|
406
|
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
|
|
|
9 from typing import Optional, Tuple, Union, List, Dict
|
|
|
10 import utils.reaction_parsing as reactionUtils
|
|
418
|
11 import utils.model_utils as modelUtils
|
|
406
|
12
|
|
|
13 ARGS : argparse.Namespace
|
|
|
14 def process_args(args: List[str] = None) -> argparse.Namespace:
|
|
|
15 """
|
|
|
16 Parse command-line arguments for CustomDataGenerator.
|
|
|
17 """
|
|
|
18
|
|
|
19 parser = argparse.ArgumentParser(
|
|
|
20 usage="%(prog)s [options]",
|
|
|
21 description="Generate custom data from a given model"
|
|
|
22 )
|
|
|
23
|
|
|
24 parser.add_argument("--out_log", type=str, required=True,
|
|
|
25 help="Output log file")
|
|
|
26
|
|
|
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)")
|
|
|
33
|
|
|
34 parser.add_argument("--medium_selector", type=str, required=True,
|
|
|
35 help="Medium selection option")
|
|
|
36
|
|
|
37 parser.add_argument("--gene_format", type=str, default="Default",
|
|
|
38 help="Gene nomenclature format: Default (original), ENSNG, HGNC_SYMBOL, HGNC_ID, ENTREZ")
|
|
|
39
|
|
|
40 parser.add_argument("--out_tabular", type=str,
|
|
|
41 help="Output file for the merged dataset (CSV or XLSX)")
|
|
|
42
|
|
|
43 parser.add_argument("--tool_dir", type=str, default=os.path.dirname(__file__),
|
|
|
44 help="Tool directory (passed from Galaxy as $__tool_directory__)")
|
|
|
45
|
|
|
46
|
|
|
47 return parser.parse_args(args)
|
|
|
48
|
|
|
49 ################################- INPUT DATA LOADING -################################
|
|
|
50 def load_custom_model(file_path :utils.FilePath, ext :Optional[utils.FileFormat] = None) -> cobra.Model:
|
|
|
51 """
|
|
|
52 Loads a custom model from a file, either in JSON or XML format.
|
|
|
53
|
|
|
54 Args:
|
|
|
55 file_path : The path to the file containing the custom model.
|
|
|
56 ext : explicit file extension. Necessary for standard use in galaxy because of its weird behaviour.
|
|
|
57
|
|
|
58 Raises:
|
|
|
59 DataErr : if the file is in an invalid format or cannot be opened for whatever reason.
|
|
|
60
|
|
|
61 Returns:
|
|
|
62 cobra.Model : the model, if successfully opened.
|
|
|
63 """
|
|
|
64 ext = ext if ext else file_path.ext
|
|
|
65 try:
|
|
|
66 if ext is utils.FileFormat.XML:
|
|
|
67 return cobra.io.read_sbml_model(file_path.show())
|
|
|
68
|
|
|
69 if ext is utils.FileFormat.JSON:
|
|
|
70 return cobra.io.load_json_model(file_path.show())
|
|
|
71
|
|
|
72 except Exception as e: raise utils.DataErr(file_path, e.__str__())
|
|
|
73 raise utils.DataErr(file_path,
|
|
|
74 f"Formato \"{file_path.ext}\" non riconosciuto, sono supportati solo file JSON e XML")
|
|
|
75
|
|
|
76
|
|
|
77 ###############################- FILE SAVING -################################
|
|
|
78 def save_as_csv_filePath(data :dict, file_path :utils.FilePath, fieldNames :Tuple[str, str]) -> None:
|
|
|
79 """
|
|
|
80 Saves any dictionary-shaped data in a .csv file created at the given file_path as FilePath.
|
|
|
81
|
|
|
82 Args:
|
|
|
83 data : the data to be written to the file.
|
|
|
84 file_path : the path to the .csv file.
|
|
|
85 fieldNames : the names of the fields (columns) in the .csv file.
|
|
|
86
|
|
|
87 Returns:
|
|
|
88 None
|
|
|
89 """
|
|
|
90 with open(file_path.show(), 'w', newline='') as csvfile:
|
|
|
91 writer = csv.DictWriter(csvfile, fieldnames = fieldNames, dialect="excel-tab")
|
|
|
92 writer.writeheader()
|
|
|
93
|
|
|
94 for key, value in data.items():
|
|
|
95 writer.writerow({ fieldNames[0] : key, fieldNames[1] : value })
|
|
|
96
|
|
|
97 def save_as_csv(data :dict, file_path :str, fieldNames :Tuple[str, str]) -> None:
|
|
|
98 """
|
|
|
99 Saves any dictionary-shaped data in a .csv file created at the given file_path as string.
|
|
|
100
|
|
|
101 Args:
|
|
|
102 data : the data to be written to the file.
|
|
|
103 file_path : the path to the .csv file.
|
|
|
104 fieldNames : the names of the fields (columns) in the .csv file.
|
|
|
105
|
|
|
106 Returns:
|
|
|
107 None
|
|
|
108 """
|
|
|
109 with open(file_path, 'w', newline='') as csvfile:
|
|
|
110 writer = csv.DictWriter(csvfile, fieldnames = fieldNames, dialect="excel-tab")
|
|
|
111 writer.writeheader()
|
|
|
112
|
|
|
113 for key, value in data.items():
|
|
|
114 writer.writerow({ fieldNames[0] : key, fieldNames[1] : value })
|
|
|
115
|
|
|
116 def save_as_tabular_df(df: pd.DataFrame, path: str) -> None:
|
|
|
117 try:
|
|
|
118 os.makedirs(os.path.dirname(path) or ".", exist_ok=True)
|
|
|
119 df.to_csv(path, sep="\t", index=False)
|
|
|
120 except Exception as e:
|
|
|
121 raise utils.DataErr(path, f"failed writing tabular output: {e}")
|
|
|
122
|
|
|
123
|
|
|
124 ###############################- ENTRY POINT -################################
|
|
|
125 def main(args:List[str] = None) -> None:
|
|
|
126 """
|
|
|
127 Initializes everything and sets the program in motion based on the fronted input arguments.
|
|
|
128
|
|
|
129 Returns:
|
|
|
130 None
|
|
|
131 """
|
|
|
132 # get args from frontend (related xml)
|
|
|
133 global ARGS
|
|
|
134 ARGS = process_args(args)
|
|
|
135
|
|
|
136
|
|
|
137 if ARGS.input:
|
|
|
138 # load custom model
|
|
|
139 model = load_custom_model(
|
|
|
140 utils.FilePath.fromStrPath(ARGS.input), utils.FilePath.fromStrPath(ARGS.name).ext)
|
|
|
141 else:
|
|
|
142 # load built-in model
|
|
|
143
|
|
|
144 try:
|
|
|
145 model_enum = utils.Model[ARGS.model] # e.g., Model['ENGRO2']
|
|
|
146 except KeyError:
|
|
|
147 raise utils.ArgsErr("model", "one of Recon/ENGRO2/HMRcore/Custom_model", ARGS.model)
|
|
|
148
|
|
|
149 # Load built-in model (Model.getCOBRAmodel uses tool_dir to locate local models)
|
|
|
150 try:
|
|
|
151 model = model_enum.getCOBRAmodel(toolDir=ARGS.tool_dir)
|
|
|
152 except Exception as e:
|
|
|
153 # Wrap/normalize load errors as DataErr for consistency
|
|
|
154 raise utils.DataErr(ARGS.model, f"failed loading built-in model: {e}")
|
|
|
155
|
|
|
156 # Determine final model name: explicit --name overrides, otherwise use the model id
|
|
|
157
|
|
|
158 model_name = ARGS.name if ARGS.name else ARGS.model
|
|
|
159
|
|
|
160 if ARGS.name == "ENGRO2" and ARGS.medium_selector != "Default":
|
|
|
161 df_mediums = pd.read_csv(ARGS.tool_dir + "/local/medium/medium.csv", index_col = 0)
|
|
|
162 ARGS.medium_selector = ARGS.medium_selector.replace("_", " ")
|
|
|
163 medium = df_mediums[[ARGS.medium_selector]]
|
|
|
164 medium = medium[ARGS.medium_selector].to_dict()
|
|
|
165
|
|
|
166 # Set all reactions to zero in the medium
|
|
|
167 for rxn_id, _ in model.medium.items():
|
|
|
168 model.reactions.get_by_id(rxn_id).lower_bound = float(0.0)
|
|
|
169
|
|
|
170 # Set medium conditions
|
|
|
171 for reaction, value in medium.items():
|
|
|
172 if value is not None:
|
|
|
173 model.reactions.get_by_id(reaction).lower_bound = -float(value)
|
|
|
174
|
|
|
175 if ARGS.name == "ENGRO2" and ARGS.gene_format != "Default":
|
|
|
176
|
|
419
|
177 model = modelUtils.convert_genes(model, ARGS.gene_format.replace("HGNC_", "HGNC "))
|
|
406
|
178
|
|
|
179 # generate data
|
|
418
|
180 rules = modelUtils.generate_rules(model, asParsed = False)
|
|
|
181 reactions = modelUtils.generate_reactions(model, asParsed = False)
|
|
|
182 bounds = modelUtils.generate_bounds(model)
|
|
|
183 medium = modelUtils.get_medium(model)
|
|
406
|
184 if ARGS.name == "ENGRO2":
|
|
418
|
185 compartments = modelUtils.generate_compartments(model)
|
|
406
|
186
|
|
|
187 df_rules = pd.DataFrame(list(rules.items()), columns = ["ReactionID", "Rule"])
|
|
|
188 df_reactions = pd.DataFrame(list(reactions.items()), columns = ["ReactionID", "Reaction"])
|
|
|
189
|
|
|
190 df_bounds = bounds.reset_index().rename(columns = {"index": "ReactionID"})
|
|
|
191 df_medium = medium.rename(columns = {"reaction": "ReactionID"})
|
|
|
192 df_medium["InMedium"] = True # flag per indicare la presenza nel medium
|
|
|
193
|
|
|
194 merged = df_reactions.merge(df_rules, on = "ReactionID", how = "outer")
|
|
|
195 merged = merged.merge(df_bounds, on = "ReactionID", how = "outer")
|
|
|
196 if ARGS.name == "ENGRO2":
|
|
|
197 merged = merged.merge(compartments, on = "ReactionID", how = "outer")
|
|
|
198 merged = merged.merge(df_medium, on = "ReactionID", how = "left")
|
|
|
199
|
|
|
200 merged["InMedium"] = merged["InMedium"].fillna(False)
|
|
|
201
|
|
|
202 merged = merged.sort_values(by = "InMedium", ascending = False)
|
|
|
203
|
|
|
204 #out_file = os.path.join(ARGS.output_path, f"{os.path.basename(ARGS.name).split('.')[0]}_custom_data")
|
|
|
205
|
|
|
206 #merged.to_csv(out_file, sep = '\t', index = False)
|
|
|
207
|
|
|
208 ####
|
|
|
209
|
|
|
210 if not ARGS.out_tabular:
|
|
|
211 raise utils.ArgsErr("out_tabular", "output path (--out_tabular) is required when output_format == tabular", ARGS.out_tabular)
|
|
|
212 save_as_tabular_df(merged, ARGS.out_tabular)
|
|
|
213 expected = ARGS.out_tabular
|
|
|
214
|
|
|
215 # verify output exists and non-empty
|
|
|
216 if not expected or not os.path.exists(expected) or os.path.getsize(expected) == 0:
|
|
|
217 raise utils.DataErr(expected, "Output non creato o vuoto")
|
|
|
218
|
|
|
219 print("CustomDataGenerator: completed successfully")
|
|
|
220
|
|
|
221 if __name__ == '__main__':
|
|
|
222 main() |
