def load_analysis_from_dirs(ctx, alignment_dir, hmmcopy_dir, annotation_dir,
id, projects, reload):
#is this for msk? for isabl or tantalus, i guess it was igo aka isabl
#andrew's function, going to keep it here for when he sends over one-off data sets
es_host = ctx.obj['host']
es_port = ctx.obj["port"]
assert reload or not _is_loaded(
id, es_host, es_port
), f'Dashboard with ID {id} already loaded. To reload, add --reload to command'
nonexistant_projects = [
project for project in projects
if not _is_project_exist(project, es_host, es_port)
]
assert len(nonexistant_projects
) == 0, f'Projects do not exist: {nonexistant_projects} '
if reload:
_clean_analysis(id, host=es_host, port=es_port)
alhena.alhena_loader.load_analysis_from_dirs(id, projects, es_host,
es_port, alignment_dir,
hmmcopy_dir, annotation_dir)
def load_analysis(ctx, data_directory, id, projects, reload):
es_host = ctx.obj['host']
es_port = ctx.obj["port"]
assert reload or not _is_loaded(
id, es_host, es_port
), f'Dashboard with ID {id} already loaded. To reload, add --reload to command'
nonexistant_projects = [
project for project in projects
if not _is_project_exist(project, es_host, es_port)
]
assert len(nonexistant_projects
) == 0, f'Projects do not exist: {nonexistant_projects} '
if reload:
_clean_analysis(id, host=es_host, port=es_port)
hmmcopy_data = _get_scgenome_colossus_tantalus_data(data_directory)
analysis_record = _get_colossus_tantalus_analysis_object(
data_directory, dashboard_id)
_load_analysis(id, hmmcopy_data, analysis_record, projects, data_directory,
es_host, es_port)
def load_merged_analysis_bccrc(ctx, data_directory, id, projects, reload):
es_host = ctx.obj['host']
es_port = ctx.obj["port"]
# get metadata.json ,sc-test.json, id.json located in the data directory
# check to see if those libraries exist
# oneline new function called bccrc_verify_libraries()
assert reload or not _is_loaded(
id, es_host, es_port
), f'Dashboard with ID {id} already loaded. To reload, add --reload to command'
nonexistant_projects = [
project for project in projects
if not _is_project_exist(project, es_host, es_port)
]
assert len(nonexistant_projects
) == 0, f'Projects do not exist: {nonexistant_projects} '
if reload:
_clean_analysis(id, host=es_host, port=es_port)
_download_libraries_for_merged(id, data_directory)
_load_merged_analysis(id, projects, data_directory, es_host, es_port)
def load_analysis_msk(ctx, id, projects, reload):
es_host = ctx.obj['host']
es_port = ctx.obj["port"]
annotation_pk = _get_scgenome_isabl_annotation_pk(id)
#hmmcopy_data = None
hmmcopy_data = _get_scgenome_isabl_data(id)
print(f'{id} is now {str(annotation_pk)}')
id = str(annotation_pk)
assert reload or not _is_loaded(
id, es_host, es_port
), f'Dashboard with ID {id} already loaded. To reload, add --reload to command'
nonexistent_projects = [
project for project in projects
if not _is_project_exist(project, es_host, es_port)
]
assert len(nonexistent_projects
) == 0, f'Projects do not exist: {nonexistent_projects} '
if reload:
_clean_analysis(id, host=es_host, port=es_port)
analysis_record = _get_isabl_analysis_object(id)
_load_analysis(dashboard_id, hmmcopy_data, analysis_record, projects, host,
port)
def load_analysis(ctx, data_directory, id, projects, reload):
es_host = ctx.obj['host']
es_port = ctx.obj["port"]
assert reload or not _is_loaded(
id, es_host, es_port
), f'Dashboard with ID {id} already loaded. To reload, add --reload to command'
nonexistant_projects = [
project for project in projects
if not _is_project_exist(project, es_host, es_port)
]
assert len(nonexistant_projects
) == 0, f'Projects do not exist: {nonexistant_projects} '
if reload:
_clean_analysis(id, host=es_host, port=es_port)
_load_analysis(id, projects, data_directory, es_host, es_port)
def add_project(ctx, project_name, dashboards):
es_host = ctx.obj['host']
es_port = ctx.obj["port"]
assert not _is_project_exist(
project_name, es_host,
es_port), f'Project with name {project_name} already exists'
if dashboards[0] == "":
dashboards = []
unloaded_dashboards = [
dashboard_id for dashboard_id in dashboards
if not _is_loaded(dashboard_id, es_host, es_port)
]
assert len(unloaded_dashboards
) == 0, f'Dashboards do not exist: {unloaded_dashboards}'
_add_project(project_name, list(dashboards), es_host, es_port)
def verify_projects(ctx, projects):
es_host = ctx.obj['host']
es_port = ctx.obj["port"]
logger = ctx.obj["logger"]
good_projects = []
bad_projects = []
for project in projects:
if _is_project_exist(project, es_host, es_port):
good_projects.append(project)
else:
bad_projects.append(project)
logger.info(f'==== Verified project names: {len(good_projects)}')
for project in good_projects:
logger.info(project)
logger.info(
f'==== Incorrect / Missing project names: {len(bad_projects)} ')
for project in bad_projects:
logger.info(project)
def load_dashboard(ctx, data_directory, id, projects, download, reload):
es_host = ctx.obj['host']
es_port = ctx.obj["port"]
nonexistant_projects = [
project for project in projects
if not _is_project_exist(project, es_host, es_port)
]
assert len(nonexistant_projects
) == 0, f'Projects do not exist: {nonexistant_projects} '
download_type = "merged" if os.path.exists(
os.path.join(data_directory, constants.MERGED_DIRECTORYNAME,
f"{id}.json")) else "single"
if download:
if download_type == "merged":
_download_libraries_for_merged(id, data_directory)
elif download_type == "single":
data_directory = _download_analysis(id, data_directory)
if reload:
_clean_analysis(id, host=es_host, port=es_port, projects=projects)
if _is_loaded(id, es_host, es_port):
_add_dashboard_to_projects(id, projects, es_host, es_port)
else:
if download_type == "merged":
_load_merged_analysis(id, projects, data_directory, es_host,
es_port)
elif download_type == "single":
hmmcopy_data = _get_scgenome_colossus_tantalus_data(data_directory)
analysis_record = __get_colossus_tantalus_analysis_object(
directory, dashboard_id)
_load_analysis(id, hmmcopy_data, analysis_record, projects,
data_directory, es_host, es_port)