{{ message }}
forked from biopython/biopython
-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathLoader.py
More file actions
1260 lines (1103 loc) · 50.4 KB
/
Copy pathLoader.py
File metadata and controls
1260 lines (1103 loc) · 50.4 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
# Copyright 2002 by Andrew Dalke. All rights reserved.
# Revisions 2007-2016 copyright by Peter Cock. All rights reserved.
# Revisions 2008 copyright by Cymon J. Cox. All rights reserved.
#
# This file is part of the Biopython distribution and governed by your
# choice of the "Biopython License Agreement" or the "BSD 3-Clause License".
# Please see the LICENSE file that should have been included as part of this
# package.
#
# Note that BioSQL (including the database schema and scripts) is
# available and licensed separately. Please consult www.biosql.org
"""Load biopython objects into a BioSQL database for persistent storage.
This code makes it possible to store biopython objects in a relational
database and then retrieve them back. You shouldn't use any of the
classes in this module directly. Rather, call the load() method on
a database object.
"""
# standard modules
from time import gmtime
from time import strftime
from Bio import Entrez
from Bio.Seq import UndefinedSequenceError
from Bio.SeqFeature import UnknownPosition
# biopython
from Bio.SeqUtils.CheckSum import crc64
class DatabaseLoader:
"""Object used to load SeqRecord objects into a BioSQL database."""
def __init__(self, adaptor, dbid, fetch_NCBI_taxonomy=False):
"""Initialize with connection information for the database.
Creating a DatabaseLoader object is normally handled via the
BioSeqDatabase DBServer object, for example::
from BioSQL import BioSeqDatabase
server = BioSeqDatabase.open_database(driver="MySQLdb",
user="gbrowse",
passwd="biosql",
host="localhost",
db="test_biosql")
try:
db = server["test"]
except KeyError:
db = server.new_database("test",
description="For testing GBrowse")
"""
self.adaptor = adaptor
self.dbid = dbid
self.fetch_NCBI_taxonomy = fetch_NCBI_taxonomy
def load_seqrecord(self, record):
"""Load a Biopython SeqRecord into the database."""
bioentry_id = self._load_bioentry_table(record)
self._load_bioentry_date(record, bioentry_id)
self._load_biosequence(record, bioentry_id)
self._load_comment(record, bioentry_id)
self._load_dbxrefs(record, bioentry_id)
references = record.annotations.get("references", ())
for reference, rank in zip(references, list(range(len(references)))):
self._load_reference(reference, rank, bioentry_id)
self._load_annotations(record, bioentry_id)
for seq_feature_num in range(len(record.features)):
seq_feature = record.features[seq_feature_num]
self._load_seqfeature(seq_feature, seq_feature_num, bioentry_id)
def _get_ontology_id(self, name, definition=None):
"""Return identifier for the named ontology (PRIVATE).
This looks through the onotology table for a the given entry name.
If it is not found, a row is added for this ontology (using the
definition if supplied). In either case, the id corresponding to
the provided name is returned, so that you can reference it in
another table.
"""
oids = self.adaptor.execute_and_fetch_col0(
"SELECT ontology_id FROM ontology WHERE name = %s", (name,)
)
if oids:
return oids[0]
self.adaptor.execute(
"INSERT INTO ontology(name, definition) VALUES (%s, %s)", (name, definition)
)
return self.adaptor.last_id("ontology")
def _get_term_id(self, name, ontology_id=None, definition=None, identifier=None):
"""Get the id that corresponds to a term (PRIVATE).
This looks through the term table for a the given term. If it
is not found, a new id corresponding to this term is created.
In either case, the id corresponding to that term is returned, so
that you can reference it in another table.
The ontology_id should be used to disambiguate the term.
"""
# try to get the term id
sql = "SELECT term_id FROM term WHERE name = %s"
fields = [name]
if ontology_id:
sql += " AND ontology_id = %s"
fields.append(ontology_id)
id_results = self.adaptor.execute_and_fetchall(sql, fields)
# something is wrong
if len(id_results) > 1:
raise ValueError(f"Multiple term ids for {name}: {id_results!r}")
elif len(id_results) == 1:
return id_results[0][0]
else:
sql = (
"INSERT INTO term (name, definition,"
" identifier, ontology_id)"
" VALUES (%s, %s, %s, %s)"
)
self.adaptor.execute(sql, (name, definition, identifier, ontology_id))
return self.adaptor.last_id("term")
def _add_dbxref(self, dbname, accession, version):
"""Insert a dbxref and return its id (PRIVATE)."""
self.adaptor.execute(
"INSERT INTO dbxref(dbname, accession, version) VALUES (%s, %s, %s)",
(dbname, accession, version),
)
return self.adaptor.last_id("dbxref")
def _get_taxon_id(self, record):
"""Get the taxon id for this record (PRIVATE).
Arguments:
- record - a SeqRecord object
This searches the taxon/taxon_name tables using the
NCBI taxon ID, scientific name and common name to find
the matching taxon table entry's id.
If the species isn't in the taxon table, and we have at
least the NCBI taxon ID, scientific name or common name,
at least a minimal stub entry is created in the table.
Returns the taxon id (database key for the taxon table,
not an NCBI taxon ID), or None if the taxonomy information
is missing.
See also the BioSQL script load_ncbi_taxonomy.pl which
will populate and update the taxon/taxon_name tables
with the latest information from the NCBI.
"""
# To find the NCBI taxid, first check for a top level annotation
ncbi_taxon_id = None
if "ncbi_taxid" in record.annotations:
# Could be a list of IDs.
if isinstance(record.annotations["ncbi_taxid"], list):
if len(record.annotations["ncbi_taxid"]) == 1:
ncbi_taxon_id = record.annotations["ncbi_taxid"][0]
else:
ncbi_taxon_id = record.annotations["ncbi_taxid"]
if not ncbi_taxon_id:
# Secondly, look for a source feature
for f in record.features:
if f.type == "source":
quals = getattr(f, "qualifiers", {})
if "db_xref" in quals:
for db_xref in f.qualifiers["db_xref"]:
if db_xref.startswith("taxon:"):
ncbi_taxon_id = int(db_xref[6:])
break
if ncbi_taxon_id:
break
try:
scientific_name = record.annotations["organism"][:255]
except KeyError:
scientific_name = None
try:
common_name = record.annotations["source"][:255]
except KeyError:
common_name = None
# Note: The maximum length for taxon names in the schema is 255.
# Cropping it now should help in getting a match when searching,
# and avoids an error if we try and add these to the database.
if ncbi_taxon_id:
# Good, we have the NCBI taxon to go on - this is unambiguous :)
# Note that the scientific name and common name will only be
# used if we have to record a stub entry.
return self._get_taxon_id_from_ncbi_taxon_id(
ncbi_taxon_id, scientific_name, common_name
)
if not common_name and not scientific_name:
# Nothing to go on... and there is no point adding
# a new entry to the database. We'll just leave this
# sequence's taxon as a NULL in the database.
return None
# Next, we'll try to find a match based on the species name
# (stored in GenBank files as the organism and/or the source).
if scientific_name:
taxa = self.adaptor.execute_and_fetch_col0(
"SELECT taxon_id FROM taxon_name"
" WHERE name_class = 'scientific name' AND name = %s",
(scientific_name,),
)
if taxa:
# Good, mapped the scientific name to a taxon table entry
return taxa[0]
# Last chance...
if common_name:
taxa = self.adaptor.execute_and_fetch_col0(
"SELECT DISTINCT taxon_id FROM taxon_name WHERE name = %s",
(common_name,),
)
# Its natural that several distinct taxa will have the same common
# name - in which case we can't resolve the taxon uniquely.
if len(taxa) > 1:
raise ValueError(
"Taxa: %d species have name %r" % (len(taxa), common_name)
)
if taxa:
# Good, mapped the common name to a taxon table entry
return taxa[0]
# At this point, as far as we can tell, this species isn't
# in the taxon table already. So we'll have to add it.
# We don't have an NCBI taxonomy ID, so if we do record just
# a stub entry, there is no simple way to fix this later.
#
# TODO - Should we try searching the NCBI taxonomy using the
# species name?
#
# OK, let's try inserting the species.
# Chances are we don't have enough information ...
# Furthermore, it won't be in the hierarchy.
lineage = []
for c in record.annotations.get("taxonomy", []):
lineage.append([None, None, c])
if lineage:
lineage[-1][1] = "genus"
lineage.append([None, "species", record.annotations["organism"]])
# XXX do we have them?
if "subspecies" in record.annotations:
lineage.append([None, "subspecies", record.annotations["subspecies"]])
if "variant" in record.annotations:
lineage.append([None, "varietas", record.annotations["variant"]])
lineage[-1][0] = ncbi_taxon_id
left_value = self.adaptor.execute_one("SELECT MAX(left_value) FROM taxon")[0]
if not left_value:
left_value = 0
left_value += 1
# XXX -- Brad: Fixing this for now in an ugly way because
# I am getting overlaps for right_values. I need to dig into this
# more to actually understand how it works. I'm not sure it is
# actually working right anyhow.
right_start_value = self.adaptor.execute_one(
"SELECT MAX(right_value) FROM taxon"
)[0]
if not right_start_value:
right_start_value = 0
right_value = right_start_value + 2 * len(lineage) - 1
parent_taxon_id = None
for taxon in lineage:
self.adaptor.execute(
"INSERT INTO taxon(parent_taxon_id, ncbi_taxon_id, node_rank,"
" left_value, right_value)"
" VALUES (%s, %s, %s, %s, %s)",
(parent_taxon_id, taxon[0], taxon[1], left_value, right_value),
)
taxon_id = self.adaptor.last_id("taxon")
self.adaptor.execute(
"INSERT INTO taxon_name(taxon_id, name, name_class)"
"VALUES (%s, %s, 'scientific name')",
(taxon_id, taxon[2][:255]),
)
# Note the name field is limited to 255, some SwissProt files
# have a multi-species name which can be longer. So truncate this.
left_value += 1
right_value -= 1
parent_taxon_id = taxon_id
if common_name:
self.adaptor.execute(
"INSERT INTO taxon_name(taxon_id, name, name_class)"
"VALUES (%s, %s, 'common name')",
(taxon_id, common_name),
)
return taxon_id
def _fix_name_class(self, entrez_name):
"""Map Entrez name terms to those used in taxdump (PRIVATE).
We need to make this conversion to match the taxon_name.name_class
values used by the BioSQL load_ncbi_taxonomy.pl script.
e.g.::
"ScientificName" -> "scientific name",
"EquivalentName" -> "equivalent name",
"Synonym" -> "synonym",
"""
# Add any special cases here:
#
# known = {}
# try:
# return known[entrez_name]
# except KeyError:
# pass
# Try automatically by adding spaces before each capital
def add_space(letter):
"""Add a space before a capital letter."""
if letter.isupper():
return " " + letter.lower()
else:
return letter
answer = "".join(add_space(letter) for letter in entrez_name).strip()
if answer != answer.lower():
raise ValueError(
f"Expected processed entrez_name, '{answer}' to only have lower case letters."
)
return answer
def _update_left_right_taxon_values(self, left_value):
"""Update the left and right taxon values in the table (PRIVATE)."""
if not left_value:
return
# Due to the UNIQUE constraint on the left and right values in the taxon
# table we cannot simply update them through an SQL statement as we risk
# colliding values. Instead we must select all of the rows that we want to
# update, modify the values in python and then update the rows
# self.adaptor.execute("UPDATE taxon SET right_value = right_value + 2 "
# "WHERE right_value >= %s", (left_value,))
# self.adaptor.execute("UPDATE taxon SET left_value = left_value + 2 "
# "WHERE left_value > %s", (left_value,))
rows = self.adaptor.execute_and_fetchall(
"SELECT left_value, right_value, taxon_id FROM taxon "
"WHERE right_value >= %s or left_value > %s",
(left_value, left_value),
)
right_rows = []
left_rows = []
for row in rows:
new_right = row[1]
new_left = row[0]
if new_right >= left_value:
new_right += 2
if new_left > left_value:
new_left += 2
right_rows.append((new_right, row[2]))
left_rows.append((new_left, row[2]))
# sort the rows based on the value from largest to smallest
# should ensure no overlaps
right_rows = sorted(right_rows, key=lambda x: x[0], reverse=True)
left_rows = sorted(left_rows, key=lambda x: x[0], reverse=True)
self.adaptor.executemany(
"UPDATE taxon SET left_value = %s WHERE taxon_id = %s", left_rows
)
self.adaptor.executemany(
"UPDATE taxon SET right_value = %s WHERE taxon_id = %s", right_rows
)
def _get_taxon_id_from_ncbi_taxon_id(
self, ncbi_taxon_id, scientific_name=None, common_name=None
):
"""Get the taxon id for record from NCBI taxon ID (PRIVATE).
Arguments:
- ncbi_taxon_id - string containing an NCBI taxon id
- scientific_name - string, used if a stub entry is recorded
- common_name - string, used if a stub entry is recorded
This searches the taxon table using ONLY the NCBI taxon ID
to find the matching taxon table entry's ID (database key).
If the species isn't in the taxon table, and the fetch_NCBI_taxonomy
flag is true, Biopython will attempt to go online using Bio.Entrez
to fetch the official NCBI lineage, recursing up the tree until an
existing entry is found in the database or the full lineage has been
fetched.
Otherwise the NCBI taxon ID, scientific name and common name are
recorded as a minimal stub entry in the taxon and taxon_name tables.
Any partial information about the lineage from the SeqRecord is NOT
recorded. This should mean that (re)running the BioSQL script
load_ncbi_taxonomy.pl can fill in the taxonomy lineage.
Returns the taxon id (database key for the taxon table, not
an NCBI taxon ID).
"""
if not ncbi_taxon_id:
raise ValueError("Expected a non-empty value for ncbi_taxon_id.")
taxon_id = self.adaptor.execute_and_fetch_col0(
"SELECT taxon_id FROM taxon WHERE ncbi_taxon_id = %s", (int(ncbi_taxon_id),)
)
if taxon_id:
# Good, we have mapped the NCBI taxid to a taxon table entry
return taxon_id[0]
# At this point, as far as we can tell, this species isn't
# in the taxon table already. So we'll have to add it.
parent_taxon_id = None
rank = "species"
genetic_code = None
mito_genetic_code = None
parent_left_value = None
parent_right_value = None
left_value = None
right_value = None
species_names = []
if scientific_name:
species_names.append(("scientific name", scientific_name))
if common_name:
species_names.append(("common name", common_name))
if self.fetch_NCBI_taxonomy:
# Go online to get the parent taxon ID!
handle = Entrez.efetch(db="taxonomy", id=ncbi_taxon_id, retmode="XML")
taxonomic_record = Entrez.read(handle)
if len(taxonomic_record) == 1:
if taxonomic_record[0]["TaxId"] != str(ncbi_taxon_id):
raise ValueError(
f"ncbi_taxon_id different from parent taxon id. {ncbi_taxon_id} versus {taxonomic_record[0]['TaxId']}"
)
(
parent_taxon_id,
parent_left_value,
parent_right_value,
) = self._get_taxon_id_from_ncbi_lineage(
taxonomic_record[0]["LineageEx"]
)
left_value = parent_right_value
right_value = parent_right_value + 1
rank = str(taxonomic_record[0]["Rank"])
genetic_code = int(taxonomic_record[0]["GeneticCode"]["GCId"])
mito_genetic_code = int(taxonomic_record[0]["MitoGeneticCode"]["MGCId"])
species_names = [
("scientific name", str(taxonomic_record[0]["ScientificName"]))
]
try:
for name_class, names in taxonomic_record[0]["OtherNames"].items():
name_class = self._fix_name_class(name_class)
if not isinstance(names, list):
# The Entrez parser seems to return single entry
# lists as just a string which is annoying.
names = [names]
for name in names:
# Want to ignore complex things like ClassCDE
# entries
if isinstance(name, str):
species_names.append((name_class, name))
except KeyError:
# OtherNames isn't always present,
# e.g. NCBI taxon 41205, Bromheadia finlaysoniana
pass
else:
pass
# If we are not allowed to go online, we will record the bare minimum;
# as long as the NCBI taxon id is present, then (re)running
# load_ncbi_taxonomy.pl should fill in the taxonomomy lineage
# (and update the species names).
#
# I am NOT going to try and record the lineage, even if it
# is in the record annotation as a list of names, as we won't
# know the NCBI taxon IDs for these parent nodes.
self._update_left_right_taxon_values(left_value)
self.adaptor.execute(
"INSERT INTO taxon(parent_taxon_id, ncbi_taxon_id, node_rank,"
" genetic_code, mito_genetic_code, left_value, right_value)"
" VALUES (%s, %s, %s, %s, %s, %s, %s)",
(
parent_taxon_id,
ncbi_taxon_id,
rank,
genetic_code,
mito_genetic_code,
left_value,
right_value,
),
)
taxon_id = self.adaptor.last_id("taxon")
# Record the scientific name, common name, etc
for name_class, name in species_names:
self.adaptor.execute(
"INSERT INTO taxon_name(taxon_id, name, name_class)"
" VALUES (%s, %s, %s)",
(taxon_id, name[:255], name_class),
)
return taxon_id
def _get_taxon_id_from_ncbi_lineage(self, taxonomic_lineage):
"""Recursive method to get taxon ID from NCBI lineage (PRIVATE).
Arguments:
- taxonomic_lineage - list of taxonomy dictionaries from Bio.Entrez
First dictionary in list is the taxonomy root, highest would be
the species. Each dictionary includes:
- TaxID (string, NCBI taxon id)
- Rank (string, e.g. "species", "genus", ..., "phylum", ...)
- ScientificName (string)
(and that is all at the time of writing)
This method will record all the lineage given, returning the taxon id
(database key, not NCBI taxon id) of the final entry (the species).
"""
ncbi_taxon_id = int(taxonomic_lineage[-1]["TaxId"])
left_value = None
right_value = None
parent_left_value = None
parent_right_value = None
# Is this in the database already? Check the taxon table...
rows = self.adaptor.execute_and_fetchall(
"SELECT taxon_id, left_value, right_value FROM taxon"
" WHERE ncbi_taxon_id=%s" % ncbi_taxon_id
)
if rows:
# we could verify that the Scientific Name etc in the database
# is the same and update it or print a warning if not...
if len(rows) != 1:
raise ValueError(f"Expected 1 response, got {len(rows)}")
return rows[0]
# We have to record this.
if len(taxonomic_lineage) > 1:
# Use recursion to find out the taxon id (database key) of the
# parent.
(
parent_taxon_id,
parent_left_value,
parent_right_value,
) = self._get_taxon_id_from_ncbi_lineage(taxonomic_lineage[:-1])
left_value = parent_right_value
right_value = parent_right_value + 1
if not isinstance(parent_taxon_id, int):
raise ValueError(
f"Expected parent_taxon_id to be an int, got {parent_taxon_id}"
)
else:
# we have reached the top of the lineage but no current taxonomy
# id has been found
parent_taxon_id = None
left_value = self.adaptor.execute_one("SELECT MAX(left_value) FROM taxon")[
0
]
if not left_value:
left_value = 0
right_value = left_value + 1
self._update_left_right_taxon_values(left_value)
# INSERT new taxon
rank = str(taxonomic_lineage[-1].get("Rank"))
self.adaptor.execute(
"INSERT INTO taxon(ncbi_taxon_id, parent_taxon_id, node_rank, "
"left_value, right_value) VALUES (%s, %s, %s, %s, %s)",
(ncbi_taxon_id, parent_taxon_id, rank, left_value, right_value),
)
taxon_id = self.adaptor.last_id("taxon")
# assert isinstance(taxon_id, int), repr(taxon_id)
# ... and its name in taxon_name
scientific_name = taxonomic_lineage[-1].get("ScientificName")
if scientific_name:
self.adaptor.execute(
"INSERT INTO taxon_name(taxon_id, name, name_class) "
"VALUES (%s, %s, 'scientific name')",
(taxon_id, scientific_name[:255]),
)
return taxon_id, left_value, right_value
def _load_bioentry_table(self, record):
"""Fill the bioentry table with sequence information (PRIVATE).
Arguments:
- record - SeqRecord object to add to the database.
"""
# get the pertinent info and insert it
if record.id.count(".") == 1: # try to get a version from the id
# This assumes the string is something like "XXXXXXXX.123"
accession, version = record.id.split(".")
try:
version = int(version)
except ValueError:
accession = record.id
version = 0
else: # otherwise just use a version of 0
accession = record.id
version = 0
if (
"accessions" in record.annotations
and isinstance(record.annotations["accessions"], list)
and record.annotations["accessions"]
):
# Take the first accession (one if there is more than one)
accession = record.annotations["accessions"][0]
# Find the taxon id (this is not just the NCBI Taxon ID)
# NOTE - If the species isn't defined in the taxon table,
# a new minimal entry is created.
taxon_id = self._get_taxon_id(record)
if "gi" in record.annotations:
identifier = record.annotations["gi"]
else:
identifier = record.id
# Allow description and division to default to NULL as in BioPerl.
description = getattr(record, "description", None)
division = record.annotations.get("data_file_division")
sql = """
INSERT INTO bioentry (
biodatabase_id,
taxon_id,
name,
accession,
identifier,
division,
description,
version)
VALUES (
%s,
%s,
%s,
%s,
%s,
%s,
%s,
%s)"""
# print(self.dbid, taxon_id, record.name, accession, identifier, \
# division, description, version)
self.adaptor.execute(
sql,
(
self.dbid,
taxon_id,
record.name,
accession,
identifier,
division,
description,
version,
),
)
# now retrieve the id for the bioentry
return self.adaptor.last_id("bioentry")
def _load_bioentry_date(self, record, bioentry_id):
"""Add the effective date of the entry into the database (PRIVATE).
record - a SeqRecord object with an annotated date
bioentry_id - corresponding database identifier
"""
# dates are GenBank style, like:
# 14-SEP-2000
date = record.annotations.get("date", strftime("%d-%b-%Y", gmtime()).upper())
if isinstance(date, list):
date = date[0]
annotation_tags_id = self._get_ontology_id("Annotation Tags")
date_id = self._get_term_id("date_changed", annotation_tags_id)
sql = (
"INSERT INTO bioentry_qualifier_value"
' (bioentry_id, term_id, value, "rank")'
" VALUES (%s, %s, %s, 1)"
)
self.adaptor.execute(sql, (bioentry_id, date_id, date))
def _load_biosequence(self, record, bioentry_id):
"""Record SeqRecord's sequence and alphabet in DB (PRIVATE).
Arguments:
- record - a SeqRecord object with a seq property
- bioentry_id - corresponding database identifier
"""
if record.seq is None:
# The biosequence table entry is optional, so if we haven't
# got a sequence, we don't need to write to the table.
return
molecule_type = record.annotations.get("molecule_type", "")
if "DNA" in molecule_type:
alphabet = "dna"
elif "RNA" in molecule_type:
alphabet = "rna"
elif "protein" in molecule_type:
alphabet = "protein"
else:
alphabet = "unknown"
try:
seq_str = str(record.seq)
except UndefinedSequenceError:
seq_str = None
sql = (
"INSERT INTO biosequence (bioentry_id, version, "
"length, seq, alphabet) "
"VALUES (%s, 0, %s, %s, %s)"
)
self.adaptor.execute(sql, (bioentry_id, len(record.seq), seq_str, alphabet))
def _load_comment(self, record, bioentry_id):
"""Record a SeqRecord's annotated comment in the database (PRIVATE).
Arguments:
- record - a SeqRecord object with an annotated comment
- bioentry_id - corresponding database identifier
"""
comments = record.annotations.get("comment")
if not comments:
return
if not isinstance(comments, list):
# It should be a string then...
comments = [comments]
for index, comment in enumerate(comments):
comment = comment.replace("\n", " ")
# TODO - Store each line as a separate entry? This would preserve
# the newlines, but we should check BioPerl etc to be consistent.
sql = (
'INSERT INTO comment (bioentry_id, comment_text, "rank")'
" VALUES (%s, %s, %s)"
)
self.adaptor.execute(sql, (bioentry_id, comment, index + 1))
def _load_annotations(self, record, bioentry_id):
"""Record a SeqRecord's misc annotations in the database (PRIVATE).
The annotation strings are recorded in the bioentry_qualifier_value
table, except for special cases like the reference, comment and
taxonomy which are handled with their own tables.
Arguments:
- record - a SeqRecord object with an annotations dictionary
- bioentry_id - corresponding database identifier
"""
mono_sql = (
"INSERT INTO bioentry_qualifier_value"
"(bioentry_id, term_id, value)"
" VALUES (%s, %s, %s)"
)
many_sql = (
"INSERT INTO bioentry_qualifier_value"
'(bioentry_id, term_id, value, "rank")'
" VALUES (%s, %s, %s, %s)"
)
tag_ontology_id = self._get_ontology_id("Annotation Tags")
for key, value in record.annotations.items():
if key in ["molecule_type", "references", "comment", "ncbi_taxid", "date"]:
# Handled separately
continue
term_id = self._get_term_id(key, ontology_id=tag_ontology_id)
if isinstance(value, (list, tuple)):
rank = 0
for entry in value:
if isinstance(entry, (str, int)):
# Easy case
rank += 1
self.adaptor.execute(
many_sql, (bioentry_id, term_id, str(entry), rank)
)
else:
pass
elif isinstance(value, (str, int)):
# Have a simple single entry, leave rank as the DB default
self.adaptor.execute(mono_sql, (bioentry_id, term_id, str(value)))
else:
pass
# print("Ignoring annotation '%s' entry of type '%s'" \
# % (key, type(value)))
def _load_reference(self, reference, rank, bioentry_id):
"""Record SeqRecord's annotated references in the database (PRIVATE).
Arguments:
- record - a SeqRecord object with annotated references
- bioentry_id - corresponding database identifier
"""
refs = None
if reference.medline_id:
refs = self.adaptor.execute_and_fetch_col0(
"SELECT reference_id"
" FROM reference JOIN dbxref USING (dbxref_id)"
" WHERE dbname = 'MEDLINE' AND accession = %s",
(reference.medline_id,),
)
if not refs and reference.pubmed_id:
refs = self.adaptor.execute_and_fetch_col0(
"SELECT reference_id"
" FROM reference JOIN dbxref USING (dbxref_id)"
" WHERE dbname = 'PUBMED' AND accession = %s",
(reference.pubmed_id,),
)
if not refs:
s = []
for f in reference.authors, reference.title, reference.journal:
s.append(f or "<undef>")
crc = crc64("".join(s))
refs = self.adaptor.execute_and_fetch_col0(
"SELECT reference_id FROM reference WHERE crc = %s", (crc,)
)
if not refs:
if reference.medline_id:
dbxref_id = self._add_dbxref("MEDLINE", reference.medline_id, 0)
elif reference.pubmed_id:
dbxref_id = self._add_dbxref("PUBMED", reference.pubmed_id, 0)
else:
dbxref_id = None
authors = reference.authors or None
title = reference.title or None
# The location/journal field cannot be Null, so default
# to an empty string rather than None:
journal = reference.journal or ""
self.adaptor.execute(
"INSERT INTO reference (dbxref_id, location,"
" title, authors, crc)"
" VALUES (%s, %s, %s, %s, %s)",
(dbxref_id, journal, title, authors, crc),
)
reference_id = self.adaptor.last_id("reference")
else:
reference_id = refs[0]
if reference.location:
start = 1 + int(str(reference.location[0].start))
end = int(str(reference.location[0].end))
else:
start = None
end = None
sql = (
"INSERT INTO bioentry_reference (bioentry_id, reference_id,"
' start_pos, end_pos, "rank") VALUES (%s, %s, %s, %s, %s)'
)
self.adaptor.execute(sql, (bioentry_id, reference_id, start, end, rank + 1))
def _load_seqfeature(self, feature, feature_rank, bioentry_id):
"""Load a biopython SeqFeature into the database (PRIVATE)."""
# records loaded from a gff file using BCBio.GFF will contain value
# of 2nd column of the gff as a feature qualifier. The BioSQL wiki
# suggests that the source should not go in with the other feature
# mappings but instead be put in the term table
# (http://www.biosql.org/wiki/Annotation_Mapping)
try:
source = feature.qualifiers["source"]
if isinstance(source, list):
source = source[0]
seqfeature_id = self._load_seqfeature_basic(
feature.type, feature_rank, bioentry_id, source=source
)
except KeyError:
seqfeature_id = self._load_seqfeature_basic(
feature.type, feature_rank, bioentry_id
)
self._load_seqfeature_locations(feature, seqfeature_id)
self._load_seqfeature_qualifiers(feature.qualifiers, seqfeature_id)
def _load_seqfeature_basic(
self, feature_type, feature_rank, bioentry_id, source="EMBL/GenBank/SwissProt"
):
"""Load the first tables of a seqfeature and returns the id (PRIVATE).
This loads the "key" of the seqfeature (ie. CDS, gene) and
the basic seqfeature table itself.
"""
ontology_id = self._get_ontology_id("SeqFeature Keys")
seqfeature_key_id = self._get_term_id(feature_type, ontology_id=ontology_id)
source_cat_id = self._get_ontology_id("SeqFeature Sources")
source_term_id = self._get_term_id(source, ontology_id=source_cat_id)
sql = (
"INSERT INTO seqfeature (bioentry_id, type_term_id, "
'source_term_id, "rank") VALUES (%s, %s, %s, %s)'
)
self.adaptor.execute(
sql, (bioentry_id, seqfeature_key_id, source_term_id, feature_rank + 1)
)
return self.adaptor.last_id("seqfeature")
def _load_seqfeature_locations(self, feature, seqfeature_id):
"""Load all of the locations for a SeqFeature into tables (PRIVATE).
This adds the locations related to the SeqFeature into the
seqfeature_location table. Fuzzies are not handled right now.
For a simple location, ie (1..2), we have a single table row
with seq_start = 1, seq_end = 2, location_rank = 1.
For split locations, ie (1..2, 3..4, 5..6) we would have three
row tables with::
start = 1, end = 2, rank = 1
start = 3, end = 4, rank = 2
start = 5, end = 6, rank = 3
"""
# TODO - Record an ontology for the locations (using location.term_id)
# which for now as in BioPerl we leave defaulting to NULL.
try:
if feature.location.operator != "join":
# e.g. order locations... we don't record "order" so it
# will become a "join" on reloading. What does BioPerl do?
import warnings
from Bio import BiopythonWarning
warnings.warn(
"%s location operators are not fully supported"
% feature.location_operator,
BiopythonWarning,
)
except AttributeError:
pass
# This will be a list of length one for a SimpleLocation:
parts = feature.location.parts
if parts and {loc.strand for loc in parts} == {-1}:
# To mimic prior behaviour of Biopython+BioSQL, reverse order
parts = parts[::-1]
# TODO - Check what BioPerl does; see also BioSeq.py code
for rank, loc in enumerate(parts):
self._insert_location(loc, rank + 1, seqfeature_id)
def _insert_location(self, location, rank, seqfeature_id):
"""Add SeqFeature location to seqfeature_location table (PRIVATE).
TODO - Add location operator to location_qualifier_value?
"""
# convert biopython locations to the 1-based location system
# used in bioSQL
# XXX This could also handle fuzzies
try:
start = int(location.start) + 1
except TypeError:
# Handle SwissProt unknown position (?)
if isinstance(location.start, UnknownPosition):
start = None
else:
raise
try:
end = int(location.end)
except TypeError:
# Handle SwissProt unknown position (?)
if isinstance(location.end, UnknownPosition):
end = None
else:
raise
# Biopython uses None when we don't know strand information but
# BioSQL requires something (non null) and sets this as zero
# So we'll use the strand or 0 if Biopython spits out None
strand = location.strand or 0
# TODO - Record an ontology term for the location (location.term_id)
# which for now like BioPerl we'll leave as NULL.
# This might allow us to record "between" positions properly, but I
# don't really see how it could work for before/after fuzzy positions
loc_term_id = None
if location.ref:
You can’t perform that action at this time.