def go_starter():
bud_session = bud_session_maker()
nex_session = nex_session_maker()
slim_ids = set()
for pieces in make_file_starter('src/sgd/convert/data/go_slim_mapping.tab.txt')():
if len(pieces) >= 6:
goid = pieces[5]
slim_ids.add(goid)
key_to_source = dict([(x.unique_key(), x) for x in nex_session.query(Source).all()])
for bud_obj in bud_session.query(Go).all():
go_id = 'GO:' + str(bud_obj.go_go_id).zfill(7)
yield {'display_name': bud_obj.go_term,
'source': key_to_source['GO'],
'description': bud_obj.go_definition,
'go_id': go_id,
'go_aspect': abbrev_to_go_aspect[bud_obj.go_aspect],
'is_slim': 1 if go_id in slim_ids else 0,
'date_created': bud_obj.date_created,
'created_by': bud_obj.created_by}
bud_session.close()
nex_session.close()
def paragraph_reference_starter():
nex_session = nex_session_maker()
references = nex_session.query(Reference).all()
key_to_paragraph = dict([(x.unique_key(), x) for x in nex_session.query(Paragraph).all()])
pubmed_id_to_reference = dict([(x.pubmed_id, x) for x in references])
sgdid_to_reference = dict([(x.sgdid, x) for x in references])
# LSP
for paragraph in key_to_paragraph.values():
if paragraph.category == "LSP":
sgdids = [x.split(">")[0] for x in paragraph.text.split("<reference:")]
for sgdid in sgdids:
if sgdid in sgdid_to_reference:
reference = sgdid_to_reference[sgdid]
yield {"paragraph_id": paragraph.id, "reference_id": reference.id}
else:
if sgdid != "<p":
print "Reference not found: " + sgdid
# Regulation
file_names = [
"src/sgd/convert/data/regulationSummaries",
"src/sgd/convert/data/15-8regulationSummaries.txt",
"src/sgd/convert/data/15-9regulationSummaries.txt",
"src/sgd/convert/data/15-10regulationSummaries.txt",
"src/sgd/convert/data/15-11regulationSummaries.txt",
]
for file_name in file_names:
for row in make_file_starter(file_name)():
paragraph_key = (row[0], "BIOENTITY", "REGULATION")
for pubmed_id in [int(x) for x in row[3].strip().split("|") if x != "references" and x != ""]:
if paragraph_key in key_to_paragraph and pubmed_id in pubmed_id_to_reference:
yield {
"paragraph_id": key_to_paragraph[paragraph_key].id,
"reference_id": pubmed_id_to_reference[pubmed_id].id,
}
else:
print "Paragraph or reference not found: " + str(paragraph_key) + " " + str(pubmed_id)
yield None
# Strain
for strain_key, paragraph in strain_paragraphs.iteritems():
paragraph_key = (strain_key, "STRAIN", None)
for pubmed_id in paragraph[1]:
if paragraph_key in key_to_paragraph and pubmed_id in pubmed_id_to_reference:
yield {
"paragraph_id": key_to_paragraph[paragraph_key].id,
"reference_id": pubmed_id_to_reference[pubmed_id].id,
}
else:
print "Paragraph or reference not found: " + str(paragraph_key) + " " + str(pubmed_id)
yield None
nex_session.close()
def tag_starter():
nex_session = nex_session_maker()
for row in make_file_starter('src/sgd/convert/data/microarray_05_14/SPELL-tags.txt')():
tag = row[2].strip()
for t in [x.strip() for x in tag.split('|')]:
if t != '':
yield {
'display_name': t,
'description': definitions.get(t)
}
nex_session.close()
def bioitem_tag_starter():
nex_session = nex_session_maker()
key_to_dataset = dict([(x.unique_key(), x) for x in nex_session.query(Dataset).all()])
key_to_tag = dict([(x.unique_key(), x) for x in nex_session.query(Tag).all()])
for row in make_file_starter('src/sgd/convert/data/microarray_05_14/SPELL-tags.txt')():
dataset_key = (row[1].strip()[:-4], 'DATASET')
tags = row[2].strip()
for t in [x.strip() for x in tags.split('|')]:
if t != '':
yield {
'bioitem': key_to_dataset[dataset_key],
'tag': key_to_tag[create_format_name(t)]
}
nex_session.close()
def datasetcolumn_starter():
nex_session = nex_session_maker()
key_to_source = dict([(x.unique_key(), x) for x in nex_session.query(Source).all()])
key_to_dataset = dict([(x.unique_key(), x) for x in nex_session.query(Dataset).all()])
#Filename to pubmed_id, geo_id, channel_count, tags, and short description
key_to_GSM = dict([((x[2], x[4].replace('delta', '').replace('sigma', '').strip()), x[5]) for x in make_file_starter(expression_dir + '/pmid_filename_gse_conds_tags_file_20150204.txt')()])
key_to_GSM.update([((x[2], x[4].strip()), x[5]) for x in make_file_starter(expression_dir + '/pmid_filename_gse_conds_tags_file_20150204.txt')()])
for path in os.listdir(expression_dir):
if os.path.isdir(expression_dir + '/' + path):
for file in os.listdir(expression_dir + '/' + path):
dataset_key = (file[:-4], 'DATASET')
if dataset_key in key_to_dataset:
f = open(expression_dir + '/' + path + '/' + file, 'r')
pieces = f.next().split('\t')
f.close()
geo_id = key_to_dataset[dataset_key].geo_id
i = 0
for piece in pieces[3:]:
column_name = piece.strip().decode('ascii','ignore')
if (geo_id, column_name) not in key_to_GSM and geo_id is not None:
print (geo_id, column_name)
#print (geo_id, column_name)
col_geo_id = None if (geo_id, column_name) not in key_to_GSM else key_to_GSM[(geo_id, column_name)]
link = None if col_geo_id is None else 'http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=' + col_geo_id
yield {
'description': column_name,
'dataset': key_to_dataset[dataset_key],
'source': key_to_source['SGD'],
'file_order': i,
'geo_id': col_geo_id,
'link': link
}
i += 1
nex_session.close()
def dataset_starter():
nex_session = nex_session_maker()
key_to_source = dict([(x.unique_key(), x) for x in nex_session.query(Source).all()])
pubmed_id_to_reference = dict([(x.pubmed_id, x) for x in nex_session.query(Reference).all()])
#Filename to pubmed_id, geo_id, channel_count, tags, and short description
filename_to_info = dict([(x[1], (int(x[0]), x[2], x[3], x[6], None)) for x in make_file_starter(expression_dir + '/pmid_filename_gse_conds_tags_file_20150204.txt')()])
for path in os.listdir(expression_dir):
if os.path.isdir(expression_dir + '/' + path):
full_description = None
geo_id = None
pubmed_id = None
state = 'BEGIN'
try:
for row in make_file_starter(expression_dir + '/' + path + '/README')():
if row[0].startswith('Full Description'):
state = 'FULL_DESCRIPTION:'
full_description = row[0][18:].strip()
elif row[0].startswith('PMID:'):
pubmed_id = int(row[0][6:].strip())
elif row[0].startswith('GEO ID:'):
geo_id = row[0][8:].strip().split('.')[0].split('GPL')[0]
elif row[0].startswith('PCL filename'):
state = 'OTHER'
elif state == 'FULL_DESCRIPTION':
full_description = full_description + row[0].strip()
elif state == 'OTHER':
pcl_filename = row[0].strip()
short_description = row[1].strip()
tag = row[3].strip()
if pcl_filename in filename_to_info:
filename_to_info[pcl_filename] = (filename_to_info[pcl_filename][0], filename_to_info[pcl_filename][1], filename_to_info[pcl_filename][2], filename_to_info[pcl_filename][3], short_description)
else:
if geo_id == 'N/A':
geo_id = None
filename_to_info[pcl_filename] = (pubmed_id, geo_id, 1, tag, short_description)
for file in os.listdir(expression_dir + '/' + path):
if file != 'README':
f = open(expression_dir + '/' + path + '/' + file, 'r')
pieces = f.next().split('\t')
f.close()
if file in filename_to_info:
pubmed_id, geo_id, channel_count, tags, short_description = filename_to_info[file]
if pubmed_id not in pubmed_id_to_reference:
print 'Warning: pubmed_id not found ' + str(pubmed_id)
yield {
'description': full_description,
'geo_id': geo_id,
'pcl_filename': file,
'short_description': short_description,
'tags': tags,
'reference': None if pubmed_id is None or pubmed_id not in pubmed_id_to_reference else pubmed_id_to_reference[pubmed_id],
'source': key_to_source['SGD'],
'channel_count': channel_count,
'condition_count': len(pieces)-3
}
else:
print 'Filename not in readme: ' + file
except:
print 'File ' + expression_dir + '/' + path + '/README' + ' not found.'
print traceback.format_exc()
nex_session.close()
def domain_starter():
bud_session = bud_session_maker()
nex_session = nex_session_maker()
key_to_source = dict([(x.unique_key(), x) for x in nex_session.query(Source).all()])
panther_id_to_description = {}
for row in make_file_starter('src/sgd/convert/data/PANTHER9.0_HMM_classifications.txt')():
panther_id_to_description[row[0]] = row[1].lower()
for row in make_file_starter('src/sgd/convert/data/domains.tab')():
source_key = row[3].strip()
if source_key.startswith('ProSite'):
source_key = 'Prosite'
if source_key.startswith('SignalP'):
source_key = 'SignalP'
if source_key.startswith('Hamap'):
source_key = 'HAMAP'
if source_key == 'Coils':
source_key = '-'
display_name = row[4].strip()
description = row[5].strip()
interpro_id = None
interpro_description = None
if len(row) == 13:
interpro_id = row[11].strip()
interpro_description = row[12].strip()
source_key = create_format_name(source_key)
source = None if source_key not in key_to_source else key_to_source[source_key]
description = None if description == '' else description
interpro_description = None if interpro_description == '' else interpro_description
interpro_id = None if interpro_id == '' else interpro_id
if source_key == 'PANTHER':
if display_name in panther_id_to_description:
yield {'display_name': display_name,
'source': source,
'description': panther_id_to_description[display_name],
'bioitem_type': source_key,
'interpro_id': interpro_id,
'interpro_description': interpro_description}
elif source_key is not None:
yield {'display_name': display_name,
'source': source,
'description': description if description is not None else interpro_description,
'bioitem_type': source_key,
'interpro_id': interpro_id,
'interpro_description': interpro_description}
else:
print 'Source not found: ' + source_key
for row in make_file_starter('src/sgd/convert/data/TF_family_class_accession04302013.txt')():
description = 'Class: ' + row[4] + ', Family: ' + row[3]
yield {'display_name': row[0],
'source': key_to_source['JASPAR'],
'description': description,
'bioitem_type': 'JASPAR'}
yield {'display_name': 'predicted signal peptide',
'source': key_to_source['SignalP'],
'description': 'predicted signal peptide',
'bioitem_type': 'SignalP'}
yield {'display_name': 'predicted transmembrane domain',
'source': key_to_source['TMHMM'],
'description': 'predicted transmembrane domain',
'bioitem_type': 'TMHMM'}
for bud_obj in bud_session.query(Dbxref).filter(or_(Dbxref.dbxref_type == 'PANTHER', Dbxref.dbxref_type == 'Prosite')).all():
dbxref_type = bud_obj.dbxref_type
source_key = create_format_name(bud_obj.source)
source = None if source_key not in key_to_source else key_to_source[source_key]
if source is None:
print source_key
yield None
bioitem_type = None
if dbxref_type == 'Prosite ID':
bioitem_type = 'Prosite'
elif dbxref_type == 'PANTHER':
bioitem_type = 'PANTHER'
if bioitem_type == 'PANTHER':
if display_name in panther_id_to_description:
yield {'display_name': bud_obj.dbxref_id,
'source': source,
'description': panther_id_to_description[bud_obj.dbxref_id],
'bioitem_type': bioitem_type}
else:
yield {'display_name': bud_obj.dbxref_id,
'source': source,
'description': bud_obj.dbxref_name,
'bioitem_type': bioitem_type}
bud_session.close()
nex_session.close()
def experiment_starter():
bud_session = bud_session_maker()
nex_session = nex_session_maker()
key_to_source = dict([(x.unique_key(), x) for x in nex_session.query(Source).all()])
for bud_obj in make_obo_file_starter('src/sgd/convert/data/eco.obo')():
description = None if 'def' not in bud_obj else bud_obj['def']
if description is not None and description.find('[') >= 0:
description = description[:description.find('[')-1]
if description is not None and description.find('"') >= 0:
description = description[1:-1]
yield {'display_name': bud_obj['name'],
'source': key_to_source['ECO'],
'description': description,
'eco_id': bud_obj['id']}
for bud_obj in bud_session.query(CVTerm).filter(CVTerm.cv_no==7).all():
format_name = create_format_name(bud_obj.name)
yield {'display_name': bud_obj.name,
'source': key_to_source['SGD'],
'description': bud_obj.definition,
'category': 'large-scale survey' if format_name in large_scale_survey else 'classical genetics' if format_name in classical_genetics else None,
'date_created': bud_obj.date_created,
'created_by': bud_obj.created_by}
for row in make_file_starter('src/sgd/convert/data/2014-05-15_reg_data/Venters_Macisaac_Hu05-12-2014_regulator_lines')():
source_key = row[11].strip()
if source_key in key_to_source:
yield {'display_name': row[4] if row[4] != '' else row[5],
'source': None if source_key not in key_to_source else key_to_source[source_key],
'eco_id': row[5]}
else:
print 'Source not found: ' + str(source_key)
for row in make_file_starter('src/sgd/convert/data/2014-05-15_reg_data/SGD_data_05_14_2014')():
source_key = row[11].strip()
if source_key in key_to_source:
yield {'display_name': row[4] if row[4] != '' else row[5],
'source': None if source_key not in key_to_source else key_to_source[source_key],
'eco_id': row[5]}
else:
print 'Source not found: ' + str(source_key)
for row in make_file_starter('src/sgd/convert/data/2014-05-15_reg_data/Madhani_fixed')():
if len(row) >= 10:
if source_key in key_to_source:
source_key = row[11].strip()
yield {'display_name': row[5] if row[5] != '' else row[4],
'source': None if source_key not in key_to_source else key_to_source[source_key],
'eco_id': row[4]}
else:
print 'Source not found: ' + str(source_key)
for row in make_file_starter('src/sgd/convert/data/2014-05-15_reg_data/Pimentel_PMID22616008.txt')():
if len(row) >= 10:
if source_key in key_to_source:
source_key = row[11].strip()
yield {'display_name': row[4] if row[4] != '' else row[5],
'source': None if source_key not in key_to_source else key_to_source[source_key],
'eco_id': row[5]}
else:
print 'Source not found: ' + str(source_key)
for row in make_file_starter('src/sgd/convert/data/yetfasco_data.txt', delimeter=';')():
expert_confidence = row[8][1:-1]
if expert_confidence == 'High':
yield {'display_name': row[9][1:-1],
'source': key_to_source['YeTFaSCo']}
yield {'display_name': 'protein abundance', 'source': key_to_source['SGD']}
yield {'display_name': 'EXP', 'source': key_to_source['GO'], 'link': 'http://www.geneontology.org/page/exp-inferred-experiment', 'description': 'Inferred from Experiment'}
yield {'display_name': 'IDA', 'source': key_to_source['GO'], 'link': 'http://www.geneontology.org/page/ida-inferred-direct-assay', 'description': 'Inferred from Direct Assay'}
yield {'display_name': 'IPI', 'source': key_to_source['GO'], 'link': 'http://www.geneontology.org/page/ipi-inferred-physical-interaction', 'description': 'Inferred from Physical Interaction'}
yield {'display_name': 'IMP', 'source': key_to_source['GO'], 'link': 'http://www.geneontology.org/page/imp-inferred-mutant-phenotype', 'description': 'Inferred from Mutant Phenotype'}
yield {'display_name': 'IGI', 'source': key_to_source['GO'], 'link': 'http://www.geneontology.org/page/igi-inferred-genetic-interaction', 'description': 'Inferred from Genetic Interaction'}
yield {'display_name': 'IEP', 'source': key_to_source['GO'], 'link': 'http://www.geneontology.org/page/iep-inferred-expression-pattern', 'description': 'Inferred from Expression Pattern'}
yield {'display_name': 'ISS', 'source': key_to_source['GO'], 'link': 'http://www.geneontology.org/page/iss-inferred-sequence-or-structural-similarity', 'description': 'Inferred from Sequence or Structural Similarity'}
yield {'display_name': 'ISA', 'source': key_to_source['GO'], 'link': 'http://www.geneontology.org/page/isa-inferred-sequence-alignment', 'description': 'Inferred from Sequence Alignment'}
yield {'display_name': 'ISO', 'source': key_to_source['GO'], 'link': 'http://www.geneontology.org/page/iso-inferred-sequence-orthology', 'description': 'Inferred from Sequence Orthology'}
yield {'display_name': 'ISM', 'source': key_to_source['GO'], 'link': 'http://www.geneontology.org/page/ism-inferred-sequence-model', 'description': 'Inferred from Sequence Model'}
yield {'display_name': 'IGC', 'source': key_to_source['GO'], 'link': 'http://www.geneontology.org/page/igc-inferred-genomic-context', 'description': 'Inferred from Genomic Context'}
yield {'display_name': 'IBA', 'source': key_to_source['GO'], 'link': 'http://www.geneontology.org/page/iba-inferred-biological-aspect-ancestor', 'description': 'Inferred from Biological aspect of Ancestor'}
yield {'display_name': 'IBD', 'source': key_to_source['GO'], 'link': 'http://www.geneontology.org/page/ibd-inferred-biological-aspect-descendent', 'description': 'Inferred from Biological aspect of Descendent'}
yield {'display_name': 'IKR', 'source': key_to_source['GO'], 'link': 'http://www.geneontology.org/page/ikr-inferred-key-residues', 'description': 'Inferred from Key Residues'}
yield {'display_name': 'IRD', 'source': key_to_source['GO'], 'link': 'http://www.geneontology.org/page/ird-inferred-rapid-divergence', 'description': 'Inferred from Rapid Divergence'}
yield {'display_name': 'RCA', 'source': key_to_source['GO'], 'link': 'http://www.geneontology.org/page/rca-inferred-reviewed-computational-analysis', 'description': 'inferred from Reviewed Computational Analysis'}
yield {'display_name': 'TAS', 'source': key_to_source['GO'], 'link': 'http://www.geneontology.org/page/tas-traceable-author-statement', 'description': 'Traceable Author Statement'}
yield {'display_name': 'NAS', 'source': key_to_source['GO'], 'link': 'http://www.geneontology.org/page/nas-non-traceable-author-statement', 'description': 'Non-traceable Author Statement'}
yield {'display_name': 'IC', 'source': key_to_source['GO'], 'link': 'http://www.geneontology.org/page/ic-inferred-curator', 'description': 'Inferred by Curator'}
yield {'display_name': 'ND', 'source': key_to_source['GO'], 'link': 'http://www.geneontology.org/page/nd-no-biological-data-available', 'description': 'No Biological Data Available'}
yield {'display_name': 'IEA', 'source': key_to_source['GO'], 'link': 'http://www.geneontology.org/page/automatically-assigned-evidence-codes', 'description': 'Inferred from Electronic Annotation'}
bud_session.close()
nex_session.close()
def paragraph_reference_starter():
nex_session = nex_session_maker()
references = nex_session.query(Reference).all()
key_to_paragraph = dict([(x.unique_key(), x) for x in nex_session.query(Paragraph).all()])
pubmed_id_to_reference = dict([(x.pubmed_id, x) for x in references])
sgdid_to_reference = dict([(x.sgdid, x) for x in references])
#LSP
for paragraph in key_to_paragraph.values():
if paragraph.category == 'LSP':
sgdids = [x.split('>')[0] for x in paragraph.text.split('<reference:')]
for sgdid in sgdids:
if sgdid in sgdid_to_reference:
reference = sgdid_to_reference[sgdid]
yield {
'paragraph_id': paragraph.id,
'reference_id': reference.id
}
else:
if sgdid != '<p':
print 'Reference not found: ' + sgdid
#Regulation
file_names = ['src/sgd/convert/data/regulationSummaries',
'src/sgd/convert/data/15-8regulationSummaries.txt',
'src/sgd/convert/data/15-9regulationSummaries.txt',
'src/sgd/convert/data/15-10regulationSummaries.txt',
'src/sgd/convert/data/15-11regulationSummaries.txt',
'src/sgd/convert/data/16-1regulationSummaries.txt',
'src/sgd/convert/data/16-2regulationSummaries.txt',
'src/sgd/convert/data/16-3regulationSummaries.txt',
'src/sgd/convert/data/16-4regulationSummaries.txt',
'src/sgd/convert/data/16-5regulationSummaries.txt']
for file_name in file_names:
for row in make_file_starter(file_name)():
paragraph_key = (row[0], 'BIOENTITY', 'REGULATION')
for pubmed_id in [int(x) for x in row[3].strip().split('|') if x != 'references' and x != '']:
if paragraph_key in key_to_paragraph and pubmed_id in pubmed_id_to_reference:
yield {
'paragraph_id': key_to_paragraph[paragraph_key].id,
'reference_id': pubmed_id_to_reference[pubmed_id].id,
}
else:
print 'Paragraph or reference not found: ' + str(paragraph_key) + ' ' + str(pubmed_id)
yield None
#Strain
for strain_key, paragraph in strain_paragraphs.iteritems():
paragraph_key = (strain_key, 'STRAIN', None)
for pubmed_id in paragraph[1]:
if paragraph_key in key_to_paragraph and pubmed_id in pubmed_id_to_reference:
yield {
'paragraph_id': key_to_paragraph[paragraph_key].id,
'reference_id': pubmed_id_to_reference[pubmed_id].id,
}
else:
print 'Paragraph or reference not found: ' + str(paragraph_key) + ' ' + str(pubmed_id)
yield None
nex_session.close()
def bioentity_paragraph_starter():
bud_session = bud_session_maker()
nex_session = nex_session_maker()
key_to_source = dict([(x.unique_key(), x) for x in nex_session.query(Source).all()])
key_to_bioentity = dict([(x.unique_key(), x) for x in nex_session.query(Locus).all()])
id_to_bioentity = dict([(x.id, x) for x in nex_session.query(Locus).all()])
sgdid_to_reference = dict([(x.sgdid, x) for x in nex_session.query(Reference).all()])
sgdid_to_bioentity = dict([(x.sgdid, x) for x in nex_session.query(Bioentity).all()])
goid_to_go = dict([(int(x.go_id[3:]), x) for x in nex_session.query(Go).all()])
#LSP
for feature in bud_session.query(Feature).all():
paragraph_feats = feature.paragraph_feats
if len(paragraph_feats) > 0 and feature.id in id_to_bioentity:
paragraph_feats.sort(key=lambda x: x.order)
paragraph_html, paragraph_text = clean_paragraph(id_to_bioentity[feature.id], '<p>' + ('</p><p>'.join([x.paragraph.text for x in paragraph_feats])) + '</p>', str([x.paragraph.id for x in paragraph_feats]), sgdid_to_reference, sgdid_to_bioentity, goid_to_go)
date_edited = None
year = 0
month = 0
day = 0
for paragraph_feat in paragraph_feats:
my_date = paragraph_feat.paragraph.date_edited
this_date = str(my_date).split(' ')[0].replace('-0', '-').split('-')
this_year = int(this_date[0])
this_month = int(this_date[1])
this_day = int(this_date[2])
if date_edited is None or datetime(this_year, this_month, this_day) > datetime(year, month, day):
date_edited = my_date
year = this_year
month = this_month
day = this_day
yield {
'bioentity': id_to_bioentity[feature.id],
'source': key_to_source['SGD'],
'text': paragraph_text,
'html': paragraph_html,
'date_edited': date_edited,
'date_created': paragraph_feats[0].paragraph.date_created,
'created_by': paragraph_feats[0].paragraph.created_by,
'category': 'LSP'
}
bioentity_key_to_date = dict()
#Go
for gofeature in bud_session.query(GoFeature).all():
bioentity_key = (gofeature.feature.name, 'LOCUS')
if gofeature.annotation_type == 'manually curated' and bioentity_key not in bioentity_key_to_date:
bioentity_key_to_date[bioentity_key] = gofeature.date_last_reviewed
for bioentity_key, date_last_reviewed in bioentity_key_to_date.iteritems():
if bioentity_key in key_to_bioentity:
yield {
'bioentity': key_to_bioentity[bioentity_key],
'source': key_to_source['SGD'],
'text': str(date_last_reviewed),
'html': str(date_last_reviewed),
'date_created': None,
'created_by': None,
'category': 'GODATE'
}
else:
#print 'Bioentity not found: ' + str(bioentity_key)
yield None
for pieces in make_file_starter('src/sgd/convert/data/gp_information.559292_sgd')():
if len(pieces) >= 8:
sgdid = pieces[8]
if sgdid.startswith('SGD:'):
sgdid = sgdid[4:]
go_annotation = [x[22:].strip() for x in pieces[9].split('|') if x.startswith('go_annotation_summary')]
if len(go_annotation) == 1:
if sgdid in sgdid_to_bioentity:
yield {
'bioentity': sgdid_to_bioentity[sgdid],
'source': key_to_source['SGD'],
'text': go_annotation[0],
'html': go_annotation[0],
'date_created': None,
'created_by': None,
'category': 'GO'
}
else:
print 'Bioentity not found: ' + sgdid
yield None
#Regulation
file_names = ['src/sgd/convert/data/regulationSummaries',
'src/sgd/convert/data/15-8regulationSummaries.txt',
'src/sgd/convert/data/15-9regulationSummaries.txt',
'src/sgd/convert/data/15-10regulationSummaries.txt',
'src/sgd/convert/data/15-11regulationSummaries.txt',
'src/sgd/convert/data/16-1regulationSummaries.txt',
'src/sgd/convert/data/16-2regulationSummaries.txt',
'src/sgd/convert/data/16-3regulationSummaries.txt',
'src/sgd/convert/data/16-4regulationSummaries.txt',
'src/sgd/convert/data/16-5regulationSummaries.txt']
for file_name in file_names:
for row in make_file_starter(file_name)():
bioentity_key = (row[0], 'LOCUS')
if bioentity_key in key_to_bioentity:
bioentity = key_to_bioentity[bioentity_key]
yield {
'bioentity': bioentity,
'source': key_to_source['SGD'],
'text': row[2],
'html': link_gene_names(row[2], {bioentity.display_name, bioentity.format_name, bioentity.display_name + 'P', bioentity.format_name + 'P'}, nex_session),
'category': 'REGULATION'
}
else:
#print 'Bioentity not found: ' + str(bioentity_key)
yield None
#Phenotype
file_names = ['src/sgd/convert/data/PhenotypeSummaries032015.txt',
'src/sgd/convert/data/15-6phenoSummariesTyposFixed.txt',
'src/sgd/convert/data/15-7phenoSummaries.txt',
'src/sgd/convert/data/15-8phenoSummaries.txt',
'src/sgd/convert/data/15-9phenoSummaries.txt',
'src/sgd/convert/data/15-10phenoSummaries.txt',
'src/sgd/convert/data/15-11phenoSummaries.txt',
'src/sgd/convert/data/15-12phenoSummaries.txt',
'src/sgd/convert/data/16-1phenoSummaries.txt',
'src/sgd/convert/data/16-2phenoSummaries.txt',
'src/sgd/convert/data/16-3phenoSummaries.txt',
'src/sgd/convert/data/16-4phenoSummaries.txt',
'src/sgd/convert/data/16-5phenoSummaries.txt',
'src/sgd/convert/data/16-6phenoSummaries.txt',
'src/sgd/convert/data/16-7phenoSummaries.txt',
'src/sgd/convert/data/16-9phenoSummaries.txt',
'src/sgd/convert/data/16-10phenoSummaries.txt']
for file_name in file_names:
for row in make_file_starter(file_name)():
bioentity_key = (row[0], 'LOCUS')
if bioentity_key in key_to_bioentity:
bioentity = key_to_bioentity[bioentity_key]
yield {
'bioentity': bioentity,
'source': key_to_source['SGD'],
'text': row[1],
'html': link_gene_names(row[1], {bioentity.display_name, bioentity.format_name, bioentity.display_name + 'P', bioentity.format_name + 'P'}, nex_session),
'category': 'PHENOTYPE'
}
else:
#print 'Bioentity not found: ' + str(bioentity_key)
yield None
bud_session.close()
nex_session.close()
def bioentity_paragraph_starter():
bud_session = bud_session_maker()
nex_session = nex_session_maker()
key_to_source = dict([(x.unique_key(), x) for x in nex_session.query(Source).all()])
key_to_bioentity = dict([(x.unique_key(), x) for x in nex_session.query(Locus).all()])
id_to_bioentity = dict([(x.id, x) for x in nex_session.query(Locus).all()])
sgdid_to_reference = dict([(x.sgdid, x) for x in nex_session.query(Reference).all()])
sgdid_to_bioentity = dict([(x.sgdid, x) for x in nex_session.query(Bioentity).all()])
goid_to_go = dict([(int(x.go_id[3:]), x) for x in nex_session.query(Go).all()])
# LSP
for feature in bud_session.query(Feature).all():
paragraph_feats = feature.paragraph_feats
if len(paragraph_feats) > 0 and feature.id in id_to_bioentity:
paragraph_feats.sort(key=lambda x: x.order)
paragraph_html, paragraph_text = clean_paragraph(
id_to_bioentity[feature.id],
"<p>" + ("</p><p>".join([x.paragraph.text for x in paragraph_feats])) + "</p>",
str([x.paragraph.id for x in paragraph_feats]),
sgdid_to_reference,
sgdid_to_bioentity,
goid_to_go,
)
yield {
"bioentity": id_to_bioentity[feature.id],
"source": key_to_source["SGD"],
"text": paragraph_text,
"html": paragraph_html,
"date_edited": paragraph_feats[0].paragraph.date_edited,
"date_created": paragraph_feats[0].paragraph.date_created,
"created_by": paragraph_feats[0].paragraph.created_by,
"category": "LSP",
}
bioentity_key_to_date = dict()
# Go
for gofeature in bud_session.query(GoFeature).all():
bioentity_key = (gofeature.feature.name, "LOCUS")
if gofeature.annotation_type == "manually curated" and bioentity_key not in bioentity_key_to_date:
bioentity_key_to_date[bioentity_key] = gofeature.date_last_reviewed
for bioentity_key, date_last_reviewed in bioentity_key_to_date.iteritems():
if bioentity_key in key_to_bioentity:
yield {
"bioentity": key_to_bioentity[bioentity_key],
"source": key_to_source["SGD"],
"text": str(date_last_reviewed),
"html": str(date_last_reviewed),
"date_created": None,
"created_by": None,
"category": "GODATE",
}
else:
# print 'Bioentity not found: ' + str(bioentity_key)
yield None
for pieces in make_file_starter("src/sgd/convert/data/gp_information.559292_sgd")():
if len(pieces) >= 8:
sgdid = pieces[8]
if sgdid.startswith("SGD:"):
sgdid = sgdid[4:]
go_annotation = [
x[22:].strip() for x in pieces[9].split("|") if x.startswith("go_annotation_summary")
]
if len(go_annotation) == 1:
if sgdid in sgdid_to_bioentity:
yield {
"bioentity": sgdid_to_bioentity[sgdid],
"source": key_to_source["SGD"],
"text": go_annotation[0],
"html": go_annotation[0],
"date_created": None,
"created_by": None,
"category": "GO",
}
else:
print "Bioentity not found: " + sgdid
yield None
# Regulation
file_names = [
"src/sgd/convert/data/regulationSummaries",
"src/sgd/convert/data/15-8regulationSummaries.txt",
"src/sgd/convert/data/15-9regulationSummaries.txt",
"src/sgd/convert/data/15-10regulationSummaries.txt",
"src/sgd/convert/data/15-11regulationSummaries.txt",
]
for file_name in file_names:
for row in make_file_starter(file_name)():
bioentity_key = (row[0], "LOCUS")
if bioentity_key in key_to_bioentity:
bioentity = key_to_bioentity[bioentity_key]
yield {
"bioentity": bioentity,
"source": key_to_source["SGD"],
"text": row[2],
"html": link_gene_names(
row[2],
{
bioentity.display_name,
bioentity.format_name,
bioentity.display_name + "P",
bioentity.format_name + "P",
},
nex_session,
),
"category": "REGULATION",
}
else:
# print 'Bioentity not found: ' + str(bioentity_key)
yield None
# Phenotype
file_names = [
"src/sgd/convert/data/PhenotypeSummaries032015.txt",
"src/sgd/convert/data/15-6phenoSummariesTyposFixed.txt",
"src/sgd/convert/data/15-7phenoSummaries.txt",
"src/sgd/convert/data/15-8phenoSummaries.txt",
"src/sgd/convert/data/15-9phenoSummaries.txt",
"src/sgd/convert/data/15-10phenoSummaries.txt",
"src/sgd/convert/data/15-11phenoSummaries.txt",
]
for file_name in file_names:
for row in make_file_starter(file_name)():
bioentity_key = (row[0], "LOCUS")
if bioentity_key in key_to_bioentity:
bioentity = key_to_bioentity[bioentity_key]
yield {
"bioentity": bioentity,
"source": key_to_source["SGD"],
"text": row[1],
"html": link_gene_names(
row[1],
{
bioentity.display_name,
bioentity.format_name,
bioentity.display_name + "P",
bioentity.format_name + "P",
},
nex_session,
),
"category": "PHENOTYPE",
}
else:
# print 'Bioentity not found: ' + str(bioentity_key)
yield None
bud_session.close()
nex_session.close()
clean_up_orphans(nex_session_maker, DNAsequenceevidence, Evidence, 'DNASEQUENCE')
for sequence_filename, coding_sequence_filename, strain_key in new_sequence_files:
do_conversion(make_new_dna_sequence_evidence_starter(nex_session_maker, strain_key, sequence_filename, coding_sequence_filename),
[Json2Obj(DNAsequenceevidence),
Obj2NexDB(nex_session_maker, lambda x: x.query(DNAsequenceevidence).filter(DNAsequenceevidence.strain_id == strain_key_to_id[strain_key]).filter(DNAsequenceevidence.dna_type != '1KB'),
name='convert.from_bud.evidence.dnasequence',
delete_untouched=True,
commit_interval=1000)])
update_contig_centromeres(nex_session_maker)
update_contig_reference_alignment(nex_session_maker)
protparam_data = dict([(row[0], row) for row in make_file_starter('src/sgd/convert/data/ProtParam.txt')()])
for sequence_filename, strain_key in protein_sequence_files:
do_conversion(make_protein_sequence_evidence_starter(nex_session_maker, strain_key, sequence_filename, protparam_data),
[Json2Obj(Proteinsequenceevidence),
Obj2NexDB(nex_session_maker, lambda x: x.query(Proteinsequenceevidence).filter(Proteinsequenceevidence.strain_id == strain_key_to_id[strain_key]), name='convert.from_bud.evidence.proteinsequence', delete_untouched=True, commit_interval=1000)])
clean_up_orphans(nex_session_maker, Proteinsequenceevidence, Evidence, 'PROTEINSEQUENCE')
do_conversion(make_kb_sequence_starter(nex_session_maker),
[Json2Obj(DNAsequenceevidence),
Obj2NexDB(nex_session_maker, lambda x: x.query(DNAsequenceevidence).filter(DNAsequenceevidence.dna_type == '1KB'),
name='convert.from_bud.evidence.1kb_dnasequence',
delete_untouched=True,
commit_interval=1000)])
