|
491
|
1 """
|
|
|
2 Scripts to generate a tabular file of a metabolic model (built-in or custom).
|
|
|
3
|
|
|
4 This script loads a COBRA model (built-in or custom), optionally applies
|
|
|
5 medium and gene nomenclature settings, derives reaction-related metadata
|
|
|
6 (GPR rules, formulas, bounds, objective coefficients, medium membership,
|
|
|
7 and compartments for ENGRO2), and writes a tabular summary.
|
|
|
8 """
|
|
|
9
|
|
|
10 import os
|
|
|
11 import csv
|
|
|
12 import cobra
|
|
|
13 import argparse
|
|
|
14 import pandas as pd
|
|
|
15 import utils.general_utils as utils
|
|
|
16 from typing import Optional, Tuple, List
|
|
|
17 import utils.model_utils as modelUtils
|
|
|
18 import logging
|
|
|
19 from pathlib import Path
|
|
|
20
|
|
|
21
|
|
|
22 ARGS : argparse.Namespace
|
|
|
23 def process_args(args: List[str] = None) -> argparse.Namespace:
|
|
|
24 """
|
|
|
25 Parse command-line arguments for metabolic_model_setting.
|
|
|
26 """
|
|
|
27
|
|
|
28 parser = argparse.ArgumentParser(
|
|
|
29 usage="%(prog)s [options]",
|
|
|
30 description="Generate custom data from a given model"
|
|
|
31 )
|
|
|
32
|
|
|
33 parser.add_argument("--out_log", type=str, required=True,
|
|
|
34 help="Output log file")
|
|
|
35
|
|
|
36 parser.add_argument("--model", type=str,
|
|
|
37 help="Built-in model identifier (e.g., ENGRO2, Recon, HMRcore)")
|
|
|
38 parser.add_argument("--input", type=str,
|
|
498
|
39 help="Custom model file (JSON, XML, MAT, YAML)")
|
|
496
|
40 parser.add_argument("--name", nargs='*', required=True,
|
|
491
|
41 help="Model name (default or custom)")
|
|
|
42
|
|
|
43 parser.add_argument("--medium_selector", type=str, required=True,
|
|
|
44 help="Medium selection option")
|
|
|
45
|
|
|
46 parser.add_argument("--gene_format", type=str, default="Default",
|
|
|
47 help="Gene nomenclature format: Default (original), ENSNG, HGNC_SYMBOL, HGNC_ID, ENTREZ")
|
|
|
48
|
|
|
49 parser.add_argument("--out_tabular", type=str,
|
|
|
50 help="Output file for the merged dataset (CSV or XLSX)")
|
|
|
51
|
|
|
52 parser.add_argument("--tool_dir", type=str, default=os.path.dirname(__file__),
|
|
|
53 help="Tool directory (passed from Galaxy as $__tool_directory__)")
|
|
|
54
|
|
|
55
|
|
|
56 return parser.parse_args(args)
|
|
|
57
|
|
|
58 ################################- INPUT DATA LOADING -################################
|
|
498
|
59 def detect_file_format(file_path: str) -> utils.FileFormat:
|
|
|
60 """
|
|
|
61 Detect file format by examining file content and extension.
|
|
|
62 Handles Galaxy .dat files by looking at content.
|
|
|
63 """
|
|
|
64 try:
|
|
|
65 with open(file_path, 'r') as f:
|
|
|
66 first_lines = ''.join([f.readline() for _ in range(5)])
|
|
|
67
|
|
|
68 # Check for XML (SBML)
|
|
|
69 if '<?xml' in first_lines or '<sbml' in first_lines:
|
|
|
70 return utils.FileFormat.XML
|
|
|
71
|
|
|
72 # Check for JSON
|
|
|
73 if first_lines.strip().startswith('{'):
|
|
|
74 return utils.FileFormat.JSON
|
|
|
75
|
|
|
76 # Check for YAML
|
|
|
77 if any(line.strip().endswith(':') for line in first_lines.split('\n')[:3]):
|
|
|
78 return utils.FileFormat.YML
|
|
|
79
|
|
|
80 except:
|
|
|
81 pass
|
|
|
82
|
|
|
83 # Fall back to extension-based detection
|
|
|
84 if file_path.endswith('.xml') or file_path.endswith('.sbml'):
|
|
|
85 return utils.FileFormat.XML
|
|
|
86 elif file_path.endswith('.json'):
|
|
|
87 return utils.FileFormat.JSON
|
|
|
88 elif file_path.endswith('.mat'):
|
|
|
89 return utils.FileFormat.MAT
|
|
|
90 elif file_path.endswith('.yml') or file_path.endswith('.yaml'):
|
|
|
91 return utils.FileFormat.YML
|
|
|
92
|
|
|
93 # Default to XML for unknown extensions
|
|
|
94 return utils.FileFormat.XML
|
|
|
95
|
|
491
|
96 def load_custom_model(file_path :utils.FilePath, ext :Optional[utils.FileFormat] = None) -> cobra.Model:
|
|
|
97 """
|
|
|
98 Loads a custom model from a file, either in JSON, XML, MAT, or YML format.
|
|
|
99
|
|
|
100 Args:
|
|
|
101 file_path : The path to the file containing the custom model.
|
|
|
102 ext : explicit file extension. Necessary for standard use in galaxy because of its weird behaviour.
|
|
|
103
|
|
|
104 Raises:
|
|
|
105 DataErr : if the file is in an invalid format or cannot be opened for whatever reason.
|
|
|
106
|
|
|
107 Returns:
|
|
|
108 cobra.Model : the model, if successfully opened.
|
|
|
109 """
|
|
|
110 ext = ext if ext else file_path.ext
|
|
|
111 try:
|
|
|
112 if ext is utils.FileFormat.XML:
|
|
|
113 return cobra.io.read_sbml_model(file_path.show())
|
|
|
114
|
|
|
115 if ext is utils.FileFormat.JSON:
|
|
|
116 return cobra.io.load_json_model(file_path.show())
|
|
|
117
|
|
|
118 if ext is utils.FileFormat.MAT:
|
|
|
119 return cobra.io.load_matlab_model(file_path.show())
|
|
|
120
|
|
|
121 if ext is utils.FileFormat.YML:
|
|
|
122 return cobra.io.load_yaml_model(file_path.show())
|
|
|
123
|
|
|
124 except Exception as e: raise utils.DataErr(file_path, e.__str__())
|
|
|
125 raise utils.DataErr(
|
|
|
126 file_path,
|
|
|
127 f"Unrecognized format '{file_path.ext}'. Only JSON, XML, MAT, YML are supported."
|
|
|
128 )
|
|
|
129
|
|
|
130
|
|
|
131 ###############################- FILE SAVING -################################
|
|
|
132 def save_as_csv_filePath(data :dict, file_path :utils.FilePath, fieldNames :Tuple[str, str]) -> None:
|
|
|
133 """
|
|
|
134 Saves any dictionary-shaped data in a .csv file created at the given file_path as FilePath.
|
|
|
135
|
|
|
136 Args:
|
|
|
137 data : the data to be written to the file.
|
|
|
138 file_path : the path to the .csv file.
|
|
|
139 fieldNames : the names of the fields (columns) in the .csv file.
|
|
|
140
|
|
|
141 Returns:
|
|
|
142 None
|
|
|
143 """
|
|
|
144 with open(file_path.show(), 'w', newline='') as csvfile:
|
|
|
145 writer = csv.DictWriter(csvfile, fieldnames = fieldNames, dialect="excel-tab")
|
|
|
146 writer.writeheader()
|
|
|
147
|
|
|
148 for key, value in data.items():
|
|
|
149 writer.writerow({ fieldNames[0] : key, fieldNames[1] : value })
|
|
|
150
|
|
|
151 def save_as_csv(data :dict, file_path :str, fieldNames :Tuple[str, str]) -> None:
|
|
|
152 """
|
|
|
153 Saves any dictionary-shaped data in a .csv file created at the given file_path as string.
|
|
|
154
|
|
|
155 Args:
|
|
|
156 data : the data to be written to the file.
|
|
|
157 file_path : the path to the .csv file.
|
|
|
158 fieldNames : the names of the fields (columns) in the .csv file.
|
|
|
159
|
|
|
160 Returns:
|
|
|
161 None
|
|
|
162 """
|
|
|
163 with open(file_path, 'w', newline='') as csvfile:
|
|
|
164 writer = csv.DictWriter(csvfile, fieldnames = fieldNames, dialect="excel-tab")
|
|
|
165 writer.writeheader()
|
|
|
166
|
|
|
167 for key, value in data.items():
|
|
|
168 writer.writerow({ fieldNames[0] : key, fieldNames[1] : value })
|
|
|
169
|
|
|
170 def save_as_tabular_df(df: pd.DataFrame, path: str) -> None:
|
|
|
171 """
|
|
|
172 Save a pandas DataFrame as a tab-separated file, creating directories as needed.
|
|
|
173
|
|
|
174 Args:
|
|
|
175 df: The DataFrame to write.
|
|
|
176 path: Destination file path (will be written as TSV).
|
|
|
177
|
|
|
178 Raises:
|
|
|
179 DataErr: If writing the output fails for any reason.
|
|
|
180
|
|
|
181 Returns:
|
|
|
182 None
|
|
|
183 """
|
|
|
184 try:
|
|
|
185 os.makedirs(os.path.dirname(path) or ".", exist_ok=True)
|
|
|
186 df.to_csv(path, sep="\t", index=False)
|
|
|
187 except Exception as e:
|
|
|
188 raise utils.DataErr(path, f"failed writing tabular output: {e}")
|
|
|
189
|
|
|
190 def is_placeholder(gid) -> bool:
|
|
|
191 """Return True if the gene id looks like a placeholder (e.g., 0/NA/NAN/empty)."""
|
|
|
192 if gid is None:
|
|
|
193 return True
|
|
|
194 s = str(gid).strip().lower()
|
|
|
195 return s in {"0", "", "na", "nan"} # lowercase for simple matching
|
|
|
196
|
|
|
197 def sample_valid_gene_ids(genes, limit=10):
|
|
|
198 """Yield up to `limit` valid gene IDs, skipping placeholders (e.g., the first 0 in RECON)."""
|
|
|
199 out = []
|
|
|
200 for g in genes:
|
|
|
201 gid = getattr(g, "id", getattr(g, "gene_id", g))
|
|
|
202 if not is_placeholder(gid):
|
|
|
203 out.append(str(gid))
|
|
|
204 if len(out) >= limit:
|
|
|
205 break
|
|
|
206 return out
|
|
|
207
|
|
|
208
|
|
|
209 ###############################- ENTRY POINT -################################
|
|
|
210 def main(args:List[str] = None) -> None:
|
|
|
211 """
|
|
|
212 Initialize and generate custom data based on the frontend input arguments.
|
|
|
213
|
|
|
214 Returns:
|
|
|
215 None
|
|
|
216 """
|
|
|
217 # Parse args from frontend (Galaxy XML)
|
|
|
218 global ARGS
|
|
|
219 ARGS = process_args(args)
|
|
|
220
|
|
496
|
221 # Convert name from list to string (handles names with spaces)
|
|
|
222 if isinstance(ARGS.name, list):
|
|
|
223 ARGS.name = ' '.join(ARGS.name)
|
|
491
|
224
|
|
|
225 if ARGS.input:
|
|
498
|
226 # Load a custom model from file with auto-detected format
|
|
|
227 detected_format = detect_file_format(ARGS.input)
|
|
|
228 model = load_custom_model(utils.FilePath.fromStrPath(ARGS.input), detected_format)
|
|
491
|
229 else:
|
|
|
230 # Load a built-in model
|
|
495
|
231 if not ARGS.model:
|
|
|
232 raise utils.ArgsErr("model", "either --model or --input must be provided", "None")
|
|
491
|
233
|
|
|
234 try:
|
|
|
235 model_enum = utils.Model[ARGS.model] # e.g., Model['ENGRO2']
|
|
|
236 except KeyError:
|
|
|
237 raise utils.ArgsErr("model", "one of Recon/ENGRO2/HMRcore/Custom_model", ARGS.model)
|
|
|
238
|
|
|
239 # Load built-in model (Model.getCOBRAmodel uses tool_dir to locate local models)
|
|
|
240 try:
|
|
|
241 model = model_enum.getCOBRAmodel(toolDir=ARGS.tool_dir)
|
|
|
242 except Exception as e:
|
|
|
243 # Wrap/normalize load errors as DataErr for consistency
|
|
|
244 raise utils.DataErr(ARGS.model, f"failed loading built-in model: {e}")
|
|
|
245
|
|
|
246 # Determine final model name: explicit --name overrides, otherwise use the model id
|
|
|
247
|
|
|
248 if ARGS.name == "ENGRO2" and ARGS.medium_selector != "Default":
|
|
|
249 df_mediums = pd.read_csv(ARGS.tool_dir + "/local/medium/medium.csv", index_col = 0)
|
|
|
250 ARGS.medium_selector = ARGS.medium_selector.replace("_", " ")
|
|
|
251 medium = df_mediums[[ARGS.medium_selector]]
|
|
|
252 medium = medium[ARGS.medium_selector].to_dict()
|
|
|
253
|
|
|
254 # Reset all medium reactions lower bound to zero
|
|
|
255 for rxn_id, _ in model.medium.items():
|
|
|
256 model.reactions.get_by_id(rxn_id).lower_bound = float(0.0)
|
|
|
257
|
|
|
258 # Apply selected medium uptake bounds (negative for uptake)
|
|
|
259 for reaction, value in medium.items():
|
|
|
260 if value is not None:
|
|
|
261 model.reactions.get_by_id(reaction).lower_bound = -float(value)
|
|
|
262
|
|
|
263 # Initialize translation_issues dictionary
|
|
|
264 translation_issues = {}
|
|
|
265
|
|
|
266 if (ARGS.name == "Recon" or ARGS.name == "ENGRO2") and ARGS.gene_format != "Default":
|
|
|
267 logging.basicConfig(level=logging.INFO)
|
|
|
268 logger = logging.getLogger(__name__)
|
|
|
269
|
|
|
270 model, translation_issues = modelUtils.translate_model_genes(
|
|
|
271 model=model,
|
|
|
272 mapping_df= pd.read_csv(ARGS.tool_dir + "/local/mappings/genes_human.csv", dtype={'entrez_id': str}),
|
|
|
273 target_nomenclature=ARGS.gene_format,
|
|
|
274 source_nomenclature='HGNC_symbol',
|
|
|
275 logger=logger
|
|
|
276 )
|
|
|
277
|
|
|
278 if ARGS.name == "Custom_model" and ARGS.gene_format != "Default":
|
|
|
279 logging.basicConfig(level=logging.INFO)
|
|
|
280 logger = logging.getLogger(__name__)
|
|
|
281
|
|
|
282 # Take a small, clean sample of gene IDs (skipping placeholders like 0)
|
|
|
283 ids_sample = sample_valid_gene_ids(model.genes, limit=10)
|
|
|
284 if not ids_sample:
|
|
|
285 raise utils.DataErr(
|
|
|
286 "Custom_model",
|
|
|
287 "No valid gene IDs found (many may be placeholders like 0)."
|
|
|
288 )
|
|
|
289
|
|
|
290 # Detect source nomenclature on the sample
|
|
|
291 types = []
|
|
|
292 for gid in ids_sample:
|
|
|
293 try:
|
|
|
294 t = modelUtils.gene_type(gid, "Custom_model")
|
|
|
295 except Exception as e:
|
|
|
296 # Keep it simple: skip problematic IDs
|
|
|
297 logger.debug(f"gene_type failed for {gid}: {e}")
|
|
|
298 t = None
|
|
|
299 if t:
|
|
|
300 types.append(t)
|
|
|
301
|
|
|
302 if not types:
|
|
|
303 raise utils.DataErr(
|
|
|
304 "Custom_model",
|
|
|
305 "Could not detect a known gene nomenclature from the sample."
|
|
|
306 )
|
|
|
307
|
|
|
308 unique_types = set(types)
|
|
|
309 if len(unique_types) > 1:
|
|
|
310 raise utils.DataErr(
|
|
|
311 "Custom_model",
|
|
|
312 "Mixed or inconsistent gene nomenclatures detected. "
|
|
|
313 "Please unify them before converting."
|
|
|
314 )
|
|
|
315
|
|
|
316 source_nomenclature = types[0]
|
|
|
317
|
|
|
318 # Convert only if needed
|
|
|
319 if source_nomenclature != ARGS.gene_format:
|
|
|
320 model, translation_issues = modelUtils.translate_model_genes(
|
|
|
321 model=model,
|
|
|
322 mapping_df= pd.read_csv(ARGS.tool_dir + "/local/mappings/genes_human.csv", dtype={'entrez_id': str}),
|
|
|
323 target_nomenclature=ARGS.gene_format,
|
|
|
324 source_nomenclature=source_nomenclature,
|
|
|
325 logger=logger
|
|
|
326 )
|
|
|
327
|
|
|
328 # generate data
|
|
|
329 rules = modelUtils.generate_rules(model, asParsed = False)
|
|
|
330 reactions = modelUtils.generate_reactions(model, asParsed = False)
|
|
|
331 bounds = modelUtils.generate_bounds(model)
|
|
|
332 medium = modelUtils.get_medium(model)
|
|
|
333 objective_function = modelUtils.extract_objective_coefficients(model)
|
|
|
334
|
|
|
335 if ARGS.name == "ENGRO2":
|
|
|
336 compartments = modelUtils.generate_compartments(model)
|
|
|
337
|
|
|
338 df_rules = pd.DataFrame(list(rules.items()), columns = ["ReactionID", "GPR"])
|
|
|
339 df_reactions = pd.DataFrame(list(reactions.items()), columns = ["ReactionID", "Formula"])
|
|
|
340
|
|
|
341 # Create DataFrame for translation issues
|
|
|
342 df_translation_issues = pd.DataFrame([
|
|
|
343 {"ReactionID": rxn_id, "TranslationIssues": issues}
|
|
|
344 for rxn_id, issues in translation_issues.items()
|
|
|
345 ])
|
|
|
346
|
|
|
347 df_bounds = bounds.reset_index().rename(columns = {"index": "ReactionID"})
|
|
|
348 df_medium = medium.rename(columns = {"reaction": "ReactionID"})
|
|
|
349 df_medium["InMedium"] = True
|
|
|
350
|
|
|
351 merged = df_reactions.merge(df_rules, on = "ReactionID", how = "outer")
|
|
|
352 merged = merged.merge(df_bounds, on = "ReactionID", how = "outer")
|
|
|
353 merged = merged.merge(objective_function, on = "ReactionID", how = "outer")
|
|
|
354 if ARGS.name == "ENGRO2":
|
|
|
355 merged = merged.merge(compartments, on = "ReactionID", how = "outer")
|
|
|
356 merged = merged.merge(df_medium, on = "ReactionID", how = "left")
|
|
|
357
|
|
|
358 # Add translation issues column
|
|
|
359 if not df_translation_issues.empty:
|
|
|
360 merged = merged.merge(df_translation_issues, on = "ReactionID", how = "left")
|
|
|
361 merged["TranslationIssues"] = merged["TranslationIssues"].fillna("")
|
|
|
362 else:
|
|
|
363 # Add empty TranslationIssues column if no issues found
|
|
|
364 #merged["TranslationIssues"] = ""
|
|
|
365 pass
|
|
|
366
|
|
|
367 merged["InMedium"] = merged["InMedium"].fillna(False)
|
|
|
368
|
|
|
369 merged = merged.sort_values(by = "InMedium", ascending = False)
|
|
|
370
|
|
|
371 if not ARGS.out_tabular:
|
|
|
372 raise utils.ArgsErr("out_tabular", "output path (--out_tabular) is required when output_format == tabular", ARGS.out_tabular)
|
|
|
373 save_as_tabular_df(merged, ARGS.out_tabular)
|
|
|
374 expected = ARGS.out_tabular
|
|
|
375
|
|
|
376 # verify output exists and non-empty
|
|
|
377 if not expected or not os.path.exists(expected) or os.path.getsize(expected) == 0:
|
|
|
378 raise utils.DataErr(expected, "Output not created or empty")
|
|
|
379
|
|
|
380 print("Metabolic_model_setting: completed successfully")
|
|
|
381
|
|
|
382 if __name__ == '__main__':
|
|
|
383
|
|
|
384 main()
|
