Skip to content

Commit

Permalink
Merge pull request #198 from RobokopU24/yeast_stuff
Browse files Browse the repository at this point in the history 
Yeast stuff
  • Loading branch information
@beasleyjonm
beasleyjonm authored Feb 5, 2024
2 parents 5ade525 + a7b53d9 commit 4dea03a
Show file tree
Hide file tree
Showing 6 changed files with 92 additions and 120 deletions.
156 changes: 63 additions & 93 deletions parsers/SGD/src/loadSGD.py
View file
Original file line number Diff line number Diff line change
@@ -1,5 +1,6 @@
import os
import enum
import requests

from parsers.SGD.src.sgd_source_retriever import retrieve_sgd_files
from Common.loader_interface import SourceDataLoader
Expand All @@ -12,6 +13,7 @@
class GENEGOTERMS_EDGEUMAN(enum.IntEnum):
GENE = 0
GOTERM = 5
GOTERM_NAME = 6
PREDICATE = 9
EVIDENCECODE = 8
EVIDENCECODETEXT = 10
Expand All @@ -21,30 +23,39 @@ class GENEGOTERMS_EDGEUMAN(enum.IntEnum):
# Maps Genes to Pathways
class GENEPATHWAYS_EDGEUMAN(enum.IntEnum):
GENE = 0
ORGANISM_NAME = 1
PATHWAY = 2
PATHWAY_NAME = 3
PATHWAY_LINK = 4

# Maps Genes to Phenotypes
class GENEPHENOTYPES_EDGEUMAN(enum.IntEnum):
GENE = 0
PHENOTYPE = 18
QUALIFIER = 8
EXPERIMENTTYPE = 5
MUTANTTYPE = 6
PHENOTYPE_NAME = 7
QUALIFIER = 8
ALLELE = 9
ALLELEDESCRIPTION = 10
STRAINBACKGROUND = 11
CHEMICAL = 12
CONDITION = 13
DETAILS = 14
EVIDENCEPMID = 15
PHENOTYPE = 18
PHENOTYPE_LINK = 19

# Maps Genes to Complexes
class GENECOMPLEXES_EDGEUMAN(enum.IntEnum):
GENE = 11
COMPLEX = 10
NAME = 0
FUNCTION = 1
SYSTEMATIC_NAME = 2
ROLE = 5
STOICHIOMETRY = 6
TYPE = 7
PROPERTIES = 9
COMPLEX = 10
GENE = 11

# Maps Complexes to GO Terms
class COMPLEXEGO_EDGEUMAN(enum.IntEnum):
Expand All @@ -53,6 +64,18 @@ class COMPLEXEGO_EDGEUMAN(enum.IntEnum):
PREDICATE = 3
COMPLEXNAME = 2


def convert_go_qualifier_to_predicate(go_qualifier):
# NOTE that there are a lot more values that go_qualifier could be, these are two currently identified as needing
# mapping before normalization (the rest will be converted succesfully with biolink lookup in edge normalization)
# better long term solution would be to add these two mappings to biolink or to map all the options from GO
if go_qualifier == 'involved in':
return 'biolink:actively_involved_in'
elif go_qualifier == 'is active in':
return 'biolink:active_in'
else:
return go_qualifier

##############
# Class: Mapping SGD Genes to SGD Associations
#
Expand All @@ -64,6 +87,7 @@ class SGDLoader(SourceDataLoader):

source_id: str = 'SGD'
provenance_id: str = 'infores:sgd'
parsing_version = '1.1'

def __init__(self, test_mode: bool = False, source_data_dir: str = None):
"""
Expand All @@ -89,13 +113,19 @@ def __init__(self, test_mode: bool = False, source_data_dir: str = None):
self.complex_to_go_term_edges_file_name
]

self.yeast_complex_base_url = "https://www.yeastgenome.org/complex/"

def get_latest_source_version(self) -> str:
"""
gets the version of the data
:return:
"""
return 'SGD_v1'
# not exactly right because we also get data from http://ontologies.berkeleybop.org/apo.obo for this currently
yeastmine_release_response = requests.get('https://yeastmine.yeastgenome.org/yeastmine/service/version/release')
release_text = yeastmine_release_response.text
release_version = release_text.split('Data Updated on:')[1].split('; GO-Release')[0].strip()
return release_version

def get_data(self) -> int:
"""
Expand Down Expand Up @@ -137,89 +167,15 @@ def parse_data(self) -> dict:
delim=',',
has_header_row=True)

#This file contains GO Term annotations for the genes.
go_term_file: str = os.path.join(self.data_path, self.genes_to_go_term_edges_file_name)
with open(go_term_file, 'r') as fp:
extractor.csv_extract(fp,
lambda line: line[5].replace(" ","_").strip(), # subject id
lambda line: None, # object id
lambda line: None, # predicate extractor
lambda line: {'name': line[6],
NODE_TYPES: [line[7]]
}, #subject props
lambda line: {}, # object props
lambda line: {PRIMARY_KNOWLEDGE_SOURCE: self.provenance_id},#edgeprops
comment_character=None,
delim=',',
has_header_row=True)

#This file is a list of pathways in yeast.
yeast_pathway_list_file: str = os.path.join(self.data_path, self.genes_to_pathway_edges_file_name)
with open(yeast_pathway_list_file, 'r') as fp:
extractor.csv_extract(fp,
lambda line: line[2].replace(" ","_").strip(), # subject id
lambda line: None, # object id
lambda line: None, # predicate extractor
lambda line: {'name': line[3],
NODE_TYPES: ['biolink:Pathway'],
'taxon': 'NCBI_Taxon:559292',
'organism': line[1],
'referenceLink': line[4]}, # subject props
lambda line: {}, # object props
lambda line: {PRIMARY_KNOWLEDGE_SOURCE: self.provenance_id}, #edgeprops
comment_character=None,
delim=',',
has_header_row=True)

#This file is a list of yeast phenotypes.
yeast_phenotype_file: str = os.path.join(self.data_path, self.genes_to_phenotype_edges_file_name)
with open(yeast_phenotype_file, 'r') as fp:
extractor.csv_extract(fp,
lambda line: line[18].replace(" ","_").strip(), # subject id
lambda line: None, # object id
lambda line: None, # predicate extractor
lambda line: {'name': line[7],
NODE_TYPES: ['biolink:PhenotypicFeature'],
'taxon': 'NCBITaxon:559292',
'organism': "S. cerevisiae",
'referenceLink': line[19]}, # subject props
lambda line: {}, # object props
lambda line: {PRIMARY_KNOWLEDGE_SOURCE: self.provenance_id},#edgeprops
comment_character=None,
delim=',',
has_header_row=True)

#This file is a list of yeast protein complexes.
yeast_complex_file: str = os.path.join(self.data_path, self.genes_to_complex_edges_file_name)
with open(yeast_complex_file, 'r') as fp:
extractor.csv_extract(fp,
lambda line: "CPX:" + line[10], # subject id
lambda line: None, # object id
lambda line: None, # predicate extractor
lambda line: {'name': line[0],
NODE_TYPES: ['biolink:MacromolecularComplexMixin'],
'function': line[1],
'systematicName': line[2],
'properties': line[9],
'SGDAccessionID': line[10],
'taxon': 'NCBITaxon:559292',
'organism': "S. cerevisiae",
'referenceLink': line[12]}, # subject props
lambda line: {}, # object props
lambda line: {PRIMARY_KNOWLEDGE_SOURCE: self.provenance_id},#edgeprops
comment_character=None,
delim=',',
has_header_row=True)

#Genes to GO Annotations. Evidence and annotation type as edge properties.
gene_to_go_term_edges_file: str = os.path.join(self.data_path, self.genes_to_go_term_edges_file_name)
with open(gene_to_go_term_edges_file, 'r') as fp:
extractor.csv_extract(fp,
lambda line: line[GENEGOTERMS_EDGEUMAN.GENE.value], #subject id
lambda line: line[GENEGOTERMS_EDGEUMAN.GOTERM.value].replace(' ','_'), # object id
lambda line: line[GENEGOTERMS_EDGEUMAN.PREDICATE.value] if line[GENEGOTERMS_EDGEUMAN.PREDICATE.value] != "involved in" else "biolink:actively_involved_in", # predicate extractor
lambda line: line[GENEGOTERMS_EDGEUMAN.GOTERM.value], # object id
lambda line: convert_go_qualifier_to_predicate(line[GENEGOTERMS_EDGEUMAN.PREDICATE.value]), # predicate extractor
lambda line: {}, # subject props
lambda line: {}, # object props
lambda line: {'name': line[GENEGOTERMS_EDGEUMAN.GOTERM_NAME.value]}, # object props
lambda line: {'evidenceCode': line[GENEGOTERMS_EDGEUMAN.EVIDENCECODE.value],
'evidenceCodeText': line[GENEGOTERMS_EDGEUMAN.EVIDENCECODETEXT.value],
'annotationType': line[GENEGOTERMS_EDGEUMAN.ANNOTATIONTYPE.value],
Expand All @@ -235,13 +191,15 @@ def parse_data(self) -> dict:
with open(gene_to_pathway_edges_file, 'r') as fp:
extractor.csv_extract(fp,
lambda line: line[GENEPATHWAYS_EDGEUMAN.GENE.value], #subject id
lambda line: line[GENEPATHWAYS_EDGEUMAN.PATHWAY.value].replace(' ','_'), # object id
lambda line: line[GENEPATHWAYS_EDGEUMAN.PATHWAY.value], # object id
lambda line: "biolink:participates_in", # predicate extractor
lambda line: {}, # subject props
lambda line: {}, # object props
lambda line: {
PRIMARY_KNOWLEDGE_SOURCE: self.provenance_id
}, #edgeprops
lambda line: {'name': line[GENEPATHWAYS_EDGEUMAN.PATHWAY_NAME.value],
NODE_TYPES: ['biolink:Pathway'],
'taxon': 'NCBI_Taxon:559292',
'organism': line[GENEPATHWAYS_EDGEUMAN.ORGANISM_NAME.value],
'referenceLink': line[GENEPATHWAYS_EDGEUMAN.PATHWAY_LINK.value]}, # object props
lambda line: {PRIMARY_KNOWLEDGE_SOURCE: self.provenance_id}, #edgeprops
comment_character=None,
delim=',',
has_header_row=True)
Expand All @@ -254,7 +212,11 @@ def parse_data(self) -> dict:
lambda line: line[GENEPHENOTYPES_EDGEUMAN.PHENOTYPE.value].replace(' ','_'), # object id
lambda line: "biolink:genetic_association", # predicate extractor
lambda line: {}, # subject props
lambda line: {}, # object props
lambda line: {'name': line[GENEPHENOTYPES_EDGEUMAN.PHENOTYPE_NAME.value],
NODE_TYPES: ['biolink:PhenotypicFeature'],
'taxon': 'NCBITaxon:559292',
'organism': "S. cerevisiae",
'referenceLink': line[GENEPHENOTYPES_EDGEUMAN.PHENOTYPE_LINK.value]}, # object props
lambda line: {'effectOnPhenotype': line[GENEPHENOTYPES_EDGEUMAN.QUALIFIER.value],
'phenotypeDetails': line[GENEPHENOTYPES_EDGEUMAN.DETAILS.value],
'experimentType': line[GENEPHENOTYPES_EDGEUMAN.EXPERIMENTTYPE.value],
Expand All @@ -275,10 +237,18 @@ def parse_data(self) -> dict:
with open(gene_to_complex_edges_file, 'r') as fp:
extractor.csv_extract(fp,
lambda line: line[GENECOMPLEXES_EDGEUMAN.GENE.value], #subject id
lambda line: line[GENECOMPLEXES_EDGEUMAN.COMPLEX.value], # object id
lambda line: f"CPX:{line[GENECOMPLEXES_EDGEUMAN.COMPLEX.value]}", # object id
lambda line: "biolink:in_complex_with", # predicate extractor
lambda line: {}, # subject props
lambda line: {}, # object props
lambda line: {'name': line[GENECOMPLEXES_EDGEUMAN.NAME.value],
NODE_TYPES: ['biolink:MacromolecularComplexMixin'],
'function': line[GENECOMPLEXES_EDGEUMAN.FUNCTION.value],
'systematicName': line[GENECOMPLEXES_EDGEUMAN.SYSTEMATIC_NAME.value],
'properties': line[GENECOMPLEXES_EDGEUMAN.PROPERTIES.value],
'SGDAccessionID': line[GENECOMPLEXES_EDGEUMAN.COMPLEX.value],
'taxon': 'NCBITaxon:559292',
'organism': "S. cerevisiae",
'referenceLink': f'{self.yeast_complex_base_url}{line[GENECOMPLEXES_EDGEUMAN.COMPLEX.value]}'}, # object props
lambda line: {'geneBiologicalRole': line[GENECOMPLEXES_EDGEUMAN.ROLE.value],
'geneStoichiometry': line[GENECOMPLEXES_EDGEUMAN.STOICHIOMETRY.value],
'interactorType': line[GENECOMPLEXES_EDGEUMAN.TYPE.value],
Expand All @@ -291,9 +261,9 @@ def parse_data(self) -> dict:
complex_to_go_term_edges_file: str = os.path.join(self.data_path, self.complex_to_go_term_edges_file_name)
with open(complex_to_go_term_edges_file, 'r') as fp:
extractor.csv_extract(fp,
lambda line: line[COMPLEXEGO_EDGEUMAN.COMPLEX.value],#.split(':')[1], #subject id
lambda line: line[COMPLEXEGO_EDGEUMAN.GOTERM.value],#.replace(' ','_'), # object id
lambda line: line[COMPLEXEGO_EDGEUMAN.PREDICATE.value] if line[COMPLEXEGO_EDGEUMAN.PREDICATE.value] != "involved in" else "biolink:actively_involved_in", # predicate extractor
lambda line: f"CPX:{line[COMPLEXEGO_EDGEUMAN.COMPLEX.value]}", #subject id
lambda line: line[COMPLEXEGO_EDGEUMAN.GOTERM.value], # object id
lambda line: line[COMPLEXEGO_EDGEUMAN.PREDICATE.value], # predicate extractor
lambda line: {}, # subject props
lambda line: {}, # object props
lambda line: {
Expand Down
36 changes: 17 additions & 19 deletions parsers/SGD/src/sgd_source_retriever.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -210,23 +210,19 @@ def SGDGene2Complex(data_directory):
"allInteractors.participant.secondaryIdentifier",
"allInteractors.biologicalRole", "allInteractors.stoichiometry",
"allInteractors.type", "identifier", "properties", "accession",
"allInteractors.participant.genes.primaryIdentifier", "complex.link"]
"allInteractors.participant.genes.primaryIdentifier"]

data = dict.fromkeys(view, [])
url = "https://www.yeastgenome.org/complex/"

for row in query.rows():
for col in range(len(view)):
key = view[col]
if key != "complex.link":
if key == "identifier":
value = "EBI:" + str(row[key])
elif key == "allInteractors.participant.genes.primaryIdentifier":
value = "SGD:" + str(row[key])
else:
value = row[key]
elif key == "complex.link":
value = url + str(row[view[10]])
for key in view:
if key == "identifier":
value = "EBI:" + str(row[key])
elif key == "allInteractors.participant.genes.primaryIdentifier":
value = "SGD:" + str(row[key])
elif key == "properties":
value = str(row[key]).replace("\r\n", " ").replace("\n", " ").strip()
else:
value = row[key]
data[key] = data[key] + [value]
gene2complexdf = pd.DataFrame(data)
gene2complexdf.fillna("?", inplace=True)
Expand Down Expand Up @@ -254,16 +250,18 @@ def SGDComplex2GOTerm(data_directory):

for row in query.rows():
for col in view:

if col == "accession":
value = "CPX:" + str(row[col])
value = str(row[col])
elif col == "goAnnotation.qualifier":
if str(row["goAnnotation.ontologyTerm.namespace"]) == "molecular_function":
ontology_term_namespace = str(row["goAnnotation.ontologyTerm.namespace"])
if ontology_term_namespace == "molecular_function":
value = "biolink:enables"
elif str(row["goAnnotation.ontologyTerm.namespace"]) == "biological_process":
elif ontology_term_namespace == "biological_process":
value = "biolink:actively_involved_in"
elif str(row["goAnnotation.ontologyTerm.namespace"]) == "cellular_component":
elif ontology_term_namespace == "cellular_component":
value = "biolink:located_in"
else:
value = ""
else:
value = str(row[col])
data[col] = data[col] + [value]
Expand Down
9 changes: 6 additions & 3 deletions parsers/yeast/src/loadCostanza2016.py
View file
Original file line number Diff line number Diff line change
@@ -1,6 +1,7 @@
import os
import enum
import csv
import requests
from Common.loader_interface import SourceDataLoader
from Common.extractor import Extractor
from Common.node_types import PRIMARY_KNOWLEDGE_SOURCE, PUBLICATIONS
Expand Down Expand Up @@ -28,7 +29,7 @@ class COSTANZA_GENEINTERACTIONS(enum.IntEnum):
##############
class Costanza2016Loader(SourceDataLoader):

source_id: str = 'Costanza2016'
source_id: str = 'Costanza2016Data'
provenance_id: str = 'infores:CostanzaGeneticInteractions'
parsing_version = '1.1'

Expand Down Expand Up @@ -56,8 +57,10 @@ def get_latest_source_version(self) -> str:
:return:
"""
# TODO - this is actually possible with https://yeastmine.yeastgenome.org/yeastmine/service/version/release
return 'yeast_v1'
yeastmine_release_response = requests.get('https://yeastmine.yeastgenome.org/yeastmine/service/version/release')
release_text = yeastmine_release_response.text
release_version = release_text.split('Data Updated on:')[1].split('; GO-Release')[0].strip()
return release_version

def get_data(self) -> int:
# Collects all data for complexes with GO Term annotations in SGD.
Expand Down
Loading

0 comments on commit 4dea03a

Please sign in to comment.