class Gene(BaseDS):
__tablename__ = 'Gene'
gene_id = Column(
String(
Constants.MAX_LEN_STRING,
collation=SQLCollationManager.get_instance().get_db_collation()))
chromosome = Column(
String(
15,
collation=SQLCollationManager.get_instance().get_db_collation()))
# Define the primary key and the unique constraint on the "primary key-like" attributes
__table_args__ = (PrimaryKeyConstraint('gene_id'), )
# Define the one-to-many relationships
DSTranscript_list = relationship('DSTranscript',
cascade='all',
backref='Gene')
GeneAlias_list = relationship('GeneAlias', cascade='all', backref='Gene')
## __eq__
# ------
#
# Tests the equality between two instances of this class.
# Two instances are considered equals if their primary keys are all equals.
#
# @param other: Gene - Another Gene object to compare to this object.
#
# @return Boolean - Are this object and 'other' equal?
#
def __eq__(self, other):
# Check if other object is of the same class
if (type(other) != type(self)):
return False
# Check if the two instances may be considered equal
elif (self.gene_id == other.gene_id):
return True
else:
return False
## __hash__
# --------
#
# Returns the hash value of a Gene object.
# The hash value of an instance is computed using its primary key.
#
# @return the hash value of the Gene object.
#
def __hash__(self):
return hash(self.gene_id)
class GeneAlias(BaseDS):
__tablename__ = 'GeneAlias'
gene_id = Column(
String(
Constants.MAX_LEN_STRING,
collation=SQLCollationManager.get_instance().get_db_collation()),
ForeignKey('Gene.gene_id', ondelete='CASCADE', onupdate='CASCADE'))
alias = Column(
String(
Constants.MAX_LEN_STRING,
collation=SQLCollationManager.get_instance().get_db_collation()))
crossref = Column(
String(
50,
collation=SQLCollationManager.get_instance().get_db_collation()))
# Define composite primary key
__table_args__ = (PrimaryKeyConstraint('gene_id', 'alias'), )
## __eq__
# ------
#
# Tests the equality between two instances of this class.
# Two instances are considered equals if their primary keys are all equals.
#
# @param other: GeneAlias - Another GeneAlias object to compare to this object.
#
# @return Boolean - Are this object and 'other' equal?
#
def __eq__(self, other):
# Check if other object is of the same class
if (type(other) != type(self)):
return False
# Check if the two instances may be considered equal
elif ((self.gene_id == other.gene_id) and (self.alias == other.alias)):
return True
else:
return False
## __hash__
# --------
#
# Returns the hash value of a GeneAlias object.
# The hash value of an instance is computed using its primary key.
#
# @return the hash value of the GeneAlias object.
#
def __hash__(self):
return hash((self.gene_id, self.alias))
class ORFAnnotationCatalog( BasePRO ):
__tablename__ = 'ORFAnnotationCatalog'
annotation = Column( String( 50, collation = SQLCollationManager.get_instance().get_db_collation() ),
primary_key = True )
family = Column( String( 50, collation = SQLCollationManager.get_instance().get_db_collation() ) )
# Define one-to-many relationship
ORFAnnotation_list = relationship( 'ORFAnnotation', cascade = 'all', backref = 'ORFAnnotationCatalog' )
class UTGeneFromAlias(BaseDS):
# =============================================================================
# /!\ This table is build from data in Gene and GeneAlias tables. It associates
# to each alias a list of genes that may correspond to it.
# =============================================================================
__tablename__ = 'UTGeneFromAlias'
alias = Column(
String(
Constants.MAX_LEN_STRING,
collation=SQLCollationManager.get_instance().get_db_collation()))
gene_ids = Column(
Text(collation=SQLCollationManager.get_instance().get_db_collation()))
# Define the primary key attributes
__table_args__ = (PrimaryKeyConstraint('alias'), )
## __eq__
# ------
#
# Tests the equality between two instances of this class.
# Two instances are considered equals if their primary keys are all equals.
#
# @param other: UTGeneFromAlias - Another UTGeneFromAlias object to compare to this object.
#
# @return Boolean - Are this object and 'other' equal?
#
def __eq__(self, other):
# Check if other object is of the same class
if (type(other) != type(self)):
return False
# Check if the two instances may be considered equal
elif (self.alias == other.alias):
return True
else:
return False
## __hash__
# --------
#
# Returns the hash value of a UTGeneFromAlias object.
# The hash value of an instance is computed using its primary key attribute.
#
# @return the hash value of the UTGeneFromAlias object.
#
def __hash__(self):
return hash(self.alias)
class Metadata(BaseDS):
__tablename__ = 'Metadata'
parameter = Column(
String(
255,
collation=SQLCollationManager.get_instance().get_db_collation()))
value = Column(
String(
255,
collation=SQLCollationManager.get_instance().get_db_collation()))
description = Column(
Text(collation=SQLCollationManager.get_instance().get_db_collation()))
# Define the primary key and the unique constraint on the "primary key-like" attributes
__table_args__ = (PrimaryKeyConstraint('parameter'), )
## __eq__
# ------
#
# Tests the equality between two instances of this class.
# Two instances are considered equals if their primary keys are equal.
#
# @param other: Metadata - Another Metadata object to compare to this object.
#
# @return Boolean - Are this object and 'other' equal?
#
def __eq__(self, other):
# Check if other object is of the same class
if (type(other) != type(self)):
return False
# Check if the two instances may be considered equal
elif (self.parameter == other.parameter):
return True
else:
return False
## __hash__
# --------
#
# Returns the hash value of a Metadata object.
# The hash value of an instance is computed using its primary key.
#
# @return the hash value of the Metadata object.
#
def __hash__(self):
return hash(self.parameter)
class ORFAnnotation( BasePRO ):
__tablename__ = 'ORFAnnotation'
orftranscriptasso_id = Column( Integer,
ForeignKey( 'ORFTranscriptAsso.id', ondelete='CASCADE', onupdate='CASCADE' ) )
orf_annotation = Column( String( 50, collation = SQLCollationManager.get_instance().get_db_collation() ),
ForeignKey( 'ORFAnnotationCatalog.annotation', ondelete='CASCADE', onupdate='CASCADE' ) )
criteria = Column( String( 50, collation = SQLCollationManager.get_instance().get_db_collation() ) )
# Define the composite primary key
__table_args__ = (
PrimaryKeyConstraint( 'orftranscriptasso_id', 'orf_annotation', 'criteria' ),
)
class PROMetadata(BasePRO):
__tablename__ = 'PROMetadata'
parameter = Column(String(
255, collation=SQLCollationManager.get_instance().get_db_collation()),
primary_key=True)
value = Column(
String(
255,
collation=SQLCollationManager.get_instance().get_db_collation()))
description = Column(
Text(collation=SQLCollationManager.get_instance().get_db_collation()))
## __eq__
# ------
#
# Tests the equality between two instances of this class.
# Two instances are considered equals if their primary keys are all equals.
#
# @param other: PROMetadata - Another PROMetadata object to compare to this object.
#
# @return Boolean - Are this object and 'other' equal?
#
def __eq__(self, other):
# Check if other object is of the same class
if (type(other) != type(self)):
return False
# Check if the two instances may be considered equal
elif (self.parameter == other.parameter):
return True
else:
return False
## __hash__
# --------
#
# Returns the hash value of a PROMetadata object.
# The hash value of an instance is computed using its primary key.
#
# @return the hash value of the PROMetadata object.
#
def __hash__(self):
return hash(self.parameter)
class UTBEDContent(BasePRO):
__tablename__ = 'UTBEDContent'
orf_id = Column(Integer,
ForeignKey('ORF.id',
ondelete='CASCADE',
onupdate='CASCADE'),
primary_key=True)
bed_index = Column(Integer)
bed_col = Column(
Text(SQLConstants.MAX_LEN_TEXT,
collation=SQLCollationManager.get_instance().get_db_collation()))
bed_add_col = Column(
Text(SQLConstants.MAX_LEN_TEXT,
collation=SQLCollationManager.get_instance().get_db_collation()))
class SpeciesCatalog(BaseDS):
__tablename__ = 'SpeciesCatalog'
name = Column(String(
50, collation=SQLCollationManager.get_instance().get_db_collation()),
primary_key=True)
class PROGeneAlias(BasePRO):
__tablename__ = 'PROGeneAlias'
gene_id = Column(
String(
Constants.MAX_LEN_STRING,
collation=SQLCollationManager.get_instance().get_db_collation()),
ForeignKey('PROGene.gene_id', ondelete='CASCADE', onupdate='CASCADE'))
alias = Column(
String(
Constants.MAX_LEN_STRING,
collation=SQLCollationManager.get_instance().get_db_collation()))
# Define composite primary key
__table_args__ = (PrimaryKeyConstraint('gene_id', 'alias'), )
class CellContext(BasePRO):
__tablename__ = 'CellContext'
orftranscriptasso_id = Column(
Integer,
ForeignKey('ORFTranscriptAsso.id',
ondelete='CASCADE',
onupdate='CASCADE'))
cell_context = Column(
String(
50,
collation=SQLCollationManager.get_instance().get_db_collation()),
ForeignKey('CellContextCatalog.context',
ondelete='CASCADE',
onupdate='CASCADE'))
# Define the composite primary key
__table_args__ = (PrimaryKeyConstraint('orftranscriptasso_id',
'cell_context'), )
## __eq__
# ------
#
# Tests the equality between two instances of this class.
# Two instances are considered equals if their "primary key-like" attributes
# (i.e. the attributes with unique constraint) are all equals.
#
# @param other: CellContext - Another CellContext object to compare to this object.
#
# @return Boolean - Are this object and 'other' equal?
#
def __eq__(self, other):
# Check if other object is of the same class
if (type(other) != type(self)):
return False
# Check if the two instances may be considered equal
elif ((self.orftranscriptasso_id == other.orftranscriptasso_id)
and (self.cell_context == other.cell_context)):
return True
else:
return False
## __hash__
# --------
#
# Returns the hash value of a CellContext object.
# The hash value of an instance is computed using its "primary key-like" attributes
# (i.e. the attributes with unique constraint).
#
# @return the hash value of the CellContext object.
#
def __hash__(self):
return hash((self.orftranscriptasso_id, self.cell_context))
class ORFTranscriptAssoDSAsso( BasePRO ):
__tablename__ = 'ORFTranscriptAssoDSAsso'
orftranscriptasso_id = Column( Integer,
ForeignKey( 'ORFTranscriptAsso.id', ondelete='CASCADE', onupdate='CASCADE' ) )
dsorftranscriptasso_id = Column( Integer )
data_source = Column( String( 50, collation = SQLCollationManager.get_instance().get_db_collation() ) )
# Define the composite primary key
__table_args__ = (
PrimaryKeyConstraint( 'orftranscriptasso_id', 'dsorftranscriptasso_id' ),
)
class ORFDSAsso(BasePRO):
__tablename__ = 'ORFDSAsso'
orf_id = Column(
Integer, ForeignKey('ORF.id', ondelete='CASCADE', onupdate='CASCADE'))
dsorf_id = Column(Integer)
data_source = Column(
String(
50,
collation=SQLCollationManager.get_instance().get_db_collation()))
ambiguous = Column(Boolean)
# Define unique constraint on "primary key-like attributes"
__table_args__ = (PrimaryKeyConstraint('orf_id', 'dsorf_id'), )
class UTRNABiotypeCatalog(BasePRO):
# =============================================================================
# /!\ This table is filled during the ComputeMissingInfo step.
# =============================================================================
__tablename__ = 'UTRNABiotypeCatalog'
biotype = Column(String(
255, collation=SQLCollationManager.get_instance().get_db_collation()),
primary_key=True)
## __eq__
# ------
#
# Tests the equality between two instances of this class.
# Two instances are considered equals if their primary keys are all equals.
#
# @param other: UTRNABiotypeCatalog - Another UTRNABiotypeCatalog object
# to compare to this object.
#
# @return Boolean - Are this object and 'other' equal?
#
def __eq__(self, other):
# Check if other object is of the same class
if (type(other) != type(self)):
return False
# Check if the two instances may be considered equal
elif (self.biotype == other.biotype):
return True
else:
return False
## __hash__
# --------
#
# Returns the hash value of a UTRNABiotypeCatalog object.
# The hash value of an instance is computed using its primary key attribute.
#
# @return the hash value of the UTRNABiotypeCatalog object.
#
def __hash__(self):
return hash(self.biotype)
class CellContextCatalog(BasePRO):
__tablename__ = 'CellContextCatalog'
context = Column(String(
50, collation=SQLCollationManager.get_instance().get_db_collation()),
primary_key=True)
# Define the one-to-many relationship
CellContext_list = relationship('CellContext',
cascade='all',
backref='CellContextCatalog')
## __eq__
# ------
#
# Tests the equality between two instances of this class.
# Two instances are considered equals if their primary keys are all equals.
#
# @param other: CellContextCatalog - Another CellContextCatalog object to compare to this object.
#
# @return Boolean - Are this object and 'other' equal?
#
def __eq__(self, other):
# Check if other object is of the same class
if (type(other) != type(self)):
return False
# Check if the two instances may be considered equal
elif (self.context == other.context):
return True
else:
return False
## __hash__
# --------
#
# Returns the hash value of a CellContextCatalog object.
# The hash value of an instance is computed using its primary key attribute.
#
# @return the hash value of the CellContextCatalog object.
#
def __hash__(self):
return hash(self.context)
class TranscriptDSAsso(BasePRO):
__tablename__ = 'TranscriptDSAsso'
transcript_id = Column(
Integer,
ForeignKey('Transcript.id', ondelete='CASCADE', onupdate='CASCADE'))
dstranscript_id = Column(Integer)
data_source = Column(
String(
50,
collation=SQLCollationManager.get_instance().get_db_collation()))
ambiguous = Column(Boolean)
# Define the composite primary key
__table_args__ = (PrimaryKeyConstraint('transcript_id',
'dstranscript_id'), )
class Transcript(BasePRO):
__tablename__ = 'Transcript'
id = Column(Integer)
transcript_id = Column(
String(
255,
collation=SQLCollationManager.get_instance().get_db_collation()))
transcript_name = Column(
String(
255,
collation=SQLCollationManager.get_instance().get_db_collation()))
gene_id = Column(
String(
Constants.MAX_LEN_STRING,
collation=SQLCollationManager.get_instance().get_db_collation()),
ForeignKey('PROGene.gene_id', ondelete='CASCADE', onupdate='CASCADE'))
strand = Column(
String(
2,
collation=SQLCollationManager.get_instance().get_db_collation()))
start_pos = Column(Integer)
end_pos = Column(Integer)
sequence = Column(
Text(SQLConstants.MAX_LEN_TEXT,
collation=SQLCollationManager.get_instance().get_db_collation()))
cds_start_pos = Column(Integer)
cds_stop_pos = Column(Integer)
cds_genomic_length = Column(Integer)
rel_cds_start_pos = Column(Integer)
rel_cds_stop_pos = Column(Integer)
rna_biotype = Column(
String(
100,
collation=SQLCollationManager.get_instance().get_db_collation()))
count_ds = Column(Integer)
count_ds_ambiguous = Column(Integer)
# Define the primary key and the unique constraint on the "primary key-like" attributes
__table_args__ = (
PrimaryKeyConstraint('id'),
UniqueConstraint('transcript_id', 'gene_id'),
)
# Define the one-to-many relationship
TranscriptDSAsso_list = relationship('TranscriptDSAsso',
cascade='all',
backref='Transcript')
ORFTranscriptAsso_list = relationship('ORFTranscriptAsso',
cascade='all',
backref='Transcript')
class FLOSSClass(BasePRO):
__tablename__ = 'FLOSSClass'
orftranscriptasso_id = Column(
Integer,
ForeignKey('ORFTranscriptAsso.id',
ondelete='CASCADE',
onupdate='CASCADE'))
floss_class = Column(
String(
100,
collation=SQLCollationManager.get_instance().get_db_collation()),
ForeignKey('FLOSSClassCatalog.floss_class',
ondelete='CASCADE',
onupdate='CASCADE'))
# Define composite primary key
__table_args__ = (PrimaryKeyConstraint('orftranscriptasso_id',
'floss_class'), )
class ProvidedCategory(BasePRO):
__tablename__ = 'ProvidedCategory'
orftranscriptasso_id = Column(
Integer,
ForeignKey('ORFTranscriptAsso.id',
ondelete='CASCADE',
onupdate='CASCADE'))
provided_category = Column(
String(
50,
collation=SQLCollationManager.get_instance().get_db_collation()),
ForeignKey('ProvidedCategoryCatalog.category',
ondelete='CASCADE',
onupdate='CASCADE'))
# Define the composite primary key
__table_args__ = (PrimaryKeyConstraint('orftranscriptasso_id',
'provided_category'), )
class ORFTranscriptAsso(BasePRO):
__tablename__ = 'ORFTranscriptAsso'
id = Column(Integer, autoincrement=True)
orf_id = Column(
Integer, ForeignKey('ORF.id', ondelete='CASCADE', onupdate='CASCADE'))
transcript_id = Column(
Integer,
ForeignKey('Transcript.id', ondelete='CASCADE', onupdate='CASCADE'))
rel_start_pos = Column(Integer)
rel_stop_pos = Column(Integer)
predicted = Column(Boolean)
ribo_seq = Column(Boolean)
ms_info = Column(Boolean)
start_codon_seq = Column(
String(
3,
collation=SQLCollationManager.get_instance().get_db_collation()))
start_flanking_seq = Column(
String(
20,
collation=SQLCollationManager.get_instance().get_db_collation()))
sequence_nt = Column(
Text(SQLConstants.MAX_LEN_TEXT,
collation=SQLCollationManager.get_instance().get_db_collation()))
sequence_aa = Column(
Text(SQLConstants.MAX_LEN_TEXT,
collation=SQLCollationManager.get_instance().get_db_collation()))
kozak_context = Column(Boolean)
kozak_context_comp = Column(
String(
30,
collation=SQLCollationManager.get_instance().get_db_collation()))
length_nt_min = Column(Integer)
length_nt_max = Column(Integer)
length_nt_median = Column(Float)
length_nt_values = Column(
Text(SQLConstants.MAX_LEN_TEXT,
collation=SQLCollationManager.get_instance().get_db_collation()))
length_aa_min = Column(Integer)
length_aa_max = Column(Integer)
length_aa_median = Column(Float)
length_aa_values = Column(
Text(SQLConstants.MAX_LEN_TEXT,
collation=SQLCollationManager.get_instance().get_db_collation()))
orf_score_min = Column(Float)
orf_score_max = Column(Float)
orf_score_median = Column(Float)
orf_score_values = Column(
Text(SQLConstants.MAX_LEN_TEXT,
collation=SQLCollationManager.get_instance().get_db_collation()))
phylocsf_min = Column(Float)
phylocsf_max = Column(Float)
phylocsf_median = Column(Float)
phylocsf_values = Column(
Text(SQLConstants.MAX_LEN_TEXT,
collation=SQLCollationManager.get_instance().get_db_collation()))
phastcons_min = Column(Float)
phastcons_max = Column(Float)
phastcons_median = Column(Float)
phastcons_values = Column(
Text(SQLConstants.MAX_LEN_TEXT,
collation=SQLCollationManager.get_instance().get_db_collation()))
floss_min = Column(Float)
floss_max = Column(Float)
floss_median = Column(Float)
floss_values = Column(
Text(SQLConstants.MAX_LEN_TEXT,
collation=SQLCollationManager.get_instance().get_db_collation()))
count_ds = Column(Integer)
count_computed_clusters = Column(Integer)
count_prov_lengths = Column(Integer)
computed_clusters = Column(
Text(SQLConstants.MAX_LEN_TEXT,
collation=SQLCollationManager.get_instance().get_db_collation()))
gen_len_eq_orf_len = Column(Boolean)
# Define the unique constraint on "primary key-like attributes"
__table_args__ = (
PrimaryKeyConstraint('id'),
UniqueConstraint('orf_id', 'transcript_id'),
)
# Define the one-to-many relationships
ORFTranscriptAssoDSAsso_list = relationship(
'ORFTranscriptAssoDSAsso',
cascade='all',
backref='ORFTranscriptAssoDSAsso')
CellContext_list = relationship('CellContext',
cascade='all',
backref='ORFTranscriptAssoDSAsso')
ProvidedCategory_list = relationship('ProvidedCategory',
cascade='all',
backref='ORFTranscriptAssoDSAsso')
FLOSSClass_list = relationship('FLOSSClass',
cascade='all',
backref='ORFTranscriptAssoDSAsso')
ORFCategory_list = relationship('ORFCategory',
cascade='all',
backref='ORFTranscriptAsso')
ORFAnnotation_list = relationship('ORFAnnotation',
cascade='all',
backref='ORFTranscriptAsso')
class DSORFTranscriptAsso(BaseDS):
__tablename__ = 'DSORFTranscriptAsso'
id = Column(Integer)
data_source = Column(
String(
50,
collation=SQLCollationManager.get_instance().get_db_collation()),
ForeignKey('DataSource.name', ondelete='CASCADE', onupdate='CASCADE'))
transcript_id = Column(
Integer,
ForeignKey('DSTranscript.id', ondelete='CASCADE', onupdate='CASCADE'))
uniq_orf_id = Column(
Integer, ForeignKey('DSORF.id', ondelete='CASCADE',
onupdate='CASCADE'))
predicted = Column(Boolean)
ribo_seq = Column(Boolean)
cell_context = Column(
String(
255,
collation=SQLCollationManager.get_instance().get_db_collation()))
orf_id = Column(
String(
255,
collation=SQLCollationManager.get_instance().get_db_collation()))
start_codon_seq = Column(
String(
3,
collation=SQLCollationManager.get_instance().get_db_collation()))
raw_sequence = Column(
Text(SQLConstants.MAX_LEN_TEXT,
collation=SQLCollationManager.get_instance().get_db_collation()))
raw_sequence_aa = Column(
Text(SQLConstants.MAX_LEN_TEXT,
collation=SQLCollationManager.get_instance().get_db_collation()))
kozak_context = Column(
String(
100,
collation=SQLCollationManager.get_instance().get_db_collation()))
orf_length_nt = Column(Integer)
orf_length = Column(Integer)
provided_category = Column(
String(
100,
collation=SQLCollationManager.get_instance().get_db_collation()))
ms_info = Column(Boolean)
orf_score = Column(Float)
phylocsf = Column(Float)
phastcons = Column(Float)
floss = Column(Float)
floss_class = Column(
String(
100,
collation=SQLCollationManager.get_instance().get_db_collation()))
gen_len_eq_orf_len = Column(Boolean)
# Define the primary key and the unique constraint on the "primary key-like" attributes
__table_args__ = (
PrimaryKeyConstraint('id'),
UniqueConstraint('data_source', 'transcript_id', 'uniq_orf_id',
'predicted', 'ribo_seq', 'cell_context'),
)
## __eq__
# ------
#
# Tests the equality between two instances of this class.
# Two instances are considered equals if their "primary key-like" attributes
# (i.e. the attributes with unique constraint) are all equals.
#
# @param other: DSORFTranscriptAsso - Another DSORFTranscriptAsso object to
# compare to this object.
#
# @return Boolean - Are this object and 'other' equal?
#
def __eq__(self, other):
# Check if other object is of the same class
if (type(other) != type(self)):
return False
# Check if the two instances may be considered equal
elif ((self.uniq_orf_id == other.uniq_orf_id)
and (self.transcript_id == other.transcript_id)
and (self.data_source == other.data_source)
and (self.cell_context == other.cell_context)
and (self.predicted == other.predicted)
and (self.ribo_seq == other.ribo_seq)):
return True
else:
return False
## __hash__
# --------
#
# Returns the hash value of a DSORFTranscriptAsso object.
# The hash value of an instance is computed using its "primary key-like" attributes
# (i.e. the attributes with unique constraint).
#
# @return the hash value of the DSTranscript object.
#
def __hash__(self):
return hash((self.data_source, self.transcript_id, self.uniq_orf_id,
self.predicted, self.ribo_seq, self.cell_context))
## update
# ------
#
# This method try to combine two instances of this class when they are "equals"
# (i.e. share the same values for "primary key-like attributes") by getting the
# missing attributes. It tests the equality between non-primary-key attributes
# of two equal instances of this class, and raise an DenCellORFObjUpdateException
# if DSORFTranscriptAsso compared have different values for their non-primary key
# attributes.
# If:
# - The self object is missing some values the other object has, then it updates
# the current object.
# - The information provided by both objects are contradictory, it logs a warning.
#
# @param other: DSORFTranscriptAsso - Another DSORFTranscriptAsso object "equal" to the
# current one.
#
# @throw DenCellORFObjUpdateException: When different values are found for the same attributes
# of two objects assumed to be the same.
#
def update(self, other):
# For each non-primary-key attribute, check if it is provided,
# try to update the current object by completing missing information
# using information provided by the other one and if there are attributes
# which have different values, raise an excpetion
attributes_to_check = [
'orf_id', 'raw_sequence', 'raw_sequence_aa', 'start_codon_seq',
'kozak_context', 'orf_length_nt', 'orf_length',
'provided_category', 'ms_info', 'orf_score', 'phylocsf',
'phastcons', 'floss', 'floss_class', 'gen_len_eq_orf_len'
]
# List of the attributes for which the value should be set to
# None if they are found different
att_to_none_if_conflict = [
'start_codon_seq', 'kozak_context', 'orf_length', 'orf_length_nt',
'ms_info', 'orf_score', 'phylocsf', 'phastcons', 'floss'
]
# List of the attributes for which the value should be set to
# 'AMBIGUOUS' if they are found different
att_to_flag_if_conflict = ['provided_category', 'floss_class']
# List of the attributes storing sequences
att_sequences = ['raw_sequence', 'raw_sequence_aa']
# Initialize a dictionary attributes for which there is a conflict
conflicting_att = {}
for att in attributes_to_check:
self_att = getattr(self, att)
other_att = getattr(other, att)
self_att_is_empty = GeneralUtil.is_empty(self_att)
other_att_is_empty = GeneralUtil.is_empty(other_att)
# If the other attribute is provided
if (not other_att_is_empty):
# and if the self attribute is empty, fill it using the information of the other object
if self_att_is_empty:
setattr(self, att, other_att)
# and if the self attribute is filled and the two values of the attribute
# are different, reset the value to 'AMBIGUOUS' or None or compute a consensus
# for the sequences and keep track of this update
# attribute and its values
elif ((not self_att_is_empty) and (other_att != self_att)):
if (att in att_to_flag_if_conflict):
setattr(self, att,
Constants.DENCELLORFOBJ_AMBIGUOUS_ATT)
conflicting_att[att] = (self_att, other_att,
getattr(self, att))
elif (att in att_sequences):
if (att == 'raw_sequence'):
sqce_type = Constants.SEQUENCE_TYPE_DNA
else:
sqce_type = Constants.SEQUENCE_TYPE_PROT
sqce_list = [self_att, other_att]
# Compute the consensus (the threshold is set to 1 as the same
# DSORFTranscriptAsso may be updated several time but sequentially.
# Setting this threshold to 1 allow to always get a N/X as soon as
# there is an ambiguity for one of the nucleotide / amino acid.
sqce_consensus = GeneticsUtil.find_sqce_consensus(
list_of_sequences=sqce_list,
sqce_type=sqce_type,
threshold=1)
setattr(self, att, sqce_consensus)
conflicting_att[att] = (self_att, other_att,
getattr(self, att))
elif (att in att_to_none_if_conflict):
setattr(self, att, None)
conflicting_att[att] = (self_att, other_att,
getattr(self, att))
# If there are attributes for which there is a conflict, raise a DenCellORFObjUpdateException
if (conflicting_att != {}):
except_message = []
for (att, val) in conflicting_att.items():
except_message.append('the attribute ' + att +
' has been found with the values "' +
str(val[0]) + '" and "' + str(val[1]) +
'",' +
' hence its value has been set to "' +
str(val[2]) + '"')
except_message = ', '.join(except_message) + '.'
raise DenCellORFObjUpdateException(except_message)
class UTDSTranscriptGeneConflict(BaseDS):
# =============================================================================
# /!\ This table is build during the insertion of data. It associates to each
# (transcript_id, data_source) unique pair for which a conflict about the
# Gene entry to which it is related, the list of gene IDs. This allows to
# manage the cases where conflicts about the gene are found several times
# for a particular DSTranscript entry.
# =============================================================================
__tablename__ = 'UTDSTranscriptGeneConflict'
transcript_id = Column(
String(
255,
collation=SQLCollationManager.get_instance().get_db_collation()))
data_source = Column(
String(
50,
collation=SQLCollationManager.get_instance().get_db_collation()))
gene_ids = Column(
Text(collation=SQLCollationManager.get_instance().get_db_collation()))
# Define the primary key attribute
__table_args__ = (PrimaryKeyConstraint('transcript_id', 'data_source'), )
## __eq__
# ------
#
# Tests the equality between two instances of this class.
# Two instances are considered equals if their primary keys are all equals.
#
# @param other: UTDSTranscriptGeneConflict - Another UTDSTranscriptGeneConflict
# object to compare to this object.
#
# @return Boolean - Are this object and 'other' equal?
#
def __eq__(self, other):
# Check if other object is of the same class
if (type(other) != type(self)):
return False
# Check if the two instances may be considered equal
elif ((self.transcript_id == other.transcript_id)
and (self.data_source == other.data_source)):
return True
else:
return False
## __hash__
# --------
#
# Returns the hash value of a UTDSTranscriptGeneConflict object.
# The hash value of an instance is computed using its primary key attributes.
#
# @return the hash value of the UTDSTranscriptGeneConflict object.
#
def __hash__(self):
return hash((self.transcript_id, self.data_source))
class DataSource(BaseDS):
__tablename__ = 'DataSource'
name = Column(
String(
50,
collation=SQLCollationManager.get_instance().get_db_collation()))
doi = Column(
String(
50,
collation=SQLCollationManager.get_instance().get_db_collation()))
description = Column(
Text(collation=SQLCollationManager.get_instance().get_db_collation()))
url = Column(
String(
255,
collation=SQLCollationManager.get_instance().get_db_collation()))
annotation_version = Column(
String(
10,
collation=SQLCollationManager.get_instance().get_db_collation()))
ensembl_release = Column(
String(
3,
collation=SQLCollationManager.get_instance().get_db_collation()))
annotation_description = Column(Text)
# Define the primary key attribute
__table_args__ = (PrimaryKeyConstraint('name'), )
# Define the one-to-many relationships
DSORF_list = relationship('DSORF', cascade='all', backref='DataSource')
DSTranscript_list = relationship('DSTranscript',
cascade='all',
backref='DataSource')
DSORFTranscriptAsso_list = relationship('DSORFTranscriptAsso',
cascade='all',
backref='DataSource')
## __eq__
# ------
#
# Tests the equality between two instances of this class.
# Two instances are considered equals if their "primary key-like" attributes
# (i.e. the attributes with unique constraint) are all equals.
#
# @param other: DataSource - Another DataSource object to compare to this object.
#
# @return Boolean - Are this object and 'other' equal?
#
def __eq__(self, other):
# Check if other object is of the same class
if (type(other) != type(self)):
return False
# Check if the two instances may be considered equal
elif (self.name == other.name):
return True
else:
return False
## __hash__
# --------
#
# Returns the hash value of a DataSource object.
# The hash value of an instance is computed using its "primary key-like" attributes
# (i.e. the attributes with unique constraint).
#
# @return the hash value of the DataSource object.
#
def __hash__(self):
return hash(self.name)
class ORF(BasePRO):
__tablename__ = 'ORF'
id = Column(Integer)
chromosome = Column(
String(
15,
collation=SQLCollationManager.get_instance().get_db_collation()))
strand = Column(
String(
2,
collation=SQLCollationManager.get_instance().get_db_collation()))
start_pos = Column(Integer)
stop_pos = Column(Integer)
spliced = Column(Boolean)
spliced_parts_count = Column(Integer)
splice_starts = Column(
String(
Constants.MAX_LEN_STRING,
collation=SQLCollationManager.get_instance().get_db_collation()))
splice_ends = Column(
String(
Constants.MAX_LEN_STRING,
collation=SQLCollationManager.get_instance().get_db_collation()))
sequence = Column(
Text(SQLConstants.MAX_LEN_TEXT,
collation=SQLCollationManager.get_instance().get_db_collation()))
sequence_aa = Column(
Text(SQLConstants.MAX_LEN_TEXT,
collation=SQLCollationManager.get_instance().get_db_collation()))
genomic_length = Column(Integer)
count_ds = Column(Integer)
count_ds_ambiguous = Column(Integer)
# Define the primary key and the unique constraint on the "primary key-like" attributes
__table_args__ = (
PrimaryKeyConstraint('id'),
UniqueConstraint('chromosome', 'strand', 'start_pos', 'stop_pos',
'spliced', 'splice_starts', 'splice_ends',
'spliced_parts_count'),
)
# Define one-to-many relationship
ORFDSAsso_list = relationship('ORFDSAsso', cascade='all', backref='ORF')
ORFTranscriptAsso_list = relationship('ORFTranscriptAsso',
cascade='all',
backref='ORF')
UTBEDContent = relationship('UTBEDContent',
cascade='all',
backref='ORF',
uselist=False)
## __eq__
# ------
#
# Tests the equality between two instances of this class.
# Two instances are considered equals if their "primary key-like" attributes
# (i.e. the attributes with unique constraint) are all equals.
#
# @param other: ORF - Another ORF object to compare to this object.
#
# @return Boolean - Are this object and 'other' equal?
#
def __eq__(self, other):
# Check if other object is of the same class
if (type(other) != type(self)):
return False
# Check if the two instances may be considered equal
elif ((self.id == other.id)
or ((self.chromosome == other.chromosome) and
(self.strand == other.strand) and
(self.start_pos == other.start_pos) and
(self.stop_pos == other.stop_pos) and
(self.spliced == other.spliced) and
(self.spliced_parts_count == other.spliced_parts_count) and
(self.splice_starts == other.splice_starts) and
(self.splice_ends == other.splice_ends))):
return True
else:
return False
## __hash__
# --------
#
# Returns the hash value of a ORF object.
# The hash value of an instance is computed using its "primary key-like" attributes
# (i.e. the attributes with unique constraint).
#
# @return the hash value of the ORF object.
#
def __hash__(self):
return hash((self.chromosome, self.strand, self.start_pos,
self.stop_pos, self.spliced, self.spliced_parts_count,
self.splice_starts, self.splice_ends))
class DSTranscript( BaseDS ):
__tablename__ = 'DSTranscript'
id = Column( Integer )
transcript_id = Column( String( 255, collation = SQLCollationManager.get_instance().get_db_collation() ) )
data_source = Column( String( 50, collation = SQLCollationManager.get_instance().get_db_collation() ),
ForeignKey( 'DataSource.name', ondelete='CASCADE', onupdate='CASCADE' ) )
gene_id = Column( String( Constants.MAX_LEN_STRING, collation = SQLCollationManager.get_instance().get_db_collation() ),
ForeignKey( 'Gene.gene_id', ondelete='CASCADE', onupdate='CASCADE' ) )
raw_strand = Column( String( 2, collation = SQLCollationManager.get_instance().get_db_collation() ) )
raw_start_pos = Column( Integer )
raw_end_pos = Column( Integer )
strand = Column( String( 2, collation = SQLCollationManager.get_instance().get_db_collation() ) )
start_pos = Column( Integer )
end_pos = Column( Integer )
raw_cds_start_pos = Column( Integer )
raw_cds_stop_pos = Column( Integer )
cds_start_pos = Column( Integer )
cds_stop_pos = Column( Integer )
rna_biotype = Column( String( 255, collation=SQLCollationManager.get_instance().get_db_collation() ) )
# Define the primary key and the unique constraint on the "primary key-like" attributes
__table_args__ = (
PrimaryKeyConstraint( 'id' ),
UniqueConstraint( 'transcript_id', 'data_source' ),
)
# Define the one-to-many relationship
DSORFTranscriptAsso_list = relationship( 'DSORFTranscriptAsso', cascade = 'all', backref = 'DSTranscript' )
## __eq__
# ------
#
# Tests the equality between two instances of this class.
# Two instances are considered equals if their "primary key-like" attributes
# (i.e. the attributes with unique constraint) are all equals.
#
# @param other: DSTranscript - Another DSTranscript object to compare to this object.
#
# @return Boolean - Are this object and the other one equal?
#
def __eq__( self, other ):
# Check if other object is of the same class
if ( type( other ) != type( self ) ):
return False
# Check if the transcript ID and the source are the same
elif ( ( self.transcript_id == other.transcript_id )
and ( self.data_source == other.data_source ) ):
return True
else:
return False
## __hash__
# --------
#
# Returns the hash value of a DSTranscript object.
# The hash value of an instance is computed using its "primary key-like" attributes
# (i.e. the attributes with unique constraint).
#
# @return the hash value of the DSTranscript object.
#
def __hash__( self ):
return hash( ( self.transcript_id, self.data_source ) )
## update
# ------
#
# Try to combine two instances of this class when they are "equals"
# (i.e. share the same values for "primary key-like attributes") by getting the
# missing attributes. It tests the equality between non-primary-key attributes
# of two equal instances of this class, and raise an DenCellORFObjUpdateException
# if DSTranscript compared have different values for their non-primary key
# attributes.
# If:
# - The self object is missing some values the other object has, then it updates
# the current object.
# - The information provided by both objects are contradictory, it logs a warning.
#
# @param other: DSTranscript - Another DSTranscript object "equal" to the current one.
# @param gene_id_conflicts: List - The list of all gene IDs that have been found associated
# to the DSTranscript entry. None by default.
#
# @throw DenCellORFTrUpdateException: When different values are found for the same attributes
# of two objects assumed to be the same.
#
def update( self, other, gene_id_conflicts=None ):
# For each non-primary-key attribute, check if it is provided,
# try to update the current object by completing missing information
# using information provided by the other one and if there are attributes
# which have different values, raise an excpetion
attributes_to_check = [ 'raw_strand', 'raw_start_pos', 'raw_end_pos',
'raw_cds_start_pos', 'raw_cds_stop_pos', 'rna_biotype' ]
# List of the attributes for which the value should be set to
# 'AMBIGUOUS' if they are found different
att_to_flag_if_conflict = [ 'rna_biotype' ]
# Initialize a dictionary attributes for which there is a conflict
conflicting_att = {}
# Check the gene IDs. If the two gene IDs are different, create a new ID
# by concatenating the names of the two genes.
self_gene = self.gene_id
other_gene = other.gene_id
if ( self_gene != other_gene ):
init_gene_id_conflicts = gene_id_conflicts
# If a conflict has already been registered for this DSTranscript,
# then get the list of genes associated to it.
# Otherwise, register this first conflict.
if gene_id_conflicts:
# Get the name of the gene which does not start with the CONFLICT prefix
if ( ( not self_gene.startswith( Constants.PREFIX_CONFLICT_GENE_TRANSCRIPT ) )
and ( self_gene not in gene_id_conflicts ) ):
gene_id_conflicts.append( self_gene )
if ( ( not other_gene.startswith( Constants.PREFIX_CONFLICT_GENE_TRANSCRIPT ) )
and ( other_gene not in gene_id_conflicts ) ):
gene_id_conflicts.append( other_gene )
else:
# Register this new conflict and order the gene IDs
gene_id_conflicts = [ self_gene, other_gene ]
# Order the IDs in order to always get the same new ID when the same gene IDs are conflicting
gene_id_conflicts.sort()
# Register the new list of conflict on gene IDs only if it has been updated
if ( gene_id_conflicts != init_gene_id_conflicts ):
# Create a new "conflict" id
new_gene_id = Constants.PREFIX_CONFLICT_GENE_TRANSCRIPT + '_'.join( gene_id_conflicts )
self.gene_id = new_gene_id
conflicting_att[ 'gene_id' ] = ( self_gene, other_gene, self.gene_id )
# Check all the other non-primary-key attributes
for att in attributes_to_check:
self_att = getattr( self, att )
other_att = getattr( other, att )
# If the attribute treated is the raw strand, the '-' value
# should not be considered as missing value
if ( att == 'raw_strand' ):
self_att_is_empty = GeneralUtil.is_empty( val = self_att,
empty_val = Constants.EMPTY_VALUES_WO_DASH )
other_att_is_empty = GeneralUtil.is_empty( val = other_att,
empty_val = Constants.EMPTY_VALUES_WO_DASH )
else:
self_att_is_empty = GeneralUtil.is_empty( self_att )
other_att_is_empty = GeneralUtil.is_empty( other_att )
# If the other attribute is provided
if ( not other_att_is_empty ):
# and if the self attribute is empty, fill it using the information of the other object
if self_att_is_empty:
setattr( self, att, other_att )
# and if the self attribute is filled and the two values of the attribute are different,
# reset the value to 'AMBIGUOUS' and keep track of this attribute and its values
elif ( ( not self_att_is_empty ) and ( other_att != self_att ) ):
if att in att_to_flag_if_conflict:
setattr( self, att, Constants.DENCELLORFOBJ_AMBIGUOUS_ATT )
else:
setattr( self, att, None )
conflicting_att[ att ] = ( self_att, other_att, getattr(self, att) )
# If there are attributes for which there is a conflict,
# raise a DenCellORFTrUpdateException
if conflicting_att != {}:
except_message = []
for (att, val) in conflicting_att.items():
except_message.append( 'the attribute ' + att + ' has been found with the values "' +
str( val[0] ) + '" and "' + str( val[1] ) + '",' +
' hence its value has been set to "' + str( val[2] ) + '"' )
except_message = ', '.join( except_message ) + '.'
# If a conflict has been found regarding the gene ID, raise the exception
# with the value of the new gene ID to create and to associate to this transcript
if ( 'gene_id' in conflicting_att.keys() ):
raise DenCellORFTrUpdateException( message = except_message,
gene_id = self.gene_id,
gene_id_conflicts = gene_id_conflicts )
else:
raise DenCellORFTrUpdateException( message = except_message )
class DSORF(BaseDS):
__tablename__ = 'DSORF'
id = Column(Integer)
data_source = Column(
String(
50,
collation=SQLCollationManager.get_instance().get_db_collation()),
ForeignKey('DataSource.name', ondelete='CASCADE', onupdate='CASCADE'))
chromosome = Column(
String(
15,
collation=SQLCollationManager.get_instance().get_db_collation()))
raw_strand = Column(
String(
2,
collation=SQLCollationManager.get_instance().get_db_collation()))
raw_start_pos = Column(Integer)
raw_stop_pos = Column(Integer)
spliced = Column(Boolean)
raw_splice_starts = Column(
String(
Constants.MAX_LEN_STRING,
collation=SQLCollationManager.get_instance().get_db_collation()))
raw_splice_ends = Column(
String(
Constants.MAX_LEN_STRING,
collation=SQLCollationManager.get_instance().get_db_collation()))
spliced_parts_count = Column(Integer)
strand = Column(
String(
2,
collation=SQLCollationManager.get_instance().get_db_collation()))
start_pos = Column(Integer)
stop_pos = Column(Integer)
splice_starts = Column(
String(
Constants.MAX_LEN_STRING,
collation=SQLCollationManager.get_instance().get_db_collation()))
splice_ends = Column(
String(
Constants.MAX_LEN_STRING,
collation=SQLCollationManager.get_instance().get_db_collation()))
raw_genomic_length = Column(Integer)
genomic_length = Column(Integer)
genomic_length_diff = Column(Integer)
liftover_succeed = Column(Boolean)
# Define the primary key and the unique constraint on the "primary key-like" attributes
__table_args__ = (
PrimaryKeyConstraint('id'),
UniqueConstraint('data_source', 'chromosome', 'raw_strand',
'raw_start_pos', 'raw_stop_pos', 'spliced',
'raw_splice_starts', 'raw_splice_ends',
'spliced_parts_count'),
)
# Define the one-to-many relationship
DSORFTranscriptAsso_list = relationship('DSORFTranscriptAsso',
cascade='all',
backref='DSORF')
## __eq__
# ------
#
# Tests the equality between two instances of this class.
# Two instances are considered equals if their "primary key-like" attributes
# (i.e. the attributes with unique constraint) are all equals.
#
# @param other: DSORF - Another DSORF object to compare to this object.
#
# @return Boolean - Are this object and 'other' equal?
#
def __eq__(self, other):
# Check if other object is of the same class
if (type(other) != type(self)):
return False
# Check if the two instances may be considered equal
elif ((self.raw_start_pos == other.raw_start_pos)
and (self.raw_stop_pos == other.raw_stop_pos)
and (self.raw_strand == other.raw_strand)
and (self.chromosome == other.chromosome)
and (self.spliced == other.spliced)
and (self.raw_splice_starts == other.raw_splice_starts)
and (self.raw_splice_ends == other.raw_splice_ends)
and (self.spliced_parts_count == other.spliced_parts_count)
and (self.data_source == other.data_source)):
return True
else:
return False
## __hash__
# --------
#
# Returns the hash value of a DSORF object.
# The hash value of an instance is computed using its "primary key-like" attributes
# (i.e. the attributes with unique constraint).
#
# @return the hash value of the DSORF object.
#
def __hash__(self):
return hash((self.data_source, self.chromosome, self.raw_strand,
self.raw_start_pos, self.raw_stop_pos, self.spliced,
self.raw_splice_starts, self.raw_splice_ends,
self.spliced_parts_count))
try:
# Set the level of verbosity of the main Logger
verbosity_level = OptionManager.get_instance().get_option(
OptionConstants.OPTION_VERBOSITY)
if (verbosity_level not in Constants.LOG_MODES.keys()):
Logger.get_instance().critical(
'The level of verbosity selected is incorrect.' +
' It must be one of: ' +
', '.join(Constants.LOG_MODES.keys()) +
'. Please see the documentation for more information.')
Logger.get_instance().set_mode(Constants.LOG_MODES[verbosity_level])
# Get the type of database, in order to set the appropriate collation for strings
SQLCollationManager.get_instance().set_db_collation(
OptionManager.get_instance().get_option(
OptionConstants.OPTION_DB_TYPE))
# NB: The execution module is only imported after this step in order to allow a
# "dynamic" creation of the model (classes inheriting from SQLAlchemy Base).
# Indeed, once the model has been loaded, this is not possible to update the
# collation of the columns.
from fr.tagc.uorf.core.execution import *
# Instantiate DenCellORF
DenCellORF = DenCellORF()
# Execute the strategy
DenCellORF.execute()
except Exception as e:
Logger.get_instance().critical(
