def format_single_qualifiers(before, after):
"""
:param before:
:param after:
:return:
"""
for name in before.keys():
if name in GENBANK_SINGLE_QUALIFIERS:
after[name] = single(before, name, on_multiple_ignore=True)
return after
def format_single_qualifiers(before, after):
"""
:param before:
:param after:
:return:
"""
for name in before.keys():
if name in GENBANK_SINGLE_QUALIFIERS:
after[name] = single(before, name, on_multiple_ignore=True)
return after
def rename_label_to_note(before, after):
"""
The `/label=""` qualifier was discontinued in 2010, but is still used frequently.
See `GenBank Release 180 <http://www.ncbi.nlm.nih.gov/genbank/release/180/>`_.
:param before:
:param after:
:return:
"""
if 'label' in before:
# TODO support cases where both a note and a label exist
after['note'] = single(before, 'label')
return after
def rename_label_to_note(before, after):
"""
The `/label=""` qualifier was discontinued in 2010, but is still used frequently.
See `GenBank Release 180 <http://www.ncbi.nlm.nih.gov/genbank/release/180/>`_.
:param before:
:param after:
:return:
"""
if 'label' in before:
# TODO support cases where both a note and a label exist
after['note'] = single(before, 'label')
return after
def format_integer_qualifiers(before, after):
"""
Formats qualifiers that must be integers as ``int`` objects.
Removes any malformed qualifiers.
:param before:
:param after:
:return:
"""
for name in before.keys():
if name in GENBANK_INTEGER_QUALIFIERS:
try:
after[name] = int(single(before, name, on_multiple_ignore=True))
except ValueError:
del after[name]
return after
def format_integer_qualifiers(before, after):
"""
Formats qualifiers that must be integers as ``int`` objects.
Removes any malformed qualifiers.
:param before:
:param after:
:return:
"""
for name in before.keys():
if name in GENBANK_INTEGER_QUALIFIERS:
try:
after[name] = int(single(before, name,
on_multiple_ignore=True))
except ValueError:
del after[name]
return after
def remove_protein_id_and_add_to_xrefs(before, after):
"""
Protein IDs are codes such as "AAF19666.1", which come from "International collaborators" and should all be
on GenBank. This translation function removes the `/protein_id=""` qualifier and instead adds a `/db_xref=""`
qualifier.
Removes any malformed protein IDs as they are useless.
:param before:
:param after:
:return:
"""
if 'protein_id' in before:
protein_id = single(before, 'protein_id')
if RE_PROTEIN_ID.match(protein_id):
protein_xref = 'GenBank:{}'.format(protein_id)
after['db_xref'] = as_set(after.get('db_xref')) | {protein_xref}
del after['protein_id']
return after
def remove_protein_id_and_add_to_xrefs(before, after):
"""
Protein IDs are codes such as "AAF19666.1", which come from "International collaborators" and should all be
on GenBank. This translation function removes the `/protein_id=""` qualifier and instead adds a `/db_xref=""`
qualifier.
Removes any malformed protein IDs as they are useless.
:param before:
:param after:
:return:
"""
if 'protein_id' in before:
protein_id = single(before, 'protein_id')
if RE_PROTEIN_ID.match(protein_id):
protein_xref = 'GenBank:{}'.format(protein_id)
after['db_xref'] = as_set(after.get('db_xref')) | {protein_xref}
del after['protein_id']
return after
def convert_feature_type(feature):
"""
Finds a Sequence Ontology term for a GenBank feature.
This function requires a :class:`SeqFeature` as opposed to just a GenBank feature key, since the type of a GenBank
feature is not always fully described by its feature key. For example a `regulatory` GenBank feature could have
a `/regulatory_class="promoter"` qualifier.
:param SeqFeature feature:
:return: a Sequence Ontology term for the type of this feature
"""
type_ = feature.type
if type_ == 'regulatory':
regulatory_class = single(feature.qualifiers, 'regulatory_class', on_multiple_ignore=True)
if regulatory_class is None:
return GENBANK_REGULATORY_DEFAULT_SO_TERM
return GENBANK_REGULATORY_CLASS_SO_TERMS.get(regulatory_class, GENBANK_REGULATORY_DEFAULT_SO_TERM)
elif type_ == 'ncRNA':
nc_rna_class = single(feature.qualifiers, 'ncRNA_class', on_multiple_ignore=True)
if nc_rna_class is None:
return GENBANK_NC_RNA_DEFAULT_SO_TERM
return GENBANK_REGULATORY_CLASS_SO_TERMS.get(nc_rna_class, GENBANK_NC_RNA_DEFAULT_SO_TERM)
elif type_ == 'mobile_element':
mobile_element_type = single(feature.qualifiers, 'mobile_element_type', on_multiple_ignore=True).split(':')[0]
if mobile_element_type is None:
return GENBANK_MOBILE_ELEMENT_DEFAULT_SO_TERM
# TODO mobile elements can also have a /rpt_type="" qualifier
return GENBANK_MOBILE_ELEMENT_TYPE_SO_TERMS.get(mobile_element_type, GENBANK_MOBILE_ELEMENT_DEFAULT_SO_TERM)
elif type_ == 'repeat_region':
# TODO other features that can also have a /rpt_type="" qualifier: 'mobile_element', 'oriT', 'telomere'
repeat_type = single(feature.qualifiers, 'rpt_type', on_multiple_ignore=True)
if repeat_type is None:
return GENBANK_REPEAT_REGION_DEFAULT_SO_TERM
repeat_type = repeat_type.lower() # /rpt_type="" is case-insensitive
return GENBANK_REPEAT_TYPE_SO_TERMS.get(repeat_type, GENBANK_REPEAT_REGION_DEFAULT_SO_TERM)
else:
if 'pseudo' in feature.qualifiers:
# TODO /pseudo="" without /pseudogene="" is not well defined and there are no matching SO terms.
# "The qualifier /pseudo should be used to describe non-functional
# genes that are not formally described as pseudogenes, e.g. CDS
# has no translation due to other reasons than pseudogenisation events.
# Other reasons may include sequencing or assembly errors.
# In order to annotate pseudogenes the qualifier /pseudogene= must be
# used indicating the TYPE which can be taken from the INSDC controlled vocabulary
# for pseudogenes."
pass
# /pseudo and /pseudogene="" used with gene
if 'pseudogene' in feature.qualifiers:
pseudogene = single(feature.qualifiers, 'pseudogene', on_multiple_ignore=True)
if type_ == 'gene':
return GENBANK_PSEUDOGENE_TYPE_SO_TERMS.get(pseudogene, GENBANK_PSEUDOGENE_DEFAULT_SO_TERM)
# The /ribosomal_slippage qualifier is used with genes that have a translational frameshift.
if 'ribosomal_slippage' in feature.qualifiers:
if type_ == 'gene':
return 'gene_with_mRNA_with_frameshift'
elif type_ == 'mRNA':
return 'mRNA_with_frameshift'
# TODO CDS with frameshift, .. have no SO terms
# /trans_splicing is used on features such as CDS, mRNA and other features that are produced as
# a result of a trans-splicing event.
if 'trans_splicing' in feature.qualifiers:
if type_ == 'mRNA':
return 'trans_spliced_mRNA'
# TODO trans-spliced CDS, tRNA, .. have no SO terms
try:
return GENBANK_FEATURE_KEY_SO_TERMS[type_]
except KeyError:
pass
if type_ in SO_TERM_GENBANK_FEATURE_KEYS:
return type_
# TODO use a sequence ontology and allow any term that inherits from sequence_feature
try:
return UNAMBIGUOUS_INVALID_KEY_SO_TERMS[type_]
except KeyError:
return DEFAULT_SO_TERM
def convert_feature_type(feature):
"""
Finds a Sequence Ontology term for a GenBank feature.
This function requires a :class:`SeqFeature` as opposed to just a GenBank feature key, since the type of a GenBank
feature is not always fully described by its feature key. For example a `regulatory` GenBank feature could have
a `/regulatory_class="promoter"` qualifier.
:param SeqFeature feature:
:return: a Sequence Ontology term for the type of this feature
"""
type_ = feature.type
if type_ == 'regulatory':
regulatory_class = single(feature.qualifiers,
'regulatory_class',
on_multiple_ignore=True)
if regulatory_class is None:
return GENBANK_REGULATORY_DEFAULT_SO_TERM
return GENBANK_REGULATORY_CLASS_SO_TERMS.get(
regulatory_class, GENBANK_REGULATORY_DEFAULT_SO_TERM)
elif type_ == 'ncRNA':
nc_rna_class = single(feature.qualifiers,
'ncRNA_class',
on_multiple_ignore=True)
if nc_rna_class is None:
return GENBANK_NC_RNA_DEFAULT_SO_TERM
return GENBANK_REGULATORY_CLASS_SO_TERMS.get(
nc_rna_class, GENBANK_NC_RNA_DEFAULT_SO_TERM)
elif type_ == 'mobile_element':
mobile_element_type = single(feature.qualifiers,
'mobile_element_type',
on_multiple_ignore=True).split(':')[0]
if mobile_element_type is None:
return GENBANK_MOBILE_ELEMENT_DEFAULT_SO_TERM
# TODO mobile elements can also have a /rpt_type="" qualifier
return GENBANK_MOBILE_ELEMENT_TYPE_SO_TERMS.get(
mobile_element_type, GENBANK_MOBILE_ELEMENT_DEFAULT_SO_TERM)
elif type_ == 'repeat_region':
# TODO other features that can also have a /rpt_type="" qualifier: 'mobile_element', 'oriT', 'telomere'
repeat_type = single(feature.qualifiers,
'rpt_type',
on_multiple_ignore=True)
if repeat_type is None:
return GENBANK_REPEAT_REGION_DEFAULT_SO_TERM
repeat_type = repeat_type.lower() # /rpt_type="" is case-insensitive
return GENBANK_REPEAT_TYPE_SO_TERMS.get(
repeat_type, GENBANK_REPEAT_REGION_DEFAULT_SO_TERM)
else:
if 'pseudo' in feature.qualifiers:
# TODO /pseudo="" without /pseudogene="" is not well defined and there are no matching SO terms.
# "The qualifier /pseudo should be used to describe non-functional
# genes that are not formally described as pseudogenes, e.g. CDS
# has no translation due to other reasons than pseudogenisation events.
# Other reasons may include sequencing or assembly errors.
# In order to annotate pseudogenes the qualifier /pseudogene= must be
# used indicating the TYPE which can be taken from the INSDC controlled vocabulary
# for pseudogenes."
pass
# /pseudo and /pseudogene="" used with gene
if 'pseudogene' in feature.qualifiers:
pseudogene = single(feature.qualifiers,
'pseudogene',
on_multiple_ignore=True)
if type_ == 'gene':
return GENBANK_PSEUDOGENE_TYPE_SO_TERMS.get(
pseudogene, GENBANK_PSEUDOGENE_DEFAULT_SO_TERM)
# The /ribosomal_slippage qualifier is used with genes that have a translational frameshift.
if 'ribosomal_slippage' in feature.qualifiers:
if type_ == 'gene':
return 'gene_with_mRNA_with_frameshift'
elif type_ == 'mRNA':
return 'mRNA_with_frameshift'
# TODO CDS with frameshift, .. have no SO terms
# /trans_splicing is used on features such as CDS, mRNA and other features that are produced as
# a result of a trans-splicing event.
if 'trans_splicing' in feature.qualifiers:
if type_ == 'mRNA':
return 'trans_spliced_mRNA'
# TODO trans-spliced CDS, tRNA, .. have no SO terms
try:
return GENBANK_FEATURE_KEY_SO_TERMS[type_]
except KeyError:
pass
if type_ in SO_TERM_GENBANK_FEATURE_KEYS:
return type_
# TODO use a sequence ontology and allow any term that inherits from sequence_feature
try:
return UNAMBIGUOUS_INVALID_KEY_SO_TERMS[type_]
except KeyError:
return DEFAULT_SO_TERM