|
93
|
1 from __future__ import division
|
|
|
2 # galaxy complains this ^^^ needs to be at the very beginning of the file, for some reason.
|
|
|
3 import sys
|
|
|
4 import argparse
|
|
|
5 import collections
|
|
|
6 import pandas as pd
|
|
|
7 import pickle as pk
|
|
|
8 import utils.general_utils as utils
|
|
|
9 import utils.rule_parsing as ruleUtils
|
|
|
10 from typing import Union, Optional, List, Dict, Tuple, TypeVar
|
|
|
11
|
|
|
12 ERRORS = []
|
|
|
13 ########################## argparse ##########################################
|
|
|
14 ARGS :argparse.Namespace
|
|
147
|
15 def process_args(args:List[str] = None) -> argparse.Namespace:
|
|
93
|
16 """
|
|
|
17 Processes command-line arguments.
|
|
|
18
|
|
|
19 Args:
|
|
|
20 args (list): List of command-line arguments.
|
|
|
21
|
|
|
22 Returns:
|
|
|
23 Namespace: An object containing parsed arguments.
|
|
|
24 """
|
|
|
25 parser = argparse.ArgumentParser(
|
|
|
26 usage = '%(prog)s [options]',
|
|
|
27 description = "process some value's genes to create a comparison's map.")
|
|
|
28
|
|
|
29 parser.add_argument(
|
|
|
30 '-rs', '--rules_selector',
|
|
265
|
31 type = utils.Model, default = utils.Model.ENGRO2, choices = list(utils.Model),
|
|
93
|
32 help = 'chose which type of dataset you want use')
|
|
|
33
|
|
|
34 parser.add_argument("-rl", "--rule_list", type = str,
|
|
|
35 help = "path to input file with custom rules, if provided")
|
|
|
36
|
|
|
37 parser.add_argument("-rn", "--rules_name", type = str, help = "custom rules name")
|
|
|
38 # ^ I need this because galaxy converts my files into .dat but I need to know what extension they were in
|
|
|
39
|
|
|
40 parser.add_argument(
|
|
|
41 '-n', '--none',
|
|
|
42 type = utils.Bool("none"), default = True,
|
|
|
43 help = 'compute Nan values')
|
|
|
44
|
|
|
45 parser.add_argument(
|
|
|
46 '-td', '--tool_dir',
|
|
|
47 type = str,
|
|
|
48 required = True, help = 'your tool directory')
|
|
|
49
|
|
|
50 parser.add_argument(
|
|
|
51 '-ol', '--out_log',
|
|
|
52 type = str,
|
|
|
53 help = "Output log")
|
|
|
54
|
|
|
55 parser.add_argument(
|
|
|
56 '-in', '--input', #id รจ diventato in
|
|
|
57 type = str,
|
|
|
58 help = 'input dataset')
|
|
|
59
|
|
|
60 parser.add_argument(
|
|
|
61 '-ra', '--ras_output',
|
|
|
62 type = str,
|
|
|
63 required = True, help = 'ras output')
|
|
147
|
64
|
|
93
|
65
|
|
147
|
66 return parser.parse_args(args)
|
|
93
|
67
|
|
|
68 ############################ dataset input ####################################
|
|
|
69 def read_dataset(data :str, name :str) -> pd.DataFrame:
|
|
|
70 """
|
|
|
71 Read a dataset from a CSV file and return it as a pandas DataFrame.
|
|
|
72
|
|
|
73 Args:
|
|
|
74 data (str): Path to the CSV file containing the dataset.
|
|
|
75 name (str): Name of the dataset, used in error messages.
|
|
|
76
|
|
|
77 Returns:
|
|
|
78 pandas.DataFrame: DataFrame containing the dataset.
|
|
|
79
|
|
|
80 Raises:
|
|
|
81 pd.errors.EmptyDataError: If the CSV file is empty.
|
|
|
82 sys.exit: If the CSV file has the wrong format, the execution is aborted.
|
|
|
83 """
|
|
|
84 try:
|
|
|
85 dataset = pd.read_csv(data, sep = '\t', header = 0, engine='python')
|
|
|
86 except pd.errors.EmptyDataError:
|
|
|
87 sys.exit('Execution aborted: wrong format of ' + name + '\n')
|
|
|
88 if len(dataset.columns) < 2:
|
|
|
89 sys.exit('Execution aborted: wrong format of ' + name + '\n')
|
|
|
90 return dataset
|
|
|
91
|
|
|
92 ############################ load id e rules ##################################
|
|
|
93 def load_id_rules(reactions :Dict[str, Dict[str, List[str]]]) -> Tuple[List[str], List[Dict[str, List[str]]]]:
|
|
|
94 """
|
|
|
95 Load IDs and rules from a dictionary of reactions.
|
|
|
96
|
|
|
97 Args:
|
|
|
98 reactions (dict): A dictionary where keys are IDs and values are rules.
|
|
|
99
|
|
|
100 Returns:
|
|
|
101 tuple: A tuple containing two lists, the first list containing IDs and the second list containing rules.
|
|
|
102 """
|
|
|
103 ids, rules = [], []
|
|
|
104 for key, value in reactions.items():
|
|
|
105 ids.append(key)
|
|
|
106 rules.append(value)
|
|
|
107 return (ids, rules)
|
|
|
108
|
|
|
109 ############################ check_methods ####################################
|
|
|
110 def gene_type(l :str, name :str) -> str:
|
|
|
111 """
|
|
|
112 Determine the type of gene ID.
|
|
|
113
|
|
|
114 Args:
|
|
|
115 l (str): The gene identifier to check.
|
|
|
116 name (str): The name of the dataset, used in error messages.
|
|
|
117
|
|
|
118 Returns:
|
|
|
119 str: The type of gene ID ('hugo_id', 'ensembl_gene_id', 'symbol', or 'entrez_id').
|
|
|
120
|
|
|
121 Raises:
|
|
|
122 sys.exit: If the gene ID type is not supported, the execution is aborted.
|
|
|
123 """
|
|
|
124 if check_hgnc(l):
|
|
|
125 return 'hugo_id'
|
|
|
126 elif check_ensembl(l):
|
|
|
127 return 'ensembl_gene_id'
|
|
|
128 elif check_symbol(l):
|
|
|
129 return 'symbol'
|
|
|
130 elif check_entrez(l):
|
|
|
131 return 'entrez_id'
|
|
|
132 else:
|
|
|
133 sys.exit('Execution aborted:\n' +
|
|
|
134 'gene ID type in ' + name + ' not supported. Supported ID'+
|
|
|
135 'types are: HUGO ID, Ensemble ID, HUGO symbol, Entrez ID\n')
|
|
|
136
|
|
|
137 def check_hgnc(l :str) -> bool:
|
|
|
138 """
|
|
|
139 Check if a gene identifier follows the HGNC format.
|
|
|
140
|
|
|
141 Args:
|
|
|
142 l (str): The gene identifier to check.
|
|
|
143
|
|
|
144 Returns:
|
|
|
145 bool: True if the gene identifier follows the HGNC format, False otherwise.
|
|
|
146 """
|
|
|
147 if len(l) > 5:
|
|
|
148 if (l.upper()).startswith('HGNC:'):
|
|
|
149 return l[5:].isdigit()
|
|
|
150 else:
|
|
|
151 return False
|
|
|
152 else:
|
|
|
153 return False
|
|
|
154
|
|
|
155 def check_ensembl(l :str) -> bool:
|
|
|
156 """
|
|
|
157 Check if a gene identifier follows the Ensembl format.
|
|
|
158
|
|
|
159 Args:
|
|
|
160 l (str): The gene identifier to check.
|
|
|
161
|
|
|
162 Returns:
|
|
|
163 bool: True if the gene identifier follows the Ensembl format, False otherwise.
|
|
|
164 """
|
|
|
165 return l.upper().startswith('ENS')
|
|
|
166
|
|
|
167
|
|
|
168 def check_symbol(l :str) -> bool:
|
|
|
169 """
|
|
|
170 Check if a gene identifier follows the symbol format.
|
|
|
171
|
|
|
172 Args:
|
|
|
173 l (str): The gene identifier to check.
|
|
|
174
|
|
|
175 Returns:
|
|
|
176 bool: True if the gene identifier follows the symbol format, False otherwise.
|
|
|
177 """
|
|
|
178 if len(l) > 0:
|
|
|
179 if l[0].isalpha() and l[1:].isalnum():
|
|
|
180 return True
|
|
|
181 else:
|
|
|
182 return False
|
|
|
183 else:
|
|
|
184 return False
|
|
|
185
|
|
|
186 def check_entrez(l :str) -> bool:
|
|
|
187 """
|
|
|
188 Check if a gene identifier follows the Entrez ID format.
|
|
|
189
|
|
|
190 Args:
|
|
|
191 l (str): The gene identifier to check.
|
|
|
192
|
|
|
193 Returns:
|
|
|
194 bool: True if the gene identifier follows the Entrez ID format, False otherwise.
|
|
|
195 """
|
|
|
196 if len(l) > 0:
|
|
|
197 return l.isdigit()
|
|
|
198 else:
|
|
|
199 return False
|
|
|
200
|
|
|
201 ############################ gene #############################################
|
|
|
202 def data_gene(gene: pd.DataFrame, type_gene: str, name: str, gene_custom: Optional[Dict[str, str]]) -> Dict[str, str]:
|
|
|
203 """
|
|
|
204 Process gene data to ensure correct formatting and handle duplicates.
|
|
|
205
|
|
|
206 Args:
|
|
|
207 gene (DataFrame): DataFrame containing gene data.
|
|
|
208 type_gene (str): Type of gene data (e.g., 'hugo_id', 'ensembl_gene_id', 'symbol', 'entrez_id').
|
|
|
209 name (str): Name of the dataset.
|
|
|
210 gene_custom (dict or None): Custom gene data dictionary if provided.
|
|
|
211
|
|
|
212 Returns:
|
|
|
213 dict: A dictionary containing gene data with gene IDs as keys and corresponding values.
|
|
|
214 """
|
|
|
215 args = process_args()
|
|
|
216 for i in range(len(gene)):
|
|
|
217 tmp = gene.iloc[i, 0]
|
|
|
218 gene.iloc[i, 0] = tmp.strip().split('.')[0]
|
|
|
219
|
|
|
220 gene_dup = [item for item, count in
|
|
|
221 collections.Counter(gene[gene.columns[0]]).items() if count > 1]
|
|
|
222 pat_dup = [item for item, count in
|
|
|
223 collections.Counter(list(gene.columns)).items() if count > 1]
|
|
260
|
224
|
|
|
225 gene_in_rule = None
|
|
259
|
226
|
|
93
|
227 if gene_dup:
|
|
|
228 if gene_custom == None:
|
|
264
|
229
|
|
265
|
230 if str(args.rules_selector) == 'HMRcore':
|
|
93
|
231 gene_in_rule = pk.load(open(args.tool_dir + '/local/pickle files/HMRcore_genes.p', 'rb'))
|
|
|
232
|
|
265
|
233 elif str(args.rules_selector) == 'Recon':
|
|
93
|
234 gene_in_rule = pk.load(open(args.tool_dir + '/local/pickle files/Recon_genes.p', 'rb'))
|
|
|
235
|
|
265
|
236 elif str(args.rules_selector) == 'ENGRO2':
|
|
93
|
237 gene_in_rule = pk.load(open(args.tool_dir + '/local/pickle files/ENGRO2_genes.p', 'rb'))
|
|
263
|
238
|
|
260
|
239 utils.logWarning(f"{args.tool_dir}'/local/pickle files/ENGRO2_genes.p'", ARGS.out_log)
|
|
259
|
240
|
|
93
|
241 gene_in_rule = gene_in_rule.get(type_gene)
|
|
|
242
|
|
|
243 else:
|
|
|
244 gene_in_rule = gene_custom
|
|
260
|
245
|
|
93
|
246 tmp = []
|
|
|
247 for i in gene_dup:
|
|
|
248 if gene_in_rule.get(i) == 'ok':
|
|
|
249 tmp.append(i)
|
|
|
250 if tmp:
|
|
|
251 sys.exit('Execution aborted because gene ID '
|
|
|
252 +str(tmp)+' in '+name+' is duplicated\n')
|
|
|
253
|
|
|
254 if pat_dup: utils.logWarning(f"Warning: duplicated label\n{pat_dup} in {name}", ARGS.out_log)
|
|
|
255 return (gene.set_index(gene.columns[0])).to_dict()
|
|
|
256
|
|
|
257 ############################ resolve ##########################################
|
|
|
258 def replace_gene_value(l :str, d :str) -> Tuple[Union[int, float], list]:
|
|
|
259 """
|
|
|
260 Replace gene identifiers with corresponding values from a dictionary.
|
|
|
261
|
|
|
262 Args:
|
|
|
263 l (str): String of gene identifier.
|
|
|
264 d (str): String corresponding to its value.
|
|
|
265
|
|
|
266 Returns:
|
|
|
267 tuple: A tuple containing two lists: the first list contains replaced values, and the second list contains any errors encountered during replacement.
|
|
|
268 """
|
|
|
269 tmp = []
|
|
|
270 err = []
|
|
|
271 while l:
|
|
|
272 if isinstance(l[0], list):
|
|
|
273 tmp_rules, tmp_err = replace_gene_value(l[0], d)
|
|
|
274 tmp.append(tmp_rules)
|
|
|
275 err.extend(tmp_err)
|
|
|
276 else:
|
|
|
277 value = replace_gene(l[0], d)
|
|
|
278 tmp.append(value)
|
|
|
279 if value == None:
|
|
|
280 err.append(l[0])
|
|
|
281 l = l[1:]
|
|
|
282 return (tmp, err)
|
|
|
283
|
|
|
284 def replace_gene(l :str, d :str) -> Union[int, float]:
|
|
|
285 """
|
|
|
286 Replace a single gene identifier with its corresponding value from a dictionary.
|
|
|
287
|
|
|
288 Args:
|
|
|
289 l (str): Gene identifier to replace.
|
|
|
290 d (str): String corresponding to its value.
|
|
|
291
|
|
|
292 Returns:
|
|
|
293 float/int: Corresponding value from the dictionary if found, None otherwise.
|
|
|
294
|
|
|
295 Raises:
|
|
|
296 sys.exit: If the value associated with the gene identifier is not valid.
|
|
|
297 """
|
|
|
298 if l =='and' or l == 'or':
|
|
|
299 return l
|
|
|
300 else:
|
|
|
301 value = d.get(l, None)
|
|
|
302 if not(value == None or isinstance(value, (int, float))):
|
|
|
303 sys.exit('Execution aborted: ' + value + ' value not valid\n')
|
|
|
304 return value
|
|
|
305
|
|
|
306 T = TypeVar("T", bound = Optional[Union[int, float]])
|
|
|
307 def computes(val1 :T, op :str, val2 :T, cn :bool) -> T:
|
|
|
308 """
|
|
|
309 Compute the RAS value between two value and an operator ('and' or 'or').
|
|
|
310
|
|
|
311 Args:
|
|
|
312 val1(Optional(Union[float, int])): First value.
|
|
|
313 op (str): Operator ('and' or 'or').
|
|
|
314 val2(Optional(Union[float, int])): Second value.
|
|
|
315 cn (bool): Control boolean value.
|
|
|
316
|
|
|
317 Returns:
|
|
|
318 Optional(Union[float, int]): Result of the computation.
|
|
|
319 """
|
|
|
320 if val1 != None and val2 != None:
|
|
|
321 if op == 'and':
|
|
|
322 return min(val1, val2)
|
|
|
323 else:
|
|
|
324 return val1 + val2
|
|
|
325 elif op == 'and':
|
|
|
326 if cn is True:
|
|
|
327 if val1 != None:
|
|
|
328 return val1
|
|
|
329 elif val2 != None:
|
|
|
330 return val2
|
|
|
331 else:
|
|
|
332 return None
|
|
|
333 else:
|
|
|
334 return None
|
|
|
335 else:
|
|
|
336 if val1 != None:
|
|
|
337 return val1
|
|
|
338 elif val2 != None:
|
|
|
339 return val2
|
|
|
340 else:
|
|
|
341 return None
|
|
|
342
|
|
|
343 # ris should be Literal[None] but Literal is not supported in Python 3.7
|
|
|
344 def control(ris, l :List[Union[int, float, list]], cn :bool) -> Union[bool, int, float]: #Union[Literal[False], int, float]:
|
|
|
345 """
|
|
|
346 Control the format of the expression.
|
|
|
347
|
|
|
348 Args:
|
|
|
349 ris: Intermediate result.
|
|
|
350 l (list): Expression to control.
|
|
|
351 cn (bool): Control boolean value.
|
|
|
352
|
|
|
353 Returns:
|
|
|
354 Union[Literal[False], int, float]: Result of the control.
|
|
|
355 """
|
|
|
356 if len(l) == 1:
|
|
|
357 if isinstance(l[0], (float, int)) or l[0] == None:
|
|
|
358 return l[0]
|
|
|
359 elif isinstance(l[0], list):
|
|
|
360 return control(None, l[0], cn)
|
|
|
361 else:
|
|
|
362 return False
|
|
|
363 elif len(l) > 2:
|
|
|
364 return control_list(ris, l, cn)
|
|
|
365 else:
|
|
|
366 return False
|
|
|
367
|
|
|
368 def control_list(ris, l :List[Optional[Union[float, int, list]]], cn :bool) -> Optional[bool]: #Optional[Literal[False]]:
|
|
|
369 """
|
|
|
370 Control the format of a list of expressions.
|
|
|
371
|
|
|
372 Args:
|
|
|
373 ris: Intermediate result.
|
|
|
374 l (list): List of expressions to control.
|
|
|
375 cn (bool): Control boolean value.
|
|
|
376
|
|
|
377 Returns:
|
|
|
378 Optional[Literal[False]]: Result of the control.
|
|
|
379 """
|
|
|
380 while l:
|
|
|
381 if len(l) == 1:
|
|
|
382 return False
|
|
|
383 elif (isinstance(l[0], (float, int)) or
|
|
|
384 l[0] == None) and l[1] in ['and', 'or']:
|
|
|
385 if isinstance(l[2], (float, int)) or l[2] == None:
|
|
|
386 ris = computes(l[0], l[1], l[2], cn)
|
|
|
387 elif isinstance(l[2], list):
|
|
|
388 tmp = control(None, l[2], cn)
|
|
|
389 if tmp is False:
|
|
|
390 return False
|
|
|
391 else:
|
|
|
392 ris = computes(l[0], l[1], tmp, cn)
|
|
|
393 else:
|
|
|
394 return False
|
|
|
395 l = l[3:]
|
|
|
396 elif l[0] in ['and', 'or']:
|
|
|
397 if isinstance(l[1], (float, int)) or l[1] == None:
|
|
|
398 ris = computes(ris, l[0], l[1], cn)
|
|
|
399 elif isinstance(l[1], list):
|
|
|
400 tmp = control(None,l[1], cn)
|
|
|
401 if tmp is False:
|
|
|
402 return False
|
|
|
403 else:
|
|
|
404 ris = computes(ris, l[0], tmp, cn)
|
|
|
405 else:
|
|
|
406 return False
|
|
|
407 l = l[2:]
|
|
|
408 elif isinstance(l[0], list) and l[1] in ['and', 'or']:
|
|
|
409 if isinstance(l[2], (float, int)) or l[2] == None:
|
|
|
410 tmp = control(None, l[0], cn)
|
|
|
411 if tmp is False:
|
|
|
412 return False
|
|
|
413 else:
|
|
|
414 ris = computes(tmp, l[1], l[2], cn)
|
|
|
415 elif isinstance(l[2], list):
|
|
|
416 tmp = control(None, l[0], cn)
|
|
|
417 tmp2 = control(None, l[2], cn)
|
|
|
418 if tmp is False or tmp2 is False:
|
|
|
419 return False
|
|
|
420 else:
|
|
|
421 ris = computes(tmp, l[1], tmp2, cn)
|
|
|
422 else:
|
|
|
423 return False
|
|
|
424 l = l[3:]
|
|
|
425 else:
|
|
|
426 return False
|
|
|
427 return ris
|
|
|
428
|
|
|
429 ResolvedRules = Dict[str, List[Optional[Union[float, int]]]]
|
|
|
430 def resolve(genes: Dict[str, str], rules: List[str], ids: List[str], resolve_none: bool, name: str) -> Tuple[Optional[ResolvedRules], Optional[list]]:
|
|
|
431 """
|
|
|
432 Resolve rules using gene data to compute scores for each rule.
|
|
|
433
|
|
|
434 Args:
|
|
|
435 genes (dict): Dictionary containing gene data with gene IDs as keys and corresponding values.
|
|
|
436 rules (list): List of rules to resolve.
|
|
|
437 ids (list): List of IDs corresponding to the rules.
|
|
|
438 resolve_none (bool): Flag indicating whether to resolve None values in the rules.
|
|
|
439 name (str): Name of the dataset.
|
|
|
440
|
|
|
441 Returns:
|
|
|
442 tuple: A tuple containing resolved rules as a dictionary and a list of gene IDs not found in the data.
|
|
|
443 """
|
|
|
444 resolve_rules = {}
|
|
|
445 not_found = []
|
|
|
446 flag = False
|
|
|
447 for key, value in genes.items():
|
|
|
448 tmp_resolve = []
|
|
|
449 for i in range(len(rules)):
|
|
|
450 tmp = rules[i]
|
|
|
451 if tmp:
|
|
|
452 tmp, err = replace_gene_value(tmp, value)
|
|
|
453 if err:
|
|
|
454 not_found.extend(err)
|
|
|
455 ris = control(None, tmp, resolve_none)
|
|
|
456 if ris is False or ris == None:
|
|
|
457 tmp_resolve.append(None)
|
|
|
458 else:
|
|
|
459 tmp_resolve.append(ris)
|
|
|
460 flag = True
|
|
|
461 else:
|
|
|
462 tmp_resolve.append(None)
|
|
|
463 resolve_rules[key] = tmp_resolve
|
|
|
464
|
|
|
465 if flag is False:
|
|
|
466 utils.logWarning(
|
|
|
467 f"Warning: no computable score (due to missing gene values) for class {name}, the class has been disregarded",
|
|
|
468 ARGS.out_log)
|
|
|
469
|
|
|
470 return (None, None)
|
|
|
471
|
|
|
472 return (resolve_rules, list(set(not_found)))
|
|
|
473 ############################ create_ras #######################################
|
|
|
474 def create_ras(resolve_rules: Optional[ResolvedRules], dataset_name: str, rules: List[str], ids: List[str], file: str) -> None:
|
|
|
475 """
|
|
|
476 Create a RAS (Reaction Activity Score) file from resolved rules.
|
|
|
477
|
|
|
478 Args:
|
|
|
479 resolve_rules (dict): Dictionary containing resolved rules.
|
|
|
480 dataset_name (str): Name of the dataset.
|
|
|
481 rules (list): List of rules.
|
|
|
482 file (str): Path to the output RAS file.
|
|
|
483
|
|
|
484 Returns:
|
|
|
485 None
|
|
|
486 """
|
|
|
487 if resolve_rules is None:
|
|
|
488 utils.logWarning(f"Couldn't generate RAS for current dataset: {dataset_name}", ARGS.out_log)
|
|
|
489
|
|
|
490 for geni in resolve_rules.values():
|
|
|
491 for i, valori in enumerate(geni):
|
|
|
492 if valori == None:
|
|
|
493 geni[i] = 'None'
|
|
|
494
|
|
|
495 output_ras = pd.DataFrame.from_dict(resolve_rules)
|
|
|
496
|
|
|
497 output_ras.insert(0, 'Reactions', ids)
|
|
|
498 output_to_csv = pd.DataFrame.to_csv(output_ras, sep = '\t', index = False)
|
|
|
499
|
|
|
500 text_file = open(file, "w")
|
|
|
501
|
|
|
502 text_file.write(output_to_csv)
|
|
|
503 text_file.close()
|
|
|
504
|
|
|
505 ################################- NEW RAS COMPUTATION -################################
|
|
|
506 Expr = Optional[Union[int, float]]
|
|
|
507 Ras = Expr
|
|
|
508 def ras_for_cell_lines(dataset: pd.DataFrame, rules: Dict[str, ruleUtils.OpList]) -> Dict[str, Dict[str, Ras]]:
|
|
|
509 """
|
|
|
510 Generates the RAS scores for each cell line found in the dataset.
|
|
|
511
|
|
|
512 Args:
|
|
|
513 dataset (pd.DataFrame): Dataset containing gene values.
|
|
|
514 rules (dict): The dict containing reaction ids as keys and rules as values.
|
|
|
515
|
|
|
516 Side effects:
|
|
|
517 dataset : mut
|
|
|
518
|
|
|
519 Returns:
|
|
|
520 dict: A dictionary where each key corresponds to a cell line name and each value is a dictionary
|
|
|
521 where each key corresponds to a reaction ID and each value is its computed RAS score.
|
|
|
522 """
|
|
|
523 ras_values_by_cell_line = {}
|
|
|
524 dataset.set_index(dataset.columns[0], inplace=True)
|
|
|
525 # Considera tutte le colonne tranne la prima in cui ci sono gli hugo quindi va scartata
|
|
|
526 for cell_line_name in dataset.columns[1:]:
|
|
|
527 cell_line = dataset[cell_line_name].to_dict()
|
|
|
528 ras_values_by_cell_line[cell_line_name]= get_ras_values(rules, cell_line)
|
|
|
529 return ras_values_by_cell_line
|
|
|
530
|
|
|
531 def get_ras_values(value_rules: Dict[str, ruleUtils.OpList], dataset: Dict[str, Expr]) -> Dict[str, Ras]:
|
|
|
532 """
|
|
|
533 Computes the RAS (Reaction Activity Score) values for each rule in the given dict.
|
|
|
534
|
|
|
535 Args:
|
|
|
536 value_rules (dict): A dictionary where keys are reaction ids and values are OpLists.
|
|
|
537 dataset : gene expression data of one cell line.
|
|
|
538
|
|
|
539 Returns:
|
|
|
540 dict: A dictionary where keys are reaction ids and values are the computed RAS values for each rule.
|
|
|
541 """
|
|
|
542 return {key: ras_op_list(op_list, dataset) for key, op_list in value_rules.items()}
|
|
|
543
|
|
|
544 def get_gene_expr(dataset :Dict[str, Expr], name :str) -> Expr:
|
|
|
545 """
|
|
|
546 Extracts the gene expression of the given gene from a cell line dataset.
|
|
|
547
|
|
|
548 Args:
|
|
|
549 dataset : gene expression data of one cell line.
|
|
|
550 name : gene name.
|
|
|
551
|
|
|
552 Returns:
|
|
|
553 Expr : the gene's expression value.
|
|
|
554 """
|
|
|
555 expr = dataset.get(name, None)
|
|
|
556 if expr is None: ERRORS.append(name)
|
|
|
557
|
|
|
558 return expr
|
|
|
559
|
|
|
560 def ras_op_list(op_list: ruleUtils.OpList, dataset: Dict[str, Expr]) -> Ras:
|
|
|
561 """
|
|
|
562 Computes recursively the RAS (Reaction Activity Score) value for the given OpList, considering the specified flag to control None behavior.
|
|
|
563
|
|
|
564 Args:
|
|
|
565 op_list (OpList): The OpList representing a rule with gene values.
|
|
|
566 dataset : gene expression data of one cell line.
|
|
|
567
|
|
|
568 Returns:
|
|
|
569 Ras: The computed RAS value for the given OpList.
|
|
|
570 """
|
|
|
571 op = op_list.op
|
|
|
572 ras_value :Ras = None
|
|
|
573 if not op: return get_gene_expr(dataset, op_list[0])
|
|
|
574 if op is ruleUtils.RuleOp.AND and not ARGS.none and None in op_list: return None
|
|
|
575
|
|
|
576 for i in range(len(op_list)):
|
|
|
577 item = op_list[i]
|
|
|
578 if isinstance(item, ruleUtils.OpList):
|
|
|
579 item = ras_op_list(item, dataset)
|
|
|
580
|
|
|
581 else:
|
|
|
582 item = get_gene_expr(dataset, item)
|
|
|
583
|
|
|
584 if item is None:
|
|
|
585 if op is ruleUtils.RuleOp.AND and not ARGS.none: return None
|
|
|
586 continue
|
|
|
587
|
|
|
588 if ras_value is None:
|
|
|
589 ras_value = item
|
|
|
590 else:
|
|
|
591 ras_value = ras_value + item if op is ruleUtils.RuleOp.OR else min(ras_value, item)
|
|
|
592
|
|
|
593 return ras_value
|
|
|
594
|
|
|
595 def save_as_tsv(rasScores: Dict[str, Dict[str, Ras]], reactions :List[str]) -> None:
|
|
|
596 """
|
|
|
597 Save computed ras scores to the given path, as a tsv file.
|
|
|
598
|
|
|
599 Args:
|
|
|
600 rasScores : the computed ras scores.
|
|
|
601 path : the output tsv file's path.
|
|
|
602
|
|
|
603 Returns:
|
|
|
604 None
|
|
|
605 """
|
|
|
606 for scores in rasScores.values(): # this is actually a lot faster than using the ootb dataframe metod, sadly
|
|
|
607 for reactId, score in scores.items():
|
|
|
608 if score is None: scores[reactId] = "None"
|
|
|
609
|
|
|
610 output_ras = pd.DataFrame.from_dict(rasScores)
|
|
|
611 output_ras.insert(0, 'Reactions', reactions)
|
|
|
612 output_ras.to_csv(ARGS.ras_output, sep = '\t', index = False)
|
|
|
613
|
|
|
614 ############################ MAIN #############################################
|
|
|
615 #TODO: not used but keep, it will be when the new translator dicts will be used.
|
|
|
616 def translateGene(geneName :str, encoding :str, geneTranslator :Dict[str, Dict[str, str]]) -> str:
|
|
|
617 """
|
|
|
618 Translate gene from any supported encoding to HugoID.
|
|
|
619
|
|
|
620 Args:
|
|
|
621 geneName (str): the name of the gene in its current encoding.
|
|
|
622 encoding (str): the encoding.
|
|
|
623 geneTranslator (Dict[str, Dict[str, str]]): the dict containing all supported gene names
|
|
|
624 and encodings in the current model, mapping each to the corresponding HugoID encoding.
|
|
|
625
|
|
|
626 Raises:
|
|
|
627 ValueError: When the gene isn't supported in the model.
|
|
|
628
|
|
|
629 Returns:
|
|
|
630 str: the gene in HugoID encoding.
|
|
|
631 """
|
|
|
632 supportedGenesInEncoding = geneTranslator[encoding]
|
|
|
633 if geneName in supportedGenesInEncoding: return supportedGenesInEncoding[geneName]
|
|
|
634 raise ValueError(f"Gene \"{geneName}\" non trovato, verifica di star utilizzando il modello corretto!")
|
|
|
635
|
|
|
636 def load_custom_rules() -> Dict[str, ruleUtils.OpList]:
|
|
|
637 """
|
|
|
638 Opens custom rules file and extracts the rules. If the file is in .csv format an additional parsing step will be
|
|
|
639 performed, significantly impacting the runtime.
|
|
|
640
|
|
|
641 Returns:
|
|
|
642 Dict[str, ruleUtils.OpList] : dict mapping reaction IDs to rules.
|
|
|
643 """
|
|
|
644 datFilePath = utils.FilePath.fromStrPath(ARGS.rule_list) # actual file, stored in galaxy as a .dat
|
|
|
645
|
|
|
646 try: filenamePath = utils.FilePath.fromStrPath(ARGS.rules_name) # file's name in input, to determine its original ext
|
|
|
647 except utils.PathErr as err:
|
|
|
648 raise utils.PathErr(filenamePath, f"Please make sure your file's name is a valid file path, {err.msg}")
|
|
|
649
|
|
|
650 if filenamePath.ext is utils.FileFormat.PICKLE: return utils.readPickle(datFilePath)
|
|
|
651
|
|
|
652 # csv rules need to be parsed, those in a pickle format are taken to be pre-parsed.
|
|
|
653 return { line[0] : ruleUtils.parseRuleToNestedList(line[1]) for line in utils.readCsv(datFilePath) }
|
|
|
654
|
|
147
|
655 def main(args:List[str] = None) -> None:
|
|
93
|
656 """
|
|
|
657 Initializes everything and sets the program in motion based on the fronted input arguments.
|
|
|
658
|
|
|
659 Returns:
|
|
|
660 None
|
|
|
661 """
|
|
|
662 # get args from frontend (related xml)
|
|
|
663 global ARGS
|
|
147
|
664 ARGS = process_args(args)
|
|
93
|
665 print(ARGS.rules_selector)
|
|
|
666 # read dataset
|
|
|
667 dataset = read_dataset(ARGS.input, "dataset")
|
|
|
668 dataset.iloc[:, 0] = (dataset.iloc[:, 0]).astype(str)
|
|
|
669
|
|
|
670 # remove versioning from gene names
|
|
|
671 dataset.iloc[:, 0] = dataset.iloc[:, 0].str.split('.').str[0]
|
|
|
672
|
|
|
673 # handle custom models
|
|
|
674 model :utils.Model = ARGS.rules_selector
|
|
|
675 if model is utils.Model.Custom:
|
|
|
676 rules = load_custom_rules()
|
|
|
677 reactions = list(rules.keys())
|
|
|
678
|
|
|
679 save_as_tsv(ras_for_cell_lines(dataset, rules), reactions)
|
|
|
680 if ERRORS: utils.logWarning(
|
|
|
681 f"The following genes are mentioned in the rules but don't appear in the dataset: {ERRORS}",
|
|
|
682 ARGS.out_log)
|
|
|
683
|
|
|
684 return
|
|
|
685
|
|
|
686 # This is the standard flow of the ras_generator program, for non-custom models.
|
|
|
687 name = "RAS Dataset"
|
|
|
688 type_gene = gene_type(dataset.iloc[0, 0], name)
|
|
|
689
|
|
|
690 rules = model.getRules(ARGS.tool_dir)
|
|
|
691 genes = data_gene(dataset, type_gene, name, None)
|
|
|
692 ids, rules = load_id_rules(rules.get(type_gene))
|
|
|
693
|
|
|
694 resolve_rules, err = resolve(genes, rules, ids, ARGS.none, name)
|
|
|
695 create_ras(resolve_rules, name, rules, ids, ARGS.ras_output)
|
|
|
696
|
|
|
697 if err: utils.logWarning(
|
|
|
698 f"Warning: gene(s) {err} not found in class \"{name}\", " +
|
|
|
699 "the expression level for this gene will be considered NaN",
|
|
|
700 ARGS.out_log)
|
|
|
701
|
|
|
702 print("Execution succeded")
|
|
|
703
|
|
|
704 ###############################################################################
|
|
|
705 if __name__ == "__main__":
|
|
|
706 main() |
