1
|
|
|
from __future__ import annotations |
|
|
|
|
2
|
|
|
|
3
|
|
|
import enum |
4
|
|
|
from collections import defaultdict |
5
|
|
|
from dataclasses import dataclass |
6
|
|
|
from functools import total_ordering |
7
|
|
|
from pathlib import Path |
8
|
|
|
from typing import FrozenSet, Iterable, List, Mapping, Optional, Sequence, Set, Union |
9
|
|
|
|
10
|
|
|
import pandas as pd |
|
|
|
|
11
|
|
|
from pocketutils.core.exceptions import LookupFailedError, DataIntegrityError |
|
|
|
|
12
|
|
|
|
13
|
|
|
from mandos.model import MultipleMatchesError |
14
|
|
|
from typeddfs import TypedDfs |
|
|
|
|
15
|
|
|
|
16
|
|
|
from mandos.model.utils.setup import logger |
|
|
|
|
17
|
|
|
from mandos.model.utils import CleverEnum |
18
|
|
|
|
19
|
|
|
|
20
|
|
|
class KnownTaxa: |
21
|
|
|
""" |
22
|
|
|
Taxa whose IDs are used in the code. |
23
|
|
|
""" |
24
|
|
|
|
25
|
|
|
biota = 131567 # 2 million nodes |
26
|
|
|
eukaryota = 2759 # 1.5 million nodes |
27
|
|
|
metazoa = 33208 # 1 million |
28
|
|
|
vertebrata = 7742 # 100,000 nodes |
29
|
|
|
euteleostomi = 117571 # 100,000 nodes |
30
|
|
|
human = 9606 |
31
|
|
|
rat = 10116 |
32
|
|
|
mouse = 10090 |
33
|
|
|
|
34
|
|
|
|
35
|
|
|
class NameType(CleverEnum): |
36
|
|
|
""" |
37
|
|
|
Scientific name, common name, or mnemonic. |
38
|
|
|
""" |
39
|
|
|
|
40
|
|
|
scientific = enum.auto() |
41
|
|
|
common = enum.auto() |
42
|
|
|
mnemonic = enum.auto() |
43
|
|
|
|
44
|
|
|
|
45
|
|
|
TaxonomyDf = ( |
46
|
|
|
TypedDfs.typed("TaxonomyDf") |
47
|
|
|
.require("taxon", "parent", dtype=int) |
48
|
|
|
.require("mnemonic", "scientific_name", "common_name", dtype=str) |
49
|
|
|
.strict() |
50
|
|
|
.secure() |
51
|
|
|
).build() |
52
|
|
|
|
53
|
|
|
|
54
|
|
|
@total_ordering |
|
|
|
|
55
|
|
|
@dataclass() |
56
|
|
|
class Taxon: |
57
|
|
|
""" """ |
58
|
|
|
|
59
|
|
|
# we can't use frozen=True because we have both parents and children |
60
|
|
|
# instead, just use properties |
61
|
|
|
__id: int |
62
|
|
|
__scientific_name: str |
63
|
|
|
__common_name: Optional[str] |
64
|
|
|
__mnemonic: Optional[str] |
65
|
|
|
__parent: Optional[Taxon] |
66
|
|
|
__children: Set[Taxon] |
67
|
|
|
|
68
|
|
|
@property |
69
|
|
|
def id(self) -> int: |
|
|
|
|
70
|
|
|
""" |
71
|
|
|
Returns the UniProt ID of this taxon. |
72
|
|
|
""" |
73
|
|
|
return self.__id |
74
|
|
|
|
75
|
|
|
@property |
76
|
|
|
def scientific_name(self) -> str: |
77
|
|
|
""" |
78
|
|
|
Returns the scientific name of this taxon. |
79
|
|
|
""" |
80
|
|
|
return self.__scientific_name |
81
|
|
|
|
82
|
|
|
@property |
83
|
|
|
def common_name(self) -> Optional[str]: |
84
|
|
|
""" |
85
|
|
|
Returns the common name of this taxon, or None if it has none. |
86
|
|
|
""" |
87
|
|
|
return self.__common_name |
88
|
|
|
|
89
|
|
|
@property |
90
|
|
|
def mnemonic(self) -> Optional[str]: |
91
|
|
|
""" |
92
|
|
|
Returns the mnemonic of this taxon, or None if it has none. |
93
|
|
|
Only ~16 taxa have mnemonics as of 2021-08. |
94
|
|
|
For example: "BOVIN" `<https://www.uniprot.org/taxonomy/9913>`_. |
95
|
|
|
""" |
96
|
|
|
return self.__mnemonic |
97
|
|
|
|
98
|
|
|
@property |
99
|
|
|
def keys(self) -> FrozenSet[Union[int, str]]: |
100
|
|
|
""" |
101
|
|
|
Returns the IDs and names that can be used to find this taxon. |
102
|
|
|
Specifically, includes the ID (int), scientific name (str), |
103
|
|
|
common name (str; if any), and mnemonic (str; if any). |
104
|
|
|
""" |
105
|
|
|
keys = {self.id, self.scientific_name, self.common_name, self.mnemonic} |
106
|
|
|
return frozenset({s for s in keys if s is not None}) |
107
|
|
|
|
108
|
|
|
@property |
109
|
|
|
def as_series(self) -> pd.Series: |
|
|
|
|
110
|
|
|
return pd.Series( |
111
|
|
|
dict( |
112
|
|
|
taxon=self.id, |
113
|
|
|
scientific_name=self.scientific_name, |
114
|
|
|
common_name=self.common_name, |
115
|
|
|
mnemonic=self.mnemonic, |
116
|
|
|
parent=self.parent.id, |
117
|
|
|
) |
118
|
|
|
) |
119
|
|
|
|
120
|
|
|
@property |
121
|
|
|
def parent(self) -> Taxon: |
122
|
|
|
""" |
123
|
|
|
Returns the parent of this taxon. |
124
|
|
|
""" |
125
|
|
|
return self.__parent |
126
|
|
|
|
127
|
|
|
@property |
128
|
|
|
def children(self) -> Set[Taxon]: |
129
|
|
|
""" |
130
|
|
|
Returns the immediate descendents of this taxon. |
131
|
|
|
""" |
132
|
|
|
return set(self.__children) |
133
|
|
|
|
134
|
|
|
@property |
135
|
|
|
def ancestors(self) -> Sequence[Taxon]: |
136
|
|
|
""" |
137
|
|
|
Returns all taxa that are ancestors of, or identical to, this taxon. |
138
|
|
|
""" |
139
|
|
|
lst = [] |
140
|
|
|
self._ancestors(lst) |
141
|
|
|
return lst |
142
|
|
|
|
143
|
|
|
@property |
144
|
|
|
def descendents(self) -> Sequence[Taxon]: |
145
|
|
|
""" |
146
|
|
|
Returns all taxa that are descendents of, or identical to, this taxon. |
147
|
|
|
""" |
148
|
|
|
lst = [] |
149
|
|
|
self._descendents(lst) |
150
|
|
|
return lst |
151
|
|
|
|
152
|
|
|
def _ancestors(self, values: List[Taxon]) -> None: |
153
|
|
|
values.append(self.parent) |
154
|
|
|
self.parent._ancestors(values) |
|
|
|
|
155
|
|
|
|
156
|
|
|
def _descendents(self, values: List[Taxon]) -> None: |
157
|
|
|
values.extend(self.children) |
158
|
|
|
for child in self.children: |
159
|
|
|
child._descendents(values) |
|
|
|
|
160
|
|
|
|
161
|
|
|
def __str__(self): |
162
|
|
|
return repr(self) |
163
|
|
|
|
164
|
|
|
def __repr__(self): |
165
|
|
|
parent = self.parent.id if self.parent else "none" |
166
|
|
|
return f"{self.__class__.__name__}({self.id}: {self.scientific_name} (parent={parent}))" |
167
|
|
|
|
168
|
|
|
def __hash__(self): |
169
|
|
|
return hash(self.id) |
170
|
|
|
|
171
|
|
|
def __eq__(self, other): |
172
|
|
|
return self.id == other.id |
173
|
|
|
|
174
|
|
|
def __lt__(self, other): |
175
|
|
|
return self.id < other.id |
176
|
|
|
|
177
|
|
|
|
178
|
|
|
TaxaIdsAndNames = Union[int, str, Taxon, Iterable[Union[int, str, Taxon]]] |
179
|
|
|
TaxonIdOrName = Union[int, str, Taxon] |
180
|
|
|
|
181
|
|
|
|
182
|
|
|
@dataclass() |
183
|
|
|
class _Taxon(Taxon): |
184
|
|
|
""" |
185
|
|
|
An internal, modifiable taxon for building the tree. |
186
|
|
|
""" |
187
|
|
|
|
188
|
|
|
def set_names(self, scientific: str, common: Optional[str], mnemonic: Optional[str]): |
|
|
|
|
189
|
|
|
self.__scientific_name = scientific |
190
|
|
|
self.__common_name = common |
191
|
|
|
self.__mnemonic = mnemonic |
192
|
|
|
|
193
|
|
|
def set_parent(self, parent: _Taxon): |
|
|
|
|
194
|
|
|
self.__parent = parent |
195
|
|
|
|
196
|
|
|
def add_child(self, child: _Taxon): |
|
|
|
|
197
|
|
|
self.__children.add(child) |
198
|
|
|
|
199
|
|
|
# weirdly these are required again -- probably an issue with dataclass |
200
|
|
|
|
201
|
|
|
def __str__(self): |
202
|
|
|
return repr(self) |
203
|
|
|
|
204
|
|
|
def __repr__(self): |
205
|
|
|
return f"{self.__class__.__name__}({self.id}: {self.scientific_name} (parent={self.parent.id if self.parent else 'none'}))" |
|
|
|
|
206
|
|
|
|
207
|
|
|
def __hash__(self): |
208
|
|
|
return hash(self.id) |
209
|
|
|
|
210
|
|
|
def __eq__(self, other): |
211
|
|
|
return self.id == other.id |
212
|
|
|
|
213
|
|
|
def __lt__(self, other): |
214
|
|
|
return self.id < other.id |
215
|
|
|
|
216
|
|
|
|
217
|
|
|
class Taxonomy: |
|
|
|
|
218
|
|
|
""" |
219
|
|
|
A taxonomic tree of organisms from UniProt. |
220
|
|
|
Elements in the tree can be looked up by name or ID using ``__getitem__`` and ``get``. |
221
|
|
|
""" |
222
|
|
|
|
223
|
|
|
def __init__(self, by_id: Mapping[int, Taxon], by_name: Mapping[str, FrozenSet[Taxon]]): |
224
|
|
|
""" |
225
|
|
|
|
226
|
|
|
Args: |
227
|
|
|
by_id: |
228
|
|
|
""" |
229
|
|
|
# constructor provided for consistency with the members |
230
|
|
|
self._by_id = dict(by_id) |
231
|
|
|
self._by_name = dict(by_name) |
232
|
|
|
# this probably isn't actually possible |
233
|
|
|
if len(self) == 0: |
234
|
|
|
logger.warning(f"{self} contains 0 taxa") |
235
|
|
|
|
236
|
|
|
@classmethod |
237
|
|
|
def from_trees(cls, taxonomies: Sequence[Taxonomy]) -> Taxonomy: |
|
|
|
|
238
|
|
|
# we need to rewrite the ancestors, which from_df already does |
239
|
|
|
# so we'll just use that |
240
|
|
|
dfs = [tree.to_df() for tree in taxonomies] |
241
|
|
|
df = TaxonomyDf(pd.concat(dfs, ignore_index=True)) |
|
|
|
|
242
|
|
|
df = df.drop_duplicates().sort_values("taxon") |
|
|
|
|
243
|
|
|
return Taxonomy.from_df(df) |
244
|
|
|
|
245
|
|
|
@classmethod |
246
|
|
|
def from_list(cls, taxa: Sequence[Taxon]) -> Taxonomy: |
|
|
|
|
247
|
|
|
by_id = {x.id: x for x in taxa} |
248
|
|
|
by_name = cls._build_by_name(by_id.values()) |
249
|
|
|
tax = Taxonomy(by_id, by_name) |
250
|
|
|
# catch duplicate values |
251
|
|
|
if len(tax._by_id) != len(taxa): |
|
|
|
|
252
|
|
|
raise DataIntegrityError(f"{len(tax._by_id)} != {len(taxa)}") |
|
|
|
|
253
|
|
|
return tax |
254
|
|
|
|
255
|
|
|
@classmethod |
256
|
|
|
def from_path(cls, path: Path) -> Taxonomy: |
257
|
|
|
""" |
258
|
|
|
Reads from a DataFrame file. |
259
|
|
|
""" |
260
|
|
|
df = TaxonomyDf.read_file(path) |
|
|
|
|
261
|
|
|
return cls.from_df(df) |
262
|
|
|
|
263
|
|
|
@classmethod |
264
|
|
|
def from_df(cls, df: TaxonomyDf) -> Taxonomy: |
|
|
|
|
265
|
|
|
""" |
266
|
|
|
Reads from a DataFrame from a file provided by a UniProt download. |
267
|
|
|
Strips any entries with missing or empty-string scientific names. |
268
|
|
|
|
269
|
|
|
Args: |
270
|
|
|
df: A TaxonomyDf DataFrame |
271
|
|
|
|
272
|
|
|
Returns: |
273
|
|
|
The corresponding taxonomic tree |
274
|
|
|
""" |
275
|
|
|
# just build up a tree, sticking the elements in by_id |
276
|
|
|
tax = {} |
277
|
|
|
for row in df.itertuples(): |
278
|
|
|
_new_child = _Taxon( |
279
|
|
|
row.taxon, row.scientific_name, row.common_name, row.mnemonic, None, set() |
280
|
|
|
) |
281
|
|
|
child = tax.setdefault(row.taxon, _new_child) |
282
|
|
|
child.set_names(row.scientific_name, row.common_name, row.mnemonic) |
283
|
|
|
if row.parent != 0: |
284
|
|
|
_new_parent = _Taxon(row.parent, "", None, None, None, set()) |
285
|
|
|
parent = tax.setdefault(row.parent, _new_parent) |
286
|
|
|
child.set_parent(parent) |
287
|
|
|
parent.add_child(child) |
288
|
|
|
bad = [t for t in tax.values() if t.scientific_name.strip() == ""] |
289
|
|
|
if len(bad) > 0: |
290
|
|
|
raise DataIntegrityError( |
291
|
|
|
f"{len(bad)} taxa with missing or empty scientific names: {bad}." |
292
|
|
|
) |
293
|
|
|
for v in tax.values(): |
|
|
|
|
294
|
|
|
v.__class__ = Taxon |
295
|
|
|
by_name = cls._build_by_name(tax.values()) |
296
|
|
|
return Taxonomy(tax, by_name) |
297
|
|
|
|
298
|
|
|
def to_df(self) -> TaxonomyDf: |
|
|
|
|
299
|
|
|
return TaxonomyDf.convert(pd.DataFrame([taxon.as_series for taxon in self.taxa])) |
300
|
|
|
|
301
|
|
|
@property |
302
|
|
|
def taxa(self) -> Sequence[Taxon]: |
303
|
|
|
""" |
304
|
|
|
Returns all taxa in the tree. |
305
|
|
|
""" |
306
|
|
|
return list(self._by_id.values()) |
307
|
|
|
|
308
|
|
|
@property |
309
|
|
|
def roots(self) -> Sequence[Taxon]: |
310
|
|
|
""" |
311
|
|
|
Returns the roots of the tree (at least 1). |
312
|
|
|
""" |
313
|
|
|
return [k for k in self.taxa if k.parent is None or k.parent not in self] |
314
|
|
|
|
315
|
|
|
@property |
316
|
|
|
def leaves(self) -> Sequence[Taxon]: |
317
|
|
|
""" |
318
|
|
|
Returns the leaves (typically species or sub-species) of the tree. |
319
|
|
|
""" |
320
|
|
|
return [k for k in self.taxa if len(k.children) == 0] |
321
|
|
|
|
322
|
|
|
def exclude_subtree(self, item: Union[int, Taxon]) -> Taxonomy: |
323
|
|
|
""" |
324
|
|
|
Returns a new tree that excludes a single specified taxon and its descendents. |
325
|
|
|
""" |
326
|
|
|
descendents = self.get_by_id_or_name(item) |
327
|
|
|
for i in set(descendents): |
328
|
|
|
descendents += i.descendents |
329
|
|
|
by_id = {d.id: d for d in descendents} |
330
|
|
|
by_name = self.__class__._build_by_name(by_id.values()) |
|
|
|
|
331
|
|
|
return Taxonomy(by_id, by_name) |
332
|
|
|
|
333
|
|
|
def exclude_subtrees_by_ids_or_names(self, items: TaxaIdsAndNames) -> Taxonomy: |
334
|
|
|
""" |
335
|
|
|
Returns a tree tree that excludes taxa that are descendents of the specified taxa. |
336
|
|
|
If a name is used in multiple taxa, all of those will be used to exclude. |
337
|
|
|
|
338
|
|
|
Arguments: |
339
|
|
|
items: A scientific name, common name, or mnemonic; or a sequence of them |
340
|
|
|
""" |
341
|
|
|
if isinstance(items, (int, str, Taxon)): |
342
|
|
|
items = [items] |
343
|
|
|
bad_taxa = self.subtrees_by_ids_or_names(items).taxa |
344
|
|
|
by_id = {i: t for i, t in self._by_id.items() if i not in bad_taxa} |
345
|
|
|
by_name = self.__class__._build_by_name(by_id.values()) |
|
|
|
|
346
|
|
|
return Taxonomy(by_id, by_name) |
347
|
|
|
|
348
|
|
|
def subtree(self, item: int) -> Taxonomy: |
349
|
|
|
""" |
350
|
|
|
Returns the tree that is rooted at a single taxon (by ID). |
351
|
|
|
""" |
352
|
|
|
item = self[item] |
353
|
|
|
descendents = {item, *item.descendents} |
354
|
|
|
by_id = {d.id: d for d in descendents} |
355
|
|
|
by_name = self.__class__._build_by_name(by_id.values()) |
|
|
|
|
356
|
|
|
return Taxonomy(by_id, by_name) |
357
|
|
|
|
358
|
|
|
def subtrees_by_ids_or_names(self, items: TaxaIdsAndNames) -> Taxonomy: |
359
|
|
|
""" |
360
|
|
|
Returns the tree that is rooted at the specified taxa (by name or ID). |
361
|
|
|
The tree will have *at most* ``len(items)`` roots. |
362
|
|
|
|
363
|
|
|
Arguments: |
364
|
|
|
items: A scientific name, common name, or mnemonic; or a sequence of them |
365
|
|
|
""" |
366
|
|
|
if isinstance(items, (int, str, Taxon)): |
367
|
|
|
items = [items] |
368
|
|
|
descendents: Set[Taxon] = set() |
369
|
|
|
for item in items: |
370
|
|
|
for taxon in self.get_by_id_or_name(item): |
371
|
|
|
descendents.update({taxon, *taxon.descendents}) |
372
|
|
|
by_id = {d.id: d for d in descendents} |
373
|
|
|
by_name = self.__class__._build_by_name(by_id.values()) |
|
|
|
|
374
|
|
|
return Taxonomy(by_id, by_name) |
375
|
|
|
|
376
|
|
|
def subtrees_by_name(self, item: str) -> Taxonomy: |
377
|
|
|
""" |
378
|
|
|
Returns the tree rooted at the taxa with the specified scientific name. |
379
|
|
|
|
380
|
|
|
Arguments: |
381
|
|
|
item: A scientific name, common name, or mnemonic |
382
|
|
|
""" |
383
|
|
|
return self.subtrees_by_names(item) |
384
|
|
|
|
385
|
|
|
def subtrees_by_names(self, items: Iterable[str]) -> Taxonomy: |
386
|
|
|
""" |
387
|
|
|
Returns the tree rooted at the specified taxa (by scientific name). |
388
|
|
|
|
389
|
|
|
Arguments: |
390
|
|
|
items: A sequence of scientific name, common name, and/or mnemonics |
391
|
|
|
""" |
392
|
|
|
descendents: Set[Taxon] = set() |
393
|
|
|
for item in items: |
394
|
|
|
for taxon in self._by_name.get(item, []): |
395
|
|
|
descendents.update({taxon, *taxon.descendents}) |
396
|
|
|
by_id = {d.id: d for d in descendents} |
397
|
|
|
by_name = self.__class__._build_by_name(by_id.values()) |
|
|
|
|
398
|
|
|
return Taxonomy(by_id, by_name) |
399
|
|
|
|
400
|
|
|
def req_one_by_name(self, item: str) -> Taxon: |
401
|
|
|
""" |
402
|
|
|
Gets a single taxon by its name. |
403
|
|
|
If there are multiple, returns the first (lowest ID). |
404
|
|
|
Raises an error if there are no matches. |
405
|
|
|
|
406
|
|
|
Arguments: |
407
|
|
|
item: A scientific name, common name, or mnemonic |
408
|
|
|
|
409
|
|
|
Raises: |
410
|
|
|
LookupError: If not found |
411
|
|
|
MultipleMatchesError: If multiple are found |
412
|
|
|
""" |
413
|
|
|
one = self.get_one_by_name(item) |
414
|
|
|
if one is None: |
415
|
|
|
raise LookupFailedError(f"No taxa for {item}") |
416
|
|
|
return one |
417
|
|
|
|
418
|
|
|
def req_only_by_name(self, item: str) -> Taxon: |
419
|
|
|
""" |
420
|
|
|
Gets a single taxon by its name. |
421
|
|
|
Raises an error if there are multiple matches for the name, or if there are no matches. |
422
|
|
|
|
423
|
|
|
Arguments: |
424
|
|
|
item: A scientific name, common name, or mnemonic |
425
|
|
|
|
426
|
|
|
Raises: |
427
|
|
|
LookupError: If not found |
428
|
|
|
MultipleMatchesError: If multiple are found |
429
|
|
|
""" |
430
|
|
|
taxa = self.get_by_name(item) |
431
|
|
|
ids = ",".join([str(t.id) for t in taxa]) |
432
|
|
|
if len(taxa) > 1: |
|
|
|
|
433
|
|
|
raise MultipleMatchesError(f"Got multiple results for {item}: {ids}") |
434
|
|
|
elif len(taxa) == 0: |
435
|
|
|
raise LookupFailedError(f"No taxa for {item}") |
436
|
|
|
return next(iter(taxa)) |
437
|
|
|
|
438
|
|
|
def get_one_by_name(self, item: str) -> Optional[Taxon]: |
439
|
|
|
""" |
440
|
|
|
Gets a single taxon by its name. |
441
|
|
|
If there are multiple, returns the first (lowest ID). |
442
|
|
|
If there are none, returns ``None``. |
443
|
|
|
Logs at warning level if multiple matched. |
444
|
|
|
|
445
|
|
|
Arguments: |
446
|
|
|
item: A scientific name, common name, or mnemonic |
447
|
|
|
""" |
448
|
|
|
taxa = self.get_by_name(item) |
449
|
|
|
ids = ",".join([str(t.id) for t in taxa]) |
450
|
|
|
if len(taxa) > 1: |
451
|
|
|
logger.warning(f"Got multiple results for {item}: {ids}") |
452
|
|
|
elif len(taxa) == 0: |
453
|
|
|
return None |
454
|
|
|
return next(iter(taxa)) |
455
|
|
|
|
456
|
|
|
def get_by_name(self, item: str) -> FrozenSet[Taxon]: |
457
|
|
|
""" |
458
|
|
|
Gets all taxa that match a scientific name. |
459
|
|
|
""" |
460
|
|
|
if isinstance(item, Taxon): |
461
|
|
|
item = item.scientific_name |
462
|
|
|
return self._by_name.get(item, frozenset(set())) |
463
|
|
|
|
464
|
|
|
def get_all_by_id_or_name(self, items: Iterable[Union[int, str, Taxon]]) -> FrozenSet[Taxon]: |
465
|
|
|
""" |
466
|
|
|
Gets all taxa that match any number of IDs or names. |
467
|
|
|
""" |
468
|
|
|
matching = [] |
469
|
|
|
for item in items: |
470
|
|
|
matching += self.get_by_id_or_name(item) |
471
|
|
|
# finally de-duplicates (making this fn useful) |
472
|
|
|
return frozenset(matching) |
473
|
|
|
|
474
|
|
|
def get_by_id_or_name(self, item: Union[int, str, Taxon]) -> FrozenSet[Taxon]: |
475
|
|
|
""" |
476
|
|
|
Gets all taxa that match an ID or name. |
477
|
|
|
""" |
478
|
|
|
if isinstance(item, Taxon): |
479
|
|
|
item = item.id |
480
|
|
|
if isinstance(item, int): |
|
|
|
|
481
|
|
|
taxon = self._by_id.get(item) |
482
|
|
|
return frozenset([]) if taxon is None else frozenset([taxon]) |
483
|
|
|
elif isinstance(item, str): |
484
|
|
|
return self._by_name.get(item, frozenset(set())) |
485
|
|
|
else: |
486
|
|
|
raise XTypeError(f"Unknown type {type(item)} of {item}") |
|
|
|
|
487
|
|
|
|
488
|
|
|
def req(self, item: int) -> Taxon: |
489
|
|
|
""" |
490
|
|
|
Gets a single taxon by its ID. |
491
|
|
|
Raises an error if it is not found. |
492
|
|
|
""" |
493
|
|
|
if isinstance(item, Taxon): |
494
|
|
|
item = item.id |
495
|
|
|
return self[item] |
496
|
|
|
|
497
|
|
|
def get(self, item: Union[int, Taxon]) -> Optional[Taxon]: |
498
|
|
|
""" |
499
|
|
|
Corresponds to ``dict.get``. |
500
|
|
|
|
501
|
|
|
Args: |
502
|
|
|
item: The scientific name or UniProt ID |
503
|
|
|
|
504
|
|
|
Returns: |
505
|
|
|
The taxon, or None if it was not found |
506
|
|
|
""" |
507
|
|
|
if isinstance(item, Taxon): |
508
|
|
|
item = item.id |
509
|
|
|
if isinstance(item, int): |
|
|
|
|
510
|
|
|
return self._by_id.get(item) |
511
|
|
|
else: |
512
|
|
|
raise XTypeError(f"Type {type(item)} of {item} not applicable") |
|
|
|
|
513
|
|
|
|
514
|
|
|
def __getitem__(self, item: int) -> Taxon: |
515
|
|
|
""" |
516
|
|
|
Corresponds to ``dict[_]``. |
517
|
|
|
|
518
|
|
|
Args: |
519
|
|
|
item: The UniProt ID |
520
|
|
|
|
521
|
|
|
Returns: |
522
|
|
|
The taxon |
523
|
|
|
|
524
|
|
|
Raises: |
525
|
|
|
KeyError: If the taxon was not found |
526
|
|
|
""" |
527
|
|
|
got = self.get(item) |
528
|
|
|
if got is None: |
529
|
|
|
raise LookupFailedError(f"{item} not found in {self}") |
530
|
|
|
return got |
531
|
|
|
|
532
|
|
|
def contains(self, item: Union[Taxon, int, str]): |
|
|
|
|
533
|
|
|
return self.get(item) is not None |
534
|
|
|
|
535
|
|
|
def n_taxa(self) -> int: |
|
|
|
|
536
|
|
|
return len(self._by_id) |
537
|
|
|
|
538
|
|
|
def __contains__(self, item: Union[Taxon, int, str]): |
539
|
|
|
return self.get(item) is not None |
540
|
|
|
|
541
|
|
|
def __len__(self) -> int: |
542
|
|
|
return len(self._by_id) |
543
|
|
|
|
544
|
|
|
def __str__(self) -> str: |
545
|
|
|
return repr(self) |
546
|
|
|
|
547
|
|
|
def __repr__(self) -> str: |
548
|
|
|
roots = ", ".join(r.scientific_name for r in self.roots) |
549
|
|
|
return f"{self.__class__.__name__}(n={len(self._by_id)} (roots={roots}) @ {hex(id(self))})" |
550
|
|
|
|
551
|
|
|
@classmethod |
552
|
|
|
def _build_by_name(cls, tax: Iterable[Taxon]) -> Mapping[str, FrozenSet[Taxon]]: |
553
|
|
|
by_name = defaultdict(set) |
554
|
|
|
# put these in the right order |
555
|
|
|
# so that we favor mnemonic, then scientific name, then common name |
556
|
|
|
for t in tax: |
|
|
|
|
557
|
|
|
if t.mnemonic is not None: |
558
|
|
|
by_name[t.mnemonic].add(t) |
559
|
|
|
for t in tax: |
|
|
|
|
560
|
|
|
by_name[t.scientific_name].add(t) |
561
|
|
|
for t in tax: |
|
|
|
|
562
|
|
|
if t.common_name is not None: |
563
|
|
|
by_name[t.common_name].add(t) |
564
|
|
|
# NOTE: lower-casing the keys for lookup |
565
|
|
|
return {k.lower(): frozenset(v) for k, v in by_name.items()} |
566
|
|
|
|
567
|
|
|
|
568
|
|
|
__all__ = ["Taxon", "Taxonomy", "TaxonomyDf", "KnownTaxa"] |
569
|
|
|
|