|
0
|
1 import os
|
|
|
2 import csv
|
|
|
3 import cobra
|
|
|
4 import pickle
|
|
|
5 import argparse
|
|
|
6 import utils.general_utils as utils
|
|
|
7 import utils.rule_parsing as rulesUtils
|
|
|
8 from typing import Optional, Tuple, Union, Dict
|
|
|
9 import utils.reaction_parsing as reactionUtils
|
|
|
10
|
|
|
11 ARGS : argparse.Namespace
|
|
|
12 def process_args() -> argparse.Namespace:
|
|
|
13 """
|
|
|
14 Interfaces the script of a module with its frontend, making the user's choices for
|
|
|
15 various parameters available as values in code.
|
|
|
16
|
|
|
17 Args:
|
|
|
18 args : Always obtained (in file) from sys.argv
|
|
|
19
|
|
|
20 Returns:
|
|
|
21 Namespace : An object containing the parsed arguments
|
|
|
22 """
|
|
|
23 parser = argparse.ArgumentParser(
|
|
|
24 usage = "%(prog)s [options]",
|
|
|
25 description = "generate custom data from a given model")
|
|
|
26
|
|
|
27 parser.add_argument("-ol", "--out_log", type = str, required = True, help = "Output log")
|
|
|
28 parser.add_argument("-id", "--input", type = str, required = True, help = "Input model")
|
|
|
29 parser.add_argument("-mn", "--name", type = str, required = True, help = "Input model name")
|
|
|
30 # ^ I need this because galaxy converts my files into .dat but I need to know what extension they were in
|
|
|
31
|
|
|
32 parser.add_argument(
|
|
|
33 "-of", "--output_format",
|
|
|
34 # vvv I have to use .fromExt because enums in python are the plague and have been implemented by a chimpanzee.
|
|
|
35 type = utils.FileFormat.fromExt, default = utils.FileFormat.PICKLE,
|
|
|
36 choices = [utils.FileFormat.CSV, utils.FileFormat.PICKLE],
|
|
|
37 # ^^^ Not all variants are valid here, otherwise list(utils.FileFormat) would be best.
|
|
|
38 required = True, help = "Extension of all output files")
|
|
|
39
|
|
|
40 argsNamespace = parser.parse_args()
|
|
|
41 argsNamespace.out_dir = "result"
|
|
|
42 # ^ can't get this one to work from xml, there doesn't seem to be a way to get the directory attribute from the collection
|
|
|
43
|
|
|
44 return argsNamespace
|
|
|
45
|
|
|
46 ################################- INPUT DATA LOADING -################################
|
|
|
47 def load_custom_model(file_path :utils.FilePath, ext :Optional[utils.FileFormat] = None) -> cobra.Model:
|
|
|
48 """
|
|
|
49 Loads a custom model from a file, either in JSON or XML format.
|
|
|
50
|
|
|
51 Args:
|
|
|
52 file_path : The path to the file containing the custom model.
|
|
|
53 ext : explicit file extension. Necessary for standard use in galaxy because of its weird behaviour.
|
|
|
54
|
|
|
55 Raises:
|
|
|
56 DataErr : if the file is in an invalid format or cannot be opened for whatever reason.
|
|
|
57
|
|
|
58 Returns:
|
|
|
59 cobra.Model : the model, if successfully opened.
|
|
|
60 """
|
|
|
61 ext = ext if ext else file_path.ext
|
|
|
62 try:
|
|
|
63 if ext is utils.FileFormat.XML:
|
|
|
64 return cobra.io.read_sbml_model(file_path.show())
|
|
|
65
|
|
|
66 if ext is utils.FileFormat.JSON:
|
|
|
67 return cobra.io.load_json_model(file_path.show())
|
|
|
68
|
|
|
69 except Exception as e: raise utils.DataErr(file_path, e.__str__())
|
|
|
70 raise utils.DataErr(file_path,
|
|
|
71 f"Formato \"{file_path.ext}\" non riconosciuto, sono supportati solo file JSON e XML")
|
|
|
72
|
|
|
73 ################################- DATA GENERATION -################################
|
|
|
74 ReactionId = str
|
|
|
75 def generate_rules(model: cobra.Model, *, asParsed = True) -> Union[Dict[ReactionId, rulesUtils.OpList], Dict[ReactionId, str]]:
|
|
|
76 """
|
|
|
77 Generates a dictionary mapping reaction ids to rules from the model.
|
|
|
78
|
|
|
79 Args:
|
|
|
80 model : the model to derive data from.
|
|
|
81 asParsed : if True parses the rules to an optimized runtime format, otherwise leaves them as strings.
|
|
|
82
|
|
|
83 Returns:
|
|
|
84 Dict[ReactionId, rulesUtils.OpList] : the generated dictionary of parsed rules.
|
|
|
85 Dict[ReactionId, str] : the generated dictionary of raw rules.
|
|
|
86 """
|
|
|
87 # Is the below approach convoluted? yes
|
|
|
88 # Ok but is it inefficient? probably
|
|
|
89 # Ok but at least I don't have to repeat the check at every rule (I'm clinically insane)
|
|
|
90 _ruleGetter = lambda reaction : reaction.gene_reaction_rule
|
|
|
91 ruleExtractor = (lambda reaction :
|
|
|
92 rulesUtils.parseRuleToNestedList(_ruleGetter(reaction))) if asParsed else _ruleGetter
|
|
|
93
|
|
|
94 return {
|
|
|
95 reaction.id : ruleExtractor(reaction)
|
|
|
96 for reaction in model.reactions
|
|
|
97 if reaction.gene_reaction_rule }
|
|
|
98
|
|
|
99 def generate_reactions(model :cobra.Model, *, asParsed = True) -> Dict[ReactionId, str]:
|
|
|
100 """
|
|
|
101 Generates a dictionary mapping reaction ids to reaction formulas from the model.
|
|
|
102
|
|
|
103 Args:
|
|
|
104 model : the model to derive data from.
|
|
|
105 asParsed : if True parses the reactions to an optimized runtime format, otherwise leaves them as they are.
|
|
|
106
|
|
|
107 Returns:
|
|
|
108 Dict[ReactionId, str] : the generated dictionary.
|
|
|
109 """
|
|
|
110
|
|
|
111 unparsedReactions = {
|
|
|
112 reaction.id : reaction.reaction
|
|
|
113 for reaction in model.reactions
|
|
|
114 if reaction.reaction
|
|
|
115 }
|
|
|
116
|
|
|
117 if not asParsed: return unparsedReactions
|
|
|
118
|
|
|
119 return reactionUtils.create_reaction_dict(unparsedReactions)
|
|
|
120
|
|
|
121 ###############################- FILE SAVING -################################
|
|
|
122 def save_as_csv(data :dict, file_path :utils.FilePath, fieldNames :Tuple[str, str]) -> None:
|
|
|
123 """
|
|
|
124 Saves any dictionary-shaped data in a .csv file created at the given file_path.
|
|
|
125
|
|
|
126 Args:
|
|
|
127 data : the data to be written to the file.
|
|
|
128 file_path : the path to the .csv file.
|
|
|
129 fieldNames : the names of the fields (columns) in the .csv file.
|
|
|
130
|
|
|
131 Returns:
|
|
|
132 None
|
|
|
133 """
|
|
|
134 with open(file_path.show(), 'w', newline='') as csvfile:
|
|
|
135 writer = csv.DictWriter(csvfile, fieldnames = fieldNames)
|
|
|
136 writer.writeheader()
|
|
|
137
|
|
|
138 for key, value in data.items():
|
|
|
139 writer.writerow({ fieldNames[0] : key, fieldNames[1] : value })
|
|
|
140
|
|
|
141 ###############################- ENTRY POINT -################################
|
|
|
142 def main() -> None:
|
|
|
143 """
|
|
|
144 Initializes everything and sets the program in motion based on the fronted input arguments.
|
|
|
145
|
|
|
146 Returns:
|
|
|
147 None
|
|
|
148 """
|
|
|
149 # get args from frontend (related xml)
|
|
|
150 global ARGS
|
|
|
151 ARGS = process_args()
|
|
|
152
|
|
|
153 # this is the worst thing I've seen so far, congrats to the former MaREA devs for suggesting this!
|
|
|
154 if os.path.isdir(ARGS.out_dir) == False: os.makedirs(ARGS.out_dir)
|
|
|
155
|
|
|
156 # load custom model
|
|
|
157 model = load_custom_model(
|
|
|
158 utils.FilePath.fromStrPath(ARGS.input), utils.FilePath.fromStrPath(ARGS.name).ext)
|
|
|
159
|
|
|
160 # generate data and save it in the desired format and in a location galaxy understands
|
|
|
161 # (it should show up as a collection in the history)
|
|
|
162 rulesPath = utils.FilePath("rules", ARGS.output_format, prefix = ARGS.out_dir)
|
|
|
163 reactionsPath = utils.FilePath("reactions", ARGS.output_format, prefix = ARGS.out_dir)
|
|
|
164
|
|
|
165 if ARGS.output_format is utils.FileFormat.PICKLE:
|
|
|
166 rules = generate_rules(model, asParsed = True)
|
|
|
167 reactions = generate_reactions(model, asParsed = True)
|
|
|
168 utils.writePickle(rulesPath, rules)
|
|
|
169 utils.writePickle(reactionsPath, reactions)
|
|
|
170
|
|
|
171 elif ARGS.output_format is utils.FileFormat.CSV:
|
|
|
172 rules = generate_rules(model, asParsed = False)
|
|
|
173 reactions = generate_reactions(model, asParsed = False)
|
|
|
174 save_as_csv(rules, rulesPath, ("ReactionID", "Rule"))
|
|
|
175 save_as_csv(reactions, reactionsPath, ("ReactionID", "Reaction"))
|
|
|
176
|
|
|
177 # ^ Please if anyone works on this after updating python to 3.12 change the if/elif into a match statement!!
|
|
|
178
|
|
|
179 if __name__ == '__main__':
|
|
|
180 main() |
