def __init__(self,
reference_name,
annotation_name,
annotation_version=None,
decompress_on_download=False,
copy_local_files_to_cache=False,
install_string_function=None,
cache_directory_path=None):
"""
Parameters
----------
reference_name : str
Name of reference genome
annotation_name : str
Name of annotation database
annotation_version : str or int, optional
Version or release of annotation database
decompress_on_download : bool, optional
If downloading a .fa.gz file, should we automatically expand it
into a decompressed FASTA file?
copy_local_files_to_cache : bool, optional
If file is on the local file system, should we still copy it
into the cache?
install_string_function : fn, optional
Function which returns an error message with
install instructions. If not provided then the error tells the
user what data is missing without install instructions.
cache_directory_path : str, optional
Where to place downloaded and temporary files, by default
inferred from reference name, annotation name, annotation version,
and the global cache directory determined by datacache.
"""
self.reference_name = reference_name
self.annotation_name = annotation_name
self.annotation_version = annotation_version
# using hidden member variable _cache_directory path since access to
# to the visible cache_directory_path (no underscore!) is combined
# with ensuring that the directpry actually exists
if cache_directory_path:
self._cache_directory_path = cache_directory_path
else:
self.cache_subdirectory = cache_subdirectory(
reference_name=reference_name,
annotation_name=annotation_name,
annotation_version=annotation_version)
self._cache_directory_path = datacache.get_data_dir(
subdir=self.cache_subdirectory)
self.decompress_on_download = decompress_on_download
self.copy_local_files_to_cache = copy_local_files_to_cache
self.install_string_function = install_string_function
def fm_index_path(genome):
"""
Returns a path for cached reference peptides, for the given genome.
"""
# if $VAXRANK_REF_PEPTIDES_DIR is set, that'll be the location of the cache
cache_dir = get_data_dir(envkey='VAXRANK_REF_PEPTIDES_DIR')
if not os.path.exists(cache_dir):
os.makedirs(cache_dir)
return os.path.join(
cache_dir, '%s_%d_%d.fm' %
(genome.species.latin_name, genome.release, 2 if six.PY2 else 3))
def make_blastdb(url, name=None, filename=None, overwrite=False):
"""Download protein sequences and a make blast db. Uses datacache module."""
import datacache
cachedir = datacache.get_data_dir()
blastdb = os.path.join(cachedir, name)
if os.path.exists(blastdb+'.phr') and overwrite==False:
#print ('blast files found')
return blastdb
filename = datacache.fetch_file(url, filename=filename, decompress=True, subdir=None)
#print filename
cmd = 'makeblastdb -dbtype prot -in %s -out %s' %(filename,blastdb)
#print cmd
tmp=subprocess.check_output(cmd, shell=True)
return blastdb
def __init__(
self,
reference_name,
annotation_name,
annotation_version=None,
decompress_on_download=False,
copy_local_files_to_cache=False,
install_string_function=None,
cache_directory_path=None):
"""
Parameters
----------
reference_name : str
Name of reference genome
annotation_name : str
Name of annotation database
annotation_version : str or int, optional
Version or release of annotation database
decompress_on_download : bool, optional
If downloading a .fa.gz file, should we automatically expand it
into a decompressed FASTA file?
copy_local_files_to_cache : bool, optional
If file is on the local file system, should we still copy it
into the cache?
install_string_function : fn, optional
Function which returns an error message with
install instructions. If not provided then the error tells the
user what data is missing without install instructions.
cache_directory_path : str, optional
Where to place downloaded and temporary files, by default
inferred from reference name, annotation name, annotation version,
and the global cache directory determined by datacache.
"""
self.reference_name = reference_name
self.annotation_name = annotation_name
self.annotation_version = annotation_version
# using hidden member variable _cache_directory path since access to
# to the visible cache_directory_path (no underscore!) is combined
# with ensuring that the directpry actually exists
if cache_directory_path:
self._cache_directory_path = cache_directory_path
else:
self.cache_subdirectory = cache_subdirectory(
reference_name=reference_name,
annotation_name=annotation_name,
annotation_version=annotation_version)
# If `CACHE_DIR_ENV_KEY` is set, the cache will be saved there
self._cache_directory_path = datacache.get_data_dir(
subdir=self.cache_subdirectory,
envkey=CACHE_DIR_ENV_KEY)
self.decompress_on_download = decompress_on_download
self.copy_local_files_to_cache = copy_local_files_to_cache
self.install_string_function = install_string_function
