def read_gaf_csv(filename, version) -> pd:
ecomapping = ecomap.EcoMap()
data_frame = pd.read_csv(filename,
comment='!',
sep='\t',
header=None,
na_filter=False,
names=["DB",
"DB_Object_ID",
"DB_Object_Symbol",
"Qualifier",
"GO_ID",
"DB_Reference",
"Evidence_code",
"With_or_From",
"Aspect",
"DB_Object_Name",
"DB_Object_Synonym",
"DB_Object_Type,"
"Taxon",
"Date",
"Assigned_By",
"Annotation_Extension",
"Gene_Product_Form_ID"]).fillna("")
new_df = data_frame.filter(['DB_Object_ID', 'Qualifier', 'GO_ID', 'Evidence_code', 'DB_Reference'], axis=1)
for eco_code in ecomapping.mappings():
for ev in new_df['Evidence_code']:
if eco_code[2] == ev:
new_df['Evidence_code'] = new_df['Evidence_code'].replace([eco_code[2]],
ecomapping.ecoclass_to_coderef(
eco_code[2])[0])
return new_df
def __init__(self,
remove_double_prefixes=False,
ontology=None,
repair_obsoletes=True,
entity_map=None,
valid_taxa=None,
class_idspaces=None,
entity_idspaces=None,
ecomap=ecomap.EcoMap(),
exclude_relations=None,
include_relations=None,
filter_out_evidence=None,
filtered_evidence_file=None,
gpi_authority_path=None,
paint=False):
self.remove_double_prefixes = remove_double_prefixes
self.ontology = ontology
self.repair_obsoletes = repair_obsoletes
self.entity_map = entity_map
self.valid_taxa = valid_taxa
self.class_idspaces = class_idspaces
self.ecomap = ecomap
self.include_relations = include_relations
self.exclude_relations = exclude_relations
self.filter_out_evidence = filter_out_evidence
self.filtered_evidence_file = filtered_evidence_file
self.gpi_authority_path = gpi_authority_path
self.paint = paint
if self.exclude_relations is None:
self.exclude_relations = []
if self.include_relations is None:
self.include_relations = []
if self.filter_out_evidence is None:
self.filter_out_evidence = []
def __init__(self, file=None, version=GPAD_1_2):
self.file = file
if version in [GPAD_1_2, GPAD_2_0]:
self.version = version
else:
self.version = GPAD_1_2
self._write("!gpa-version: {}\n".format(self.version))
self.ecomap = ecomap.EcoMap()
def __init__(self,
remove_double_prefixes=False,
ontology=None,
repair_obsoletes=True,
entity_map=None,
valid_taxa=None,
class_idspaces=None,
entity_idspaces=None,
group_idspace=None,
ecomap=ecomap.EcoMap(),
exclude_relations=None,
include_relations=None,
filter_out_evidence=None,
filtered_evidence_file=None,
gpi_authority_path=None,
paint=False,
rule_metadata=dict(),
goref_metadata=None,
group_metadata=None,
dbxrefs=None,
suppress_rule_reporting_tags=[],
annotation_inferences=None,
extensions_constraints=None,
rule_contexts=[]):
self.remove_double_prefixes = remove_double_prefixes
self.ontology = ontology
self.repair_obsoletes = repair_obsoletes
self.entity_map = entity_map
self.valid_taxa = valid_taxa
self.class_idspaces = class_idspaces
self.ecomap = ecomap
self.include_relations = include_relations
self.exclude_relations = exclude_relations
self.filter_out_evidence = filter_out_evidence
self.filtered_evidence_file = filtered_evidence_file
self.gpi_authority_path = gpi_authority_path
self.paint = paint
self.rule_metadata = rule_metadata
self.goref_metadata = goref_metadata
self.group_metadata = group_metadata
self.suppress_rule_reporting_tags = suppress_rule_reporting_tags
self.annotation_inferences = annotation_inferences
self.entity_idspaces = entity_idspaces
self.extensions_constraints = AssocParserConfig._compute_constraint_subclasses(
extensions_constraints, ontology)
self.group_idspace = None if group_idspace is None else set(
group_idspace)
self.rule_contexts = rule_contexts
# This is a dictionary from ruleid: `gorule-0000001` to title strings
if self.exclude_relations is None:
self.exclude_relations = []
if self.include_relations is None:
self.include_relations = []
if self.filter_out_evidence is None:
self.filter_out_evidence = []
def __init__(self,
remove_double_prefixes=False,
ontology=None,
repair_obsoletes=True,
entity_map=None,
valid_taxa=None,
class_idspaces=None,
entity_idspaces=None,
group_idspace=None,
ecomap=ecomap.EcoMap(),
exclude_relations=None,
include_relations=None,
filter_out_evidence=None,
filtered_evidence_file=None,
gpi_authority_path=None,
paint=False,
rule_titles=None,
dbxrefs=None):
self.remove_double_prefixes=remove_double_prefixes
self.ontology=ontology
self.repair_obsoletes=repair_obsoletes
self.entity_map=entity_map
self.valid_taxa=valid_taxa
self.class_idspaces=class_idspaces
self.ecomap=ecomap
self.include_relations=include_relations
self.exclude_relations=exclude_relations
self.filter_out_evidence = filter_out_evidence
self.filtered_evidence_file = filtered_evidence_file
self.gpi_authority_path = gpi_authority_path
self.paint = paint
self.rule_titles = rule_titles
self.entity_idspaces = None if entity_idspaces is None else set(entity_idspaces)
self.group_idspace = None if group_idspace is None else set(group_idspace)
# This is a dictionary from ruleid: `gorule-0000001` to title strings
if self.exclude_relations is None:
self.exclude_relations = []
if self.include_relations is None:
self.include_relations = []
if self.filter_out_evidence is None:
self.filter_out_evidence = []
def read_gpad_csv(filename, version) -> pd:
if version.startswith("1"):
data_frame = pd.read_csv(filename,
comment='!',
sep='\t',
header=None,
na_filter=False,
names=gpad_1_2_format).fillna("")
df = data_frame.filter(['db', 'subject', 'qualifiers', 'relation', 'object', 'evidence_code', 'reference'], axis=1)
concat_column = df['db'] + ":" + df['subject']
df['concat_column'] = concat_column
filtered_df = df.filter(['concat_column', 'qualifiers', 'relation', 'object', 'evidence_code', 'reference'])
filtered_df.rename(columns={'concat_column': 'subject'}, inplace=True)
new_df = filtered_df
else:
data_frame = pd.read_csv(filename,
comment='!',
sep='\t',
header=None,
na_filter=False,
names=gpad_2_0_format).fillna("")
new_df = data_frame.filter(['subject', 'negation', 'relation', 'object', 'evidence_code', 'reference'], axis=1)
ecomapping = ecomap.EcoMap()
for eco_code in ecomapping.mappings():
for ev in new_df['evidence_code']:
if eco_code[2] == ev:
new_df['evidence_code'] = new_df['evidence_code'].replace([eco_code[2]],
ecomapping.ecoclass_to_coderef(eco_code[2])[0])
# normalize ids
config = assocparser.AssocParserConfig()
config.remove_double_prefixes = True
parser = gpadparser.GpadParser(config=config)
for i, r in enumerate(new_df['subject']):
r1 = parser._normalize_id(r)
new_df.at[i, 'subject'] = r1
return new_df
import logging
from typing import List
from ontobio.io import assocparser
from ontobio.io.gpadparser import GpadParser
from ontobio.model.association import GoAssociation, Date
from ontobio.rdfgen.gocamgen import errors
from ontobio import ecomap
logger = logging.getLogger(__name__)
ecomapping = ecomap.EcoMap()
ipi_eco = ecomapping.coderef_to_ecoclass("IPI")
GPAD_PARSER = GpadParser()
BINDING_ROOT = "GO:0005488" # binding
IPI_ECO_CODE = ipi_eco
class GoAssocWithFrom:
"""
Separate with/from column values into
header vs line arrangement.
Used for explicit placement in
annotation assertions.
"""
def __init__(self, header=None, line=None):
if header is None:
header = []
if line is None:
line = []
self.header = sorted(header)
def __init__(self, file=None):
self.file = file
self._write("!gpa-version: 1.1\n")
self.ecomap = ecomap.EcoMap()
def __init__(self):
super().__init__(
"GORULE:0000043",
"Check for valid combination of evidence code and GO_REF",
FailMode.SOFT)
self.ecomapping = ecomap.EcoMap()
def __init__(self,
remove_double_prefixes=False,
ontology=None,
repair_obsoletes=True,
entity_map=None,
valid_taxa=None,
class_idspaces=None,
entity_idspaces=None,
group_idspace=None,
ecomap=ecomap.EcoMap(),
exclude_relations=None,
include_relations=None,
filter_out_evidence=None,
filtered_evidence_file=None,
gpi_authority_path=None,
paint=False,
rule_metadata=dict(),
goref_metadata=None,
group_metadata=None,
dbxrefs=None,
suppress_rule_reporting_tags=[],
annotation_inferences=None,
extensions_constraints=None,
rule_contexts=[],
rule_set=None):
self.remove_double_prefixes = remove_double_prefixes
self.ontology = ontology
self.repair_obsoletes = repair_obsoletes
self.entity_map = entity_map
self.valid_taxa = valid_taxa
self.class_idspaces = class_idspaces
self.ecomap = ecomap
self.include_relations = include_relations
self.exclude_relations = exclude_relations
self.filter_out_evidence = filter_out_evidence
self.filtered_evidence_file = filtered_evidence_file
self.gpi_authority_path = gpi_authority_path
self.paint = paint
self.rule_metadata = rule_metadata
self.goref_metadata = goref_metadata
self.group_metadata = group_metadata
self.suppress_rule_reporting_tags = suppress_rule_reporting_tags
self.annotation_inferences = annotation_inferences
self.entity_idspaces = entity_idspaces
self.extensions_constraints = AssocParserConfig._compute_constraint_subclasses(
extensions_constraints, ontology)
self.group_idspace = None if group_idspace is None else set(
group_idspace)
self.rule_contexts = rule_contexts
# We'll say that the default None should run no rules, so let's set the rule_set to []
# print("Rule Set is {}".format(rule_set))
if rule_set == None:
self.rule_set = RuleSet([])
elif rule_set == RuleSet.ALL:
# None here means all rules
self.rule_set = RuleSet(None)
else:
self.rule_set = RuleSet(rule_set)
# This is a dictionary from ruleid: `gorule-0000001` to title strings
if self.exclude_relations is None:
self.exclude_relations = []
if self.include_relations is None:
self.include_relations = []
if self.filter_out_evidence is None:
self.filter_out_evidence = []
