Skip to content

Commit 0e14b51

Browse files
fescudiefescudie
committed
Manage unstandard case.
1 parent 213dc09 commit 0e14b51

1 file changed

Lines changed: 34 additions & 44 deletions

File tree

‎bin/buildKnownBNDDb.py‎

Lines changed: 34 additions & 44 deletions
Original file line numberDiff line numberDiff line change
@@ -51,8 +51,8 @@ def aliasesBySymbolsFromEnsembl(in_aliases):
5151
aliases_by_symbol = {}
5252
with HashedSVIO(in_aliases) as reader:
5353
for record in reader:
54-
name = record["Gene name"]
55-
alias = record["Gene Synonym"]
54+
name = record["Gene name"].upper()
55+
alias = record["Gene Synonym"].upper()
5656
if name not in aliases_by_symbol:
5757
aliases_by_symbol[name] = [name, alias]
5858
else:
@@ -76,8 +76,8 @@ def aliasesBySymbolsFromNCBI(in_aliases):
7676
aliases_by_symbol = {}
7777
with HashedSVIO(in_aliases) as reader:
7878
for record in reader:
79-
name = record["Symbol"]
80-
aliases = record["Synonyms"].split("|")
79+
name = record["Symbol"].upper()
80+
aliases = [elt.upper() for elt in record["Synonyms"].split("|")]
8181
if name not in aliases_by_symbol:
8282
aliases_by_symbol[name] = [name] + aliases
8383
else:
@@ -96,16 +96,18 @@ def annotSymbols(in_annotations):
9696
9797
:param in_annotations: Path to the genes annotations file.
9898
:type in_annotations: str
99-
:return: List of genes names used in genes annotations file.
100-
:rtype: set
99+
:return: By upper gene name the gene name used in genes annotations file.
100+
:rtype: dict
101101
"""
102-
annotation_symbols = set()
102+
annotation_symbols = dict()
103103
with GTFIO(in_annotations) as reader:
104104
for record in reader:
105105
if "gene_name" in record.annot:
106-
annotation_symbols.add(record.annot["gene_name"])
106+
name = record.annot["gene_name"]
107+
annotation_symbols[name.upper()] = name
107108
elif "gene" in record.annot:
108-
annotation_symbols.add(record.annot["gene"])
109+
name = record.annot["gene"]
110+
annotation_symbols[name.upper()] = name
109111
return annotation_symbols
110112

111113

@@ -115,27 +117,23 @@ def selectAnnotSymbol(gene_symbol, annotation_symbols, aliases_by_symbol):
115117
116118
:param gene_symbol: Gene name.
117119
:type gene_symbol: str
118-
:param annotation_symbols: List of genes names known in genes annotations file.
119-
:type annotation_symbols: set
120+
:param annotation_symbols: By upper gene name the gene name used in genes annotations file.
121+
:type annotation_symbols: dict
120122
:param aliases_by_symbol: Gene name aliases by symbol.
121123
:type aliases_by_symbol: dict
122124
:return: Alias of the gene symbol used in genes annotations file.
123125
:rtype: str
124126
"""
125127
retained_name = None
126-
if gene_symbol not in aliases_by_symbol:
127-
if gene_symbol.upper() in aliases_by_symbol:
128-
gene_symbol = gene_symbol.upper()
129-
elif gene_symbol.replace("ORF", "orf") in aliases_by_symbol:
130-
gene_symbol = gene_symbol.replace("ORF", "orf")
131-
aliases = [gene_symbol]
132-
if gene_symbol in aliases_by_symbol:
133-
aliases = aliases_by_symbol[gene_symbol]
128+
uc_symbol = gene_symbol.upper()
129+
aliases = [uc_symbol]
130+
if uc_symbol in aliases_by_symbol:
131+
aliases = aliases_by_symbol[uc_symbol]
134132
for curr_name in aliases:
135133
if curr_name in annotation_symbols:
136-
retained_name = curr_name
134+
retained_name = annotation_symbols[curr_name]
137135
if retained_name is None:
138-
raise Exception("The gene with aliases {} cannot be found in genes annotations.".format(aliases))
136+
raise Exception(f"The gene with aliases {aliases} cannot be found in genes annotations.")
139137
return retained_name
140138

141139

@@ -195,7 +193,7 @@ def loadChimerdb(db_path, db_version, fusions_by_partners, aliases_by_symbol, an
195193
:param annotation_symbols: List of genes names known in genes annotations file.
196194
:type annotation_symbols: set
197195
"""
198-
# id Source webSource Fusion_pair H_gene H_chr H_position H_strand T_gene T_chr T_position T_strand Breakpoint_Type Genome_Build_Version PMID Disease Validation Kinase Oncogene Tumor_suppressor Receptor Transcription_Factor ChimerPub ChimerSeq
196+
# id ChimerDB_Type Source webSource Fusion_pair 5Gene_Junction 3Gene_Junction H_gene H_chr H_position H_strand T_gene T_chr T_position T_strand Genomic_breakpoint Exonic_breakpoint Breakpoint_Type Genome_Build_Version PMID Disease Validation Frame Chr_info Kinase Oncogene Tumor_suppressor Receptor Transcription_FactorChimerPub ChimerSeq ChimerSeq+
199197
with HashedSVIO(db_path) as reader:
200198
for record in reader:
201199
up_gene = None
@@ -239,18 +237,14 @@ def loadCosmic(db_path, db_version, fusions_by_partners, aliases_by_symbol, anno
239237
:param annotation_symbols: List of genes names known in genes annotations file.
240238
:type annotation_symbols: set
241239
"""
242-
# Sample ID Sample name Primary site Site subtype 1 Site subtype 2 Site subtype 3 Primary histology Histology subtype 1 Histology subtype 2 Histology subtype 3 Fusion ID Translocation Name 5'_CHROMOSOME 5'_GENOME_START_FROM 5'_GENOME_START_TO 5'_GENOME_STOP_FROM 5'_GENOME_STOP_TO 5'_STRAND 3'_CHROMOSOME 3'_GENOME_START_FROM 3'_GENOME_START_TO 3'_GENOME_STOP_FROM 3'_GENOME_STOP_TO 3'_STRAND Fusion type Pubmed_PMID
240+
# SAMPLE_ID SAMPLE_NAME PRIMARY_SITE SITE_SUBTYPE_1 SITE_SUBTYPE_2 SITE_SUBTYPE_3 PRIMARY_HISTOLOGY HISTOLOGY_SUBTYPE_1 HISTOLOGY_SUBTYPE_2 HISTOLOGY_SUBTYPE_3 FUSION_ID TRANSLOCATION_NAME 5'_CHROMOSOME 5'_STRAND 5'_GENE_ID 5'_GENE_NAME 5'_LAST_OBSERVED_EXON 5'_GENOME_START_FROM 5'_GENOME_START_TO 5'_GENOME_STOP_FROM 5'_GENOME_STOP_TO 3'_CHROMOSOME3'_STRAND 3'_GENE_ID 3'_GENE_NAME 3'_FIRST_OBSERVED_EXON 3'_GENOME_START_FROM 3'_GENOME_START_TO 3'_GENOME_STOP_FROM 3'_GENOME_STOP_TO FUSION_TYPE PUBMED_PMID
243241
with HashedSVIO(db_path) as reader:
244-
reader.titles = [elt.upper().replace(" ", "_") for elt in reader.titles]
245242
for record in reader:
246243
if record["TRANSLOCATION_NAME"] != "":
247244
matches = re.fullmatch(r"ENS.+\((.+)\):.+_ENS.+\((.+)\):.+", record["TRANSLOCATION_NAME"]) # ENST00000324093.4(PLXND1):r.1_2864_ENST00000393238.3(TMCC1):r.918_5992
248245
if matches is None:
249246
log.warning(
250-
"Error to parse gene names {} from cosmic (PMID: {}).".format(
251-
record["TRANSLOCATION_NAME"],
252-
record["PUBMED_PMID"]
253-
)
247+
f"Error to parse gene names {record['TRANSLOCATION_NAME']} from cosmic (PMID: {record['PUBMED_PMID']})."
254248
)
255249
else:
256250
up_gene, down_gene = matches.groups()
@@ -333,10 +327,12 @@ def loadMitelman(db_path, db_version, fusions_by_partners, aliases_by_symbol, an
333327
if "," in db_path:
334328
mbca_path, ref_path = db_path.split(",")
335329
pubmed_by_fusion = pubmedByFusion(ref_path)
336-
# MolClin RefNo InvNo Morph Topo Immunology GeneLength GeneShort GeneLong KaryLength KaryShort KaryLong
330+
# MolClin RefNo InvNo Morph Topo Immunology GeneLength GeneShort GeneLong KaryLength KaryShort KaryLong
337331
with HashedSVIO(mbca_path) as reader:
338-
for record in reader:
339-
if record["GeneShort"] != "":
332+
for row_idx, record in enumerate(reader):
333+
if "GeneShort" not in record:
334+
log.warning(f"Invalid record line {row_idx} in {mbca_path}.")
335+
elif record["GeneShort"] != "":
340336
for fusion in record["GeneShort"].split(","):
341337
if "/" in fusion:
342338
genes = fusion.replace("+", "").split("/") # PDRG1/ARF3/RUNX1 => fusion between 3 genes
@@ -348,9 +344,7 @@ def loadMitelman(db_path, db_version, fusions_by_partners, aliases_by_symbol, an
348344
found = True
349345
except Exception:
350346
log.warning(
351-
"Error to parse gene names [{}, {}] from Mitelman (PMID: {}).".format(
352-
up_gene, down_gene, record["RefNo"]
353-
)
347+
f"Error to parse gene names [{up_gene}, {down_gene}] from Mitelman (PMID: {record['RefNo']})."
354348
)
355349
if found:
356350
fusion_partners = "{}_@_{}".format(up_gene, down_gene)
@@ -375,7 +369,7 @@ def pubmedByFusion(in_ref):
375369
:return: Pubmed IDs by fusion partners.
376370
:rtype: dict
377371
"""
378-
# RefNo TitleLength TitleShort TitleLong Volume Year Journal Text Abbreviation AuthorsLength AuthorsShort AuthorsLong Flag Pubmed
372+
# RefNo TitleLength TitleShort TitleLong Volume Year Journal Text Abbreviation AuthorsLength AuthorsShort AuthorsLong Flag Pubmed
379373
pubmed_by_fusion = {}
380374
with HashedSVIO(in_ref) as reader:
381375
for record in reader:
@@ -422,16 +416,12 @@ def __call__(self, parser, namespace, values, option_string=None):
422416
for db_arg in values:
423417
if db_arg.count(":") != 2:
424418
raise argparse.ArgumentTypeError(
425-
'Argument "{}" is invalid. The format must be: "MODEL:VERSION:PATH".'.format(
426-
db_arg
427-
)
419+
f'Argument "{db_arg}" is invalid. The format must be: "MODEL:VERSION:PATH".'
428420
)
429421
model, version, path = db_arg.split(":")
430422
if model.lower() not in fct_by_model:
431423
raise argparse.ArgumentTypeError(
432-
'Database model "{}" is invalid. It must be selected in {}.'.format(
433-
model, sorted(fct_by_model.keys())
434-
)
424+
f'Database model "{model}" is invalid. It must be selected in {sorted(fct_by_model.keys())}.'
435425
)
436426
databases.append({
437427
"model": model,
@@ -451,11 +441,11 @@ def __call__(self, parser, namespace, values, option_string=None):
451441
# Manage parameters
452442
parser = argparse.ArgumentParser(description='Write unique known fusions partners database from multiple databases.')
453443
parser.add_argument('-v', '--version', action='version', version=__version__)
454-
group_input = parser.add_argument_group('Inputs') # Inputs
444+
group_input = parser.add_argument_group('Inputs')
455445
group_input.add_argument('-l', '--input-aliases', required=True, help="Path to the file containing aliases between genes symbols (format: TSV). Each line contains a symbol (Gene name) with one of these aliases (Gene Synonym). This file can be obtain from Ensembl's biomart.")
456446
group_input.add_argument('-a', '--input-annotations', required=True, help='Path to the file containing the genes annotations used in analysis (format: GTF).')
457447
group_input.add_argument('-d', '--inputs-databases', action=InputsDatabases, nargs='+', required=True, help='Paths to databases format: "model:version:path". Model must be in ["babiceanu", "bodymap", "chimerdb", "cosmic", "mitelman"]. Example: -d cosmic:91:~/cosmic91_fusion.tsv chimerdb:Kb3.0:~/chimerKb.tsv mitelman:2019:~/MBCA.TXT.DATA,~/REF.TXT.DATA.')
458-
group_output = parser.add_argument_group('Outputs') # Outputs
448+
group_output = parser.add_argument_group('Outputs')
459449
group_output.add_argument('-o', '--output-database', required=True, help='Path to the fusions partners database (format: TSV).')
460450
args = parser.parse_args()
461451

@@ -474,4 +464,4 @@ def __call__(self, parser, namespace, values, option_string=None):
474464
log.info("Load {}:{}".format(db["model"], db["version"]))
475465
db["parser"](db["path"], db["version"], fusions_by_partners, aliases_by_symbol, annotation_symbols)
476466
writePartnersDb(args.output_database, fusions_by_partners)
477-
log.info("End of job")
467+
log.info("End of job")

0 commit comments

Comments
 (0)