 rxns = []
|
|
|
146 for reaction in model.reactions:
|
|
|
147 rxns.append(reaction.id)
|
|
|
148
|
|
|
149 bounds = pd.DataFrame(columns = ["lower_bound", "upper_bound"], index=rxns)
|
|
|
150
|
|
|
151 for reaction in model.reactions:
|
|
|
152 bounds.loc[reaction.id] = [reaction.lower_bound, reaction.upper_bound]
|
|
|
153 return bounds
|
|
|
154
|
|
|
155
|
|
|
156 ###############################- FILE SAVING -################################
|
|
|
157 def save_as_csv_filePath(data :dict, file_path :utils.FilePath, fieldNames :Tuple[str, str]) -> None:
|
|
|
158 """
|
|
|
159 Saves any dictionary-shaped data in a .csv file created at the given file_path as FilePath.
|
|
|
160
|
|
|
161 Args:
|
|
|
162 data : the data to be written to the file.
|
|
|
163 file_path : the path to the .csv file.
|
|
|
164 fieldNames : the names of the fields (columns) in the .csv file.
|
|
|
165
|
|
|
166 Returns:
|
|
|
167 None
|
|
|
168 """
|
|
|
169 with open(file_path.show(), 'w', newline='') as csvfile:
|
|
|
170 writer = csv.DictWriter(csvfile, fieldnames = fieldNames, dialect="excel-tab")
|
|
|
171 writer.writeheader()
|
|
|
172
|
|
|
173 for key, value in data.items():
|
|
|
174 writer.writerow({ fieldNames[0] : key, fieldNames[1] : value })
|
|
|
175
|
|
|
176 def save_as_csv(data :dict, file_path :str, fieldNames :Tuple[str, str]) -> None:
|
|
|
177 """
|
|
|
178 Saves any dictionary-shaped data in a .csv file created at the given file_path as string.
|
|
|
179
|
|
|
180 Args:
|
|
|
181 data : the data to be written to the file.
|
|
|
182 file_path : the path to the .csv file.
|
|
|
183 fieldNames : the names of the fields (columns) in the .csv file.
|
|
|
184
|
|
|
185 Returns:
|
|
|
186 None
|
|
|
187 """
|
|
|
188 with open(file_path, 'w', newline='') as csvfile:
|
|
|
189 writer = csv.DictWriter(csvfile, fieldnames = fieldNames, dialect="excel-tab")
|
|
|
190 writer.writeheader()
|
|
|
191
|
|
|
192 for key, value in data.items():
|
|
|
193 writer.writerow({ fieldNames[0] : key, fieldNames[1] : value })
|
|
|
194
|
|
377
|
195 def save_as_tabular_df(df: pd.DataFrame, path: str) -> None:
|
|
|
196 try:
|
|
|
197 os.makedirs(os.path.dirname(path) or ".", exist_ok=True)
|
|
|
198 df.to_csv(path, sep="\t", index=False)
|
|
|
199 except Exception as e:
|
|
|
200 raise utils.DataErr(path, f"failed writing tabular output: {e}")
|
|
|
201
|
|
|
202 def save_as_xlsx_df(df: pd.DataFrame, path: str) -> None:
|
|
|
203 try:
|
|
|
204 if not path.lower().endswith(".xlsx"):
|
|
|
205 path += ".xlsx"
|
|
|
206 os.makedirs(os.path.dirname(path) or ".", exist_ok=True)
|
|
|
207 df.to_excel(path, index=False)
|
|
|
208 except Exception as e:
|
|
|
209 raise utils.DataErr(path, f"failed writing xlsx output: {e}")
|
|
|
210
|
|
93
|
211 ###############################- ENTRY POINT -################################
|
|
147
|
212 def main(args:List[str] = None) -> None:
|
|
93
|
213 """
|
|
|
214 Initializes everything and sets the program in motion based on the fronted input arguments.
|
|
|
215
|
|
|
216 Returns:
|
|
|
217 None
|
|
|
218 """
|
|
|
219 # get args from frontend (related xml)
|
|
|
220 global ARGS
|
|
147
|
221 ARGS = process_args(args)
|
|
93
|
222
|
|
343
|
223
|
|
350
|
224 if ARGS.input:
|
|
343
|
225 # load custom model
|
|
|
226 model = load_custom_model(
|
|
|
227 utils.FilePath.fromStrPath(ARGS.input), utils.FilePath.fromStrPath(ARGS.name).ext)
|
|
|
228 else:
|
|
|
229 # load built-in model
|
|
93
|
230
|
|
343
|
231 try:
|
|
|
232 model_enum = utils.Model[ARGS.model] # e.g., Model['ENGRO2']
|
|
|
233 except KeyError:
|
|
|
234 raise utils.ArgsErr("model", "one of Recon/ENGRO2/HMRcore/Custom_model", ARGS.model)
|
|
|
235
|
|
|
236 # Load built-in model (Model.getCOBRAmodel uses tool_dir to locate local models)
|
|
|
237 try:
|
|
353
|
238 model = model_enum.getCOBRAmodel(toolDir=ARGS.tool_dir)
|
|
343
|
239 except Exception as e:
|
|
|
240 # Wrap/normalize load errors as DataErr for consistency
|
|
|
241 raise utils.DataErr(ARGS.model, f"failed loading built-in model: {e}")
|
|
|
242
|
|
|
243 # Determine final model name: explicit --name overrides, otherwise use the model id
|
|
|
244 model_name = ARGS.name if ARGS.name else ARGS.model
|
|
93
|
245
|
|
|
246 # generate data
|
|
|
247 rules = generate_rules(model, asParsed = False)
|
|
|
248 reactions = generate_reactions(model, asParsed = False)
|
|
|
249 bounds = generate_bounds(model)
|
|
|
250 medium = get_medium(model)
|
|
|
251
|
|
343
|
252 df_rules = pd.DataFrame(list(rules.items()), columns = ["ReactionID", "Rule"])
|
|
|
253 df_reactions = pd.DataFrame(list(reactions.items()), columns = ["ReactionID", "Reaction"])
|
|
|
254
|
|
|
255 df_bounds = bounds.reset_index().rename(columns = {"index": "ReactionID"})
|
|
|
256 df_medium = medium.rename(columns = {"reaction": "ReactionID"})
|
|
|
257 df_medium["InMedium"] = True # flag per indicare la presenza nel medium
|
|
|
258
|
|
|
259 merged = df_reactions.merge(df_rules, on = "ReactionID", how = "outer")
|
|
|
260 merged = merged.merge(df_bounds, on = "ReactionID", how = "outer")
|
|
|
261
|
|
|
262 merged = merged.merge(df_medium, on = "ReactionID", how = "left")
|
|
|
263
|
|
|
264 merged["InMedium"] = merged["InMedium"].fillna(False)
|
|
|
265
|
|
|
266 merged = merged.sort_values(by = "InMedium", ascending = False)
|
|
|
267
|
|
359
|
268 #out_file = os.path.join(ARGS.output_path, f"{os.path.basename(ARGS.name).split('.')[0]}_custom_data")
|
|
343
|
269
|
|
|
270 #merged.to_csv(out_file, sep = '\t', index = False)
|
|
|
271
|
|
|
272
|
|
|
273 ####
|
|
|
274
|
|
377
|
275 # write only the requested output
|
|
343
|
276 if ARGS.output_format == "xlsx":
|
|
375
|
277 if not ARGS.out_xlsx:
|
|
|
278 raise utils.ArgsErr("out_xlsx", "output path (--out_xlsx) is required when output_format == xlsx", ARGS.out_xlsx)
|
|
377
|
279 save_as_xlsx_df(merged, ARGS.out_xlsx)
|
|
|
280 expected = ARGS.out_xlsx
|
|
343
|
281 else:
|
|
375
|
282 if not ARGS.out_tabular:
|
|
|
283 raise utils.ArgsErr("out_tabular", "output path (--out_tabular) is required when output_format == tabular", ARGS.out_tabular)
|
|
377
|
284 save_as_tabular_df(merged, ARGS.out_tabular)
|
|
|
285 expected = ARGS.out_tabular
|
|
|
286
|
|
|
287 # verify output exists and non-empty
|
|
|
288 if not expected or not os.path.exists(expected) or os.path.getsize(expected) == 0:
|
|
|
289 raise utils.DataErr(expected, "Output non creato o vuoto")
|
|
343
|
290
|
|
367
|
291 print("CustomDataGenerator: completed successfully")
|
|
93
|
292
|
|
|
293 if __name__ == '__main__':
|
|
|
294 main() |
