def list_traits(self):
sq = sql_client.sqlClient(self.file)
traits = sq.get_traits()
#traits = []
#for group in self.groups:
# traits.extend(get_data(hdf=self.file, key=group, fields=['phenotype_id'])['phenotype_id'].drop_duplicates().values.tolist())
return list(set(traits))
def chrom_from_trait(self, trait):
sq = sql_client.sqlClient(self.file)
chroms_found = sq.get_chrom_from_trait(trait)
#for group in self.groups:
# chroms_found.extend(self.file.select(group, where='phenotype_id == trait', columns=['chromosome'], index=False).drop_duplicates().values.tolist())
chroms_found = list(set(chroms_found)) # remove dupes
return chroms_found
def chrom_from_gene(self, gene):
sq = sql_client.sqlClient(self.file)
chroms_found = sq.get_chrom_from_gene(gene)
#for group in self.groups:
# chroms_found.extend(self.file.select(group, where='gene_id == gene', columns=['chromosome'], index=False).drop_duplicates().values.tolist())
#chroms_found.extend(get_data(hdf=self.file, key=group, condition=condition, fields=['chromosome'])['chromosome'].drop_duplicates().values.tolist())
chroms_found = list(set(chroms_found)) # remove dupes
return chroms_found
def get_trait_of_study(self, study_to_find):
sq = sql_client.sqlClient(self.sqlite_db)
traits = sq.get_traits_for_study(study_to_find)
if traits:
return sorted(list(set(traits)))
else:
# study not found
raise NotFoundError("Study " + study_to_find)
def map_snp_to_location(self):
try:
snp_no_prefix = re.search(r"[a-zA-Z]+([0-9]+)", self.snp).group(1)
sql = sq.sqlClient(self.snpdb)
mapping = sql.get_chr_pos(snp_no_prefix)
chromosome, position = mapping[0] if mapping else (None, None)
bp_interval = ':'.join([str(position), str(position)])
return (chromosome, bp_interval)
except AttributeError:
return (None, None)
def main():
argparser = argparse.ArgumentParser()
argparser.add_argument('-vcf',
help='The name of the vcf to be processed',
required=False)
argparser.add_argument('-db',
help='The name of the database to load to',
required=True)
argparser.add_argument('-index',
help='create index on the rsid',
required=False,
action='store_true')
args = argparser.parse_args()
db = args.db
if args.vcf:
vcf = args.vcf
vcfdf = pd.read_csv(vcf,
sep='\t',
comment='#',
header=None,
dtype=str,
usecols=[0, 1, 2],
names=['CHROM', 'POS', 'RSID'])
vcfdf.RSID = vcfdf.RSID.str.replace("rs", "")
sql = sq.sqlClient(db)
sql.drop_rsid_index()
list_of_tuples = list(vcfdf.itertuples(index=False, name=None))
sql.cur.execute('BEGIN TRANSACTION')
sql.cur.executemany(
"insert or ignore into snp(chr, position, rsid) values (?, ?, ?)",
list_of_tuples)
sql.cur.execute('COMMIT')
if args.index:
sql = sq.sqlClient(db)
sql.drop_rsid_index()
sql.create_rsid_index()
else:
print("nothing left to do")
def load(self):
sqlcl = sqlClient(database=self.database)
print('starting load...')
print('database: {}'.format(self.database))
try:
sqlcl.drop_rsid_index()
except sqlite3.OperationalError as e:
print(e)
for index, snp in enumerate(self.datasets[SNP_DSET]):
data = (snp, self.datasets[CHR_DSET][index],
self.datasets[BP_DSET][index])
print(data)
sqlcl.insert_snp_row(data)
sqlcl.create_rsid_index()
sqlcl.commit()
def _get_study_metadata(self, key):
sql = sq.sqlClient(self.database)
metadata_dict = sql.get_study_context_meta(key)
return metadata_dict
def get_qtl_list(self):
sq = sql_client.sqlClient(self.sqlite_db)
qtl_list = sq.get_qtl_list()
return sorted(list(set(qtl_list)))
def get_tissue_ont_dict(self):
sq = sql_client.sqlClient(self.sqlite_db)
tissue_ont_dict = sq.get_tissue_ont_dict()
return tissue_ont_dict
def get_list_of_tissues(self):
sq = sql_client.sqlClient(self.sqlite_db)
tissues = sq.get_tissue_ontos()
return sorted(list(set(tissues)))
def get_list_of_studies(self):
sq = sql_client.sqlClient(self.sqlite_db)
studies = sq.get_studies()
return sorted(list(set(studies)))
def check_study(self, study):
sq = sql_client.sqlClient(self.sqlite_db)
if sq.check_study(study):
return True
raise NotFoundError("Study " + study)
def _narrow_hdf_pool(self):
if self.snp:
logger.debug("snp")
self._chr_bp_from_snp()
# narrow by tissue
if self.tissue and self.study:
logger.debug("tissue and study")
sql = sq.sqlClient(self.database)
file_ids = []
resp = sql.get_file_ids_for_study_tissue(self.study, self.tissue,
self.quant_method)
if resp:
file_ids.extend(resp)
if not self._narrow_by_chromosome(file_ids):
raise RequestedNotFound(
"Study :{} with tissue: {} and chr {}".format(
self.study, self.tissue, self.chromosome))
else:
raise RequestedNotFound(
"Study :{} with tissue: {} and quantification method: {}".
format(self.study, self.tissue, self.quant_method))
if self.tissue and not self.study:
logger.debug("tissue")
sql = sq.sqlClient(self.database)
file_ids = []
resp = sql.get_file_ids_for_tissue(self.tissue, self.quant_method)
if resp:
file_ids.extend(resp)
if not self._narrow_by_chromosome(file_ids):
raise RequestedNotFound("Tissue: {} with chr {}".format(
self.tissue, self.chromosome))
else:
raise RequestedNotFound(
"Tissue: {} with quantification method: {}".format(
self.tissue, self.quant_method))
# narrow by qtl group
if self.qtl_group and self.study:
logger.debug("qtl_group and study")
sql = sq.sqlClient(self.database)
file_ids = []
resp = sql.get_file_ids_for_study_qtl_group(
self.study, self.qtl_group, self.quant_method)
if resp:
file_ids.extend(resp)
if not self._narrow_by_chromosome(file_ids):
raise RequestedNotFound(
"Study :{} with qtl_group: {} and chr {}".format(
self.study, self.qtl_group, self.chromosome))
else:
raise RequestedNotFound(
"Study :{} with qtl_group: {} and quantification method: {}"
.format(self.study, self.qtl_group, self.quant_method))
if self.qtl_group and not self.study:
logger.debug("qtl_group")
sql = sq.sqlClient(self.database)
file_ids = []
resp = sql.get_file_ids_for_qtl_group(self.qtl_group,
self.quant_method)
if resp:
file_ids.extend(resp)
if not self._narrow_by_chromosome(file_ids):
raise RequestedNotFound("QTL group: {} with chr {}".format(
self.qtl_group, self.chromosome))
else:
raise RequestedNotFound(
"QTL group: {} with quantification method: {}".format(
self.qtl_group, self.quant_method))
# narrow by anything else
if self.study and not (self.qtl_group or self.tissue):
logger.debug("study")
sql = sq.sqlClient(self.database)
file_ids = []
resp = sql.get_file_id_for_study(self.study, self.quant_method)
if resp:
file_ids.extend(resp)
if not self._narrow_by_chromosome(file_ids):
raise RequestedNotFound("Study :{} with chr {}".format(
self.study, self.chromosome))
else:
raise RequestedNotFound(
"Study :{} with quantification method: {}".format(
self.study, self.quant_method))
if self.quant_method in ["ge", "microarray"]:
if self.snp:
self.hdfs = glob.glob(
os.path.join(self.search_path, self.chr_dir) + "/" +
"/file_" + str(self.chromosome) + "." +
str(self.quant_method) + ".h5")
return "chr"
if self.trait and not (self.study or self.tissue
or self.qtl_group):
logger.debug("phen")
self.chrom_for_trait()
self.hdfs = glob.glob(
os.path.join(self.search_path, self.chr_dir) + "/" +
"/file_" + str(self.chromosome) + "." +
str(self.quant_method) + ".h5")
return "chr"
if self.gene and not (self.study or self.tissue or self.qtl_group):
logger.debug("gene")
self.chrom_for_gene()
self.hdfs = glob.glob(
os.path.join(self.search_path, self.chr_dir) + "/" +
"/file_" + str(self.chromosome) + "." +
str(self.quant_method) + ".h5")
return "chr"
if self.chromosome and all(
v is None
for v in [self.study, self.qtl_group, self.tissue]):
logger.debug("bp/chr")
self.hdfs = glob.glob(
os.path.join(self.search_path, self.chr_dir) + "/" +
"/file_" + str(self.chromosome) + "." +
str(self.quant_method) + ".h5")
return "chr"
if all(v is None for v in [
self.chromosome, self.study, self.gene, self.trait,
self.tissue, self.qtl_group
]):
print("all")
logger.debug("all")
self.hdfs = glob.glob(
os.path.join(self.search_path, self.study_dir) +
"/*/file_*+" + str(self.quant_method) + ".h5")
return "study"
else:
# block for tx/exon/txrev
if self.trait and not (self.study or self.tissue):
logger.debug("phen")
self.chrom_for_trait()
self.hdfs = glob.glob(
os.path.join(self.search_path, self.study_dir) + "/" +
str(self.chromosome) + "/file_*+" +
str(self.quant_method) + ".h5")
if self.gene and not (self.study or self.tissue):
logger.debug("gene")
self.chrom_for_gene()
self.hdfs = glob.glob(
os.path.join(self.search_path, self.study_dir) + "/" +
str(self.chromosome) + "/file_*+" +
str(self.quant_method) + ".h5")
if self.chromosome and all(
v is None
for v in [self.study, self.trait, self.gene, self.tissue]):
logger.debug("bp/chr")
self.hdfs = glob.glob(
os.path.join(self.search_path, self.study_dir) + "/" +
str(self.chromosome) + "/file_*+" +
str(self.quant_method) + ".h5")
if all(v is None for v in [
self.chromosome, self.study, self.gene, self.trait,
self.tissue, self.qtl_group
]):
logger.debug("all")
self.hdfs = glob.glob(
os.path.join(self.search_path, self.study_dir) +
"/*/file_*+" + str(self.quant_method) + ".h5")
return "study"
def has_gene(self, gene):
sq = sql_client.sqlClient(self.file)
search = sq.get_gene(gene)
if search:
return True
return False
def has_trait(self, trait):
sq = sql_client.sqlClient(self.file)
search = sq.get_trait(trait)
if search:
return True
return False
def list_genes(self):
sq = sql_client.sqlClient(self.file)
genes = sq.get_genes()
return list(set(genes))
