def print_hgnc(out_dir):
"""Print HPO files to a directory
Args:
out_dir(Path)
"""
file_name = "hgnc.txt"
file_path = out_dir / file_name
LOG.info("Downloads HGNC genes to %s", file_path)
with file_path.open("w", encoding="utf-8") as outfile:
for line in fetch_hgnc():
outfile.write(line + "\n")
def test_fetch_hgnc(hgnc_file, mocker):
"""Test fetch hgnc"""
# GIVEN file with hgnc info
mocker.patch.object(scout_requests.urllib.request, "urlopen")
with open(hgnc_file, "rb") as hgnc_handle:
hgnc_info = hgnc_handle.read()
with tempfile.TemporaryFile() as temp:
temp.write(hgnc_info)
temp.seek(0)
scout_requests.urllib.request.urlopen.return_value = temp
# WHEN fetching the resource
data = scout_requests.fetch_hgnc()
# THEN assert that the HGNC header is there
assert "hgnc_id\tsymbol" in data[0]
def generate_hgnc(genes):
"""Generate lines from a file with reduced hgnc information
Args:
genes(dict): A dictionary with hgnc_id as key and hgnc_symbol as value
outpath(str): Defaults to hgnc_reduced_path
Yields:
print_line(str): Lines from the reduced file
"""
LOG.info("Generating new hgnc reduced file")
# fetch the latest hgnc file here
hgnc_gene_lines = fetch_hgnc()
header = None
genes_found = 0
# Loop over all hgnc gene lines
for i, line in enumerate(hgnc_gene_lines):
line = line.rstrip()
# Skip lines that are empty
if not len(line) > 0:
continue
# If we are reading the header, print it
if i == 0:
header = line.split("\t")
yield line
continue
# Parse the hgnc gene line
gene = parse_hgnc_line(line, header)
if not gene:
continue
hgnc_id = int(gene["hgnc_id"])
# Check if the gene is in the reduced
if hgnc_id in genes:
genes_found += 1
yield line
LOG.info("Number of genes printed to file: %s", genes_found)
def load_hgnc_genes(
adapter,
genes=None,
ensembl_lines=None,
hgnc_lines=None,
exac_lines=None,
mim2gene_lines=None,
genemap_lines=None,
hpo_lines=None,
build="37",
omim_api_key="",
):
"""Load genes into the database
link_genes will collect information from all the different sources and
merge it into a dictionary with hgnc_id as key and gene information as values.
Args:
adapter(scout.adapter.MongoAdapter)
genes(dict): If genes are already parsed
ensembl_lines(iterable(str)): Lines formated with ensembl gene information
hgnc_lines(iterable(str)): Lines with gene information from genenames.org
exac_lines(iterable(str)): Lines with information pLi-scores from ExAC
mim2gene(iterable(str)): Lines with map from omim id to gene symbol
genemap_lines(iterable(str)): Lines with information of omim entries
hpo_lines(iterable(str)): Lines information about map from hpo terms to genes
build(str): What build to use. Defaults to '37'
Returns:
gene_objects(list): A list with all gene_objects that was loaded into database
"""
gene_objects = list()
if not genes:
# Fetch the resources if not provided
if ensembl_lines is None:
ensembl_lines = fetch_ensembl_genes(build=build)
hgnc_lines = hgnc_lines or fetch_hgnc()
exac_lines = exac_lines or fetch_exac_constraint()
if not (mim2gene_lines and genemap_lines):
if not omim_api_key:
LOG.warning("No omim api key provided!")
else:
mim_files = fetch_mim_files(omim_api_key,
mim2genes=True,
genemap2=True)
mim2gene_lines = mim_files["mim2genes"]
genemap_lines = mim_files["genemap2"]
if not hpo_lines:
hpo_files = fetch_hpo_files(hpogenes=True)
hpo_lines = hpo_files["hpogenes"]
# Link the resources
genes = link_genes(
ensembl_lines=ensembl_lines,
hgnc_lines=hgnc_lines,
exac_lines=exac_lines,
hpo_lines=hpo_lines,
mim2gene_lines=mim2gene_lines,
genemap_lines=genemap_lines,
)
non_existing = 0
nr_genes = len(genes)
with progressbar(genes.values(), label="Building genes",
length=nr_genes) as bar:
for gene_data in bar:
if not gene_data.get("chromosome"):
LOG.debug(
"skipping gene: %s. No coordinates found",
gene_data.get("hgnc_symbol", "?"),
)
non_existing += 1
continue
gene_obj = build_hgnc_gene(gene_data, build=build)
gene_objects.append(gene_obj)
LOG.info("Loading genes build %s", build)
adapter.load_hgnc_bulk(gene_objects)
LOG.info("Loading done. %s genes loaded", len(gene_objects))
LOG.info("Nr of genes without coordinates in build %s: %s", build,
non_existing)
return gene_objects
def setup_scout(
adapter,
institute_id="cust000",
user_name="Clark Kent",
user_mail="*****@*****.**",
api_key=None,
demo=False,
resource_files=None,
):
"""Function to setup a working scout instance.
WARNING: If the instance is populated all collections will be deleted
Build insert a institute and an admin user.
There are multiple sources of information that is used by scout and that needs to exist for
scout to work proper.
Genes:
Scout uses HGNC as the source for gene identifiers en ensembl as source for coordinates.
Additional information of disease connections for genes if fetched from OMIM.
Link between hpo terms and genes is fetched from HPO
For more details check the documentation.
"""
LOG.info("Check if there was a database, delete if existing")
existing_database = False
for collection_name in adapter.db.list_collection_names():
if collection_name.startswith("system"):
continue
LOG.info("Deleting collection %s", collection_name)
adapter.db.drop_collection(collection_name)
existing_database = True
if existing_database:
LOG.info("Database deleted")
institute_obj = build_institute(
internal_id=institute_id,
display_name=institute_id,
sanger_recipients=[user_mail],
)
adapter.add_institute(institute_obj)
user_obj = dict(
_id=user_mail,
email=user_mail,
name=user_name,
roles=["admin"],
institutes=[institute_id],
)
adapter.add_user(user_obj)
resource_files = resource_files or {}
if demo:
resource_files = demo_files
mim2gene_lines = None
genemap_lines = None
mim2gene_path = resource_files.get("mim2gene_path")
genemap_path = resource_files.get("genemap_path")
if genemap_path and mim2gene_path:
mim2gene_lines = [line for line in get_file_handle(mim2gene_path)]
genemap_lines = [line for line in get_file_handle(genemap_path)]
if (genemap_lines is None) and api_key:
try:
mim_files = fetch_mim_files(api_key, mim2genes=True, genemap2=True)
except Exception as err:
LOG.warning(err)
raise err
mim2gene_lines = mim_files["mim2genes"]
genemap_lines = mim_files["genemap2"]
if resource_files.get("hpogenes_path"):
hpo_gene_lines = [
line
for line in get_file_handle(resource_files.get("hpogenes_path"))
]
else:
hpo_gene_lines = fetch_genes_to_hpo_to_disease()
if resource_files.get("hgnc_path"):
hgnc_lines = [
line for line in get_file_handle(resource_files.get("hgnc_path"))
]
else:
hgnc_lines = fetch_hgnc()
if resource_files.get("exac_path"):
exac_lines = [
line for line in get_file_handle(resource_files.get("exac_path"))
]
else:
exac_lines = fetch_exac_constraint()
# Load cytobands into cytoband collection
for genome_build, cytobands_path in cytoband_files.items():
load_cytobands(cytobands_path, genome_build, adapter)
builds = ["37", "38"]
for build in builds:
genes_path = "genes{}_path".format(build)
if resource_files.get(genes_path):
ensembl_genes = get_file_handle(resource_files[genes_path])
else:
ensembl_genes = fetch_ensembl_genes(build=build)
hgnc_genes = load_hgnc_genes(
adapter=adapter,
ensembl_lines=ensembl_genes,
hgnc_lines=hgnc_lines,
exac_lines=exac_lines,
mim2gene_lines=mim2gene_lines,
genemap_lines=genemap_lines,
hpo_lines=hpo_gene_lines,
build=build,
)
# Create a map from ensembl ids to gene objects
ensembl_genes = {}
for gene_obj in hgnc_genes:
ensembl_id = gene_obj["ensembl_id"]
ensembl_genes[ensembl_id] = gene_obj
tx_path = "transcripts{}_path".format(build)
if resource_files.get(tx_path):
ensembl_transcripts = get_file_handle(resource_files[tx_path])
else:
ensembl_transcripts = fetch_ensembl_transcripts(build=build)
# Load the transcripts for a certain build
transcripts = load_transcripts(adapter, ensembl_transcripts, build,
ensembl_genes)
hpo_terms_handle = None
if resource_files.get("hpoterms_path"):
hpo_terms_handle = get_file_handle(resource_files["hpoterms_path"])
hpo_to_genes_handle = None
if resource_files.get("hpo_to_genes_path"):
hpo_to_genes_handle = get_file_handle(
resource_files["hpo_to_genes_path"])
hpo_disease_handle = None
if resource_files.get("hpo_disease_path"):
hpo_disease_handle = get_file_handle(
resource_files["hpo_disease_path"])
load_hpo(
adapter=adapter,
disease_lines=genemap_lines,
hpo_lines=hpo_terms_handle,
hpo_gene_lines=hpo_to_genes_handle,
)
# If demo we load a gene panel and some case information
if demo:
parsed_panel = parse_gene_panel(
path=panel_path,
institute="cust000",
panel_id="panel1",
version=1.0,
display_name="Test panel",
)
adapter.load_panel(parsed_panel)
case_handle = get_file_handle(load_path)
case_data = yaml.load(case_handle, Loader=yaml.FullLoader)
config_data = parse_case_data(config=case_data)
adapter.load_case(config_data)
LOG.info("Creating indexes")
adapter.load_indexes()
LOG.info("Scout instance setup successful")
def genes(build, api_key):
"""
Load the hgnc aliases to the mongo database.
"""
LOG.info("Running scout update genes")
adapter = store
# Fetch the omim information
api_key = api_key or current_app.config.get("OMIM_API_KEY")
mim_files = {}
if not api_key:
LOG.warning("No omim api key provided, Please not that some information will be missing")
else:
try:
mim_files = fetch_mim_files(api_key, mim2genes=True, morbidmap=True, genemap2=True)
except Exception as err:
LOG.warning(err)
raise click.Abort()
LOG.warning("Dropping all gene information")
adapter.drop_genes(build)
LOG.info("Genes dropped")
LOG.warning("Dropping all transcript information")
adapter.drop_transcripts(build)
LOG.info("transcripts dropped")
hpo_genes = fetch_genes_to_hpo_to_disease()
if build:
builds = [build]
else:
builds = ["37", "38"]
hgnc_lines = fetch_hgnc()
exac_lines = fetch_exac_constraint()
for build in builds:
ensembl_genes = fetch_ensembl_genes(build=build)
# load the genes
hgnc_genes = load_hgnc_genes(
adapter=adapter,
ensembl_lines=ensembl_genes,
hgnc_lines=hgnc_lines,
exac_lines=exac_lines,
mim2gene_lines=mim_files.get("mim2genes"),
genemap_lines=mim_files.get("genemap2"),
hpo_lines=hpo_genes,
build=build,
)
ensembl_genes = {}
for gene_obj in hgnc_genes:
ensembl_id = gene_obj["ensembl_id"]
ensembl_genes[ensembl_id] = gene_obj
# Fetch the transcripts from ensembl
ensembl_transcripts = fetch_ensembl_transcripts(build=build)
transcripts = load_transcripts(adapter, ensembl_transcripts, build, ensembl_genes)
adapter.update_indexes()
LOG.info("Genes, transcripts and Exons loaded")