def _write_feature_qualifier(self, key, value=None, quote=None):
if value is None:
# Value-less entry like /pseudo
self.handle.write("%s/%s\n" % (self.QUALIFIER_INDENT_STR, key))
return
# Quick hack with no line wrapping, may be useful for testing:
# self.handle.write('%s/%s="%s"\n' % (self.QUALIFIER_INDENT_STR, key, value))
if quote is None:
# Try to mimic unwritten rules about when quotes can be left out:
if _is_int_or_long(value) or key in self.FTQUAL_NO_QUOTE:
quote = False
else:
quote = True
if quote:
line = '%s/%s="%s"' % (self.QUALIFIER_INDENT_STR, key, value)
else:
line = '%s/%s=%s' % (self.QUALIFIER_INDENT_STR, key, value)
if len(line) <= self.MAX_WIDTH:
self.handle.write(line + "\n")
return
while line.lstrip():
if len(line) <= self.MAX_WIDTH:
self.handle.write(line + "\n")
return
# Insert line break...
for index in range(min(len(line) - 1, self.MAX_WIDTH),
self.QUALIFIER_INDENT + 1, -1):
if line[index] == " ":
break
if line[index] != " ":
# No nice place to break...
index = self.MAX_WIDTH
assert index <= self.MAX_WIDTH
self.handle.write(line[:index] + "\n")
line = self.QUALIFIER_INDENT_STR + line[index:].lstrip()
def _write_feature_qualifier(self, key, value=None, quote=None):
if value is None:
# Value-less entry like /pseudo
self.handle.write("%s/%s\n" % (self.QUALIFIER_INDENT_STR, key))
return
# Quick hack with no line wrapping, may be useful for testing:
# self.handle.write('%s/%s="%s"\n' % (self.QUALIFIER_INDENT_STR, key, value))
if quote is None:
# Try to mimic unwritten rules about when quotes can be left out:
if _is_int_or_long(value) or key in self.FTQUAL_NO_QUOTE:
quote = False
else:
quote = True
if quote:
line = '%s/%s="%s"' % (self.QUALIFIER_INDENT_STR, key, value)
else:
line = '%s/%s=%s' % (self.QUALIFIER_INDENT_STR, key, value)
if len(line) <= self.MAX_WIDTH:
self.handle.write(line + "\n")
return
while line.lstrip():
if len(line) <= self.MAX_WIDTH:
self.handle.write(line + "\n")
return
# Insert line break...
for index in range(min(len(line) - 1, self.MAX_WIDTH),
self.QUALIFIER_INDENT + 1, -1):
if line[index] == " ":
break
if line[index] != " ":
# No nice place to break...
index = self.MAX_WIDTH
assert index <= self.MAX_WIDTH
self.handle.write(line[:index] + "\n")
line = self.QUALIFIER_INDENT_STR + line[index:].lstrip()
def add_entry(self, entry):
"""Add an Entry element to the pathway."""
# We insist that the node ID is an integer
assert _is_int_or_long(entry.id), \
"Node ID must be an integer, got %s (%s)" % (type(entry.id),
entry.id)
entry._pathway = self # Let the entry know about the pathway
self.entries[entry.id] = entry
def add_entry(self, entry):
"""Add an Entry element to the pathway."""
# We insist that the node ID is an integer
assert _is_int_or_long(entry.id), \
"Node ID must be an integer, got %s (%s)" % (type(entry.id),
entry.id)
entry._pathway = self # Let the entry know about the pathway
self.entries[entry.id] = entry
def remove_reaction(self, reaction):
"""Remove a Reaction element from the pathway."""
assert _is_int_or_long(reaction.id), \
"Node ID must be an integer, got %s (%s)" % (type(reaction.id),
reaction.id)
# We need to remove the reaction from any other elements that may
# contain it, which means removing those elements
# TODO
del self._reactions[reaction.id]
def remove_entry(self, entry):
"""Remove an Entry element from the pathway."""
assert _is_int_or_long(entry.id), \
"Node ID must be an integer, got %s (%s)" % (type(entry.id),
entry.id)
# We need to remove the entry from any other elements that may
# contain it, which means removing those elements
# TODO
del self.entries[entry.id]
def remove_entry(self, entry):
"""Remove an Entry element from the pathway."""
assert _is_int_or_long(entry.id), \
"Node ID must be an integer, got %s (%s)" % (type(entry.id),
entry.id)
# We need to remove the entry from any other elements that may
# contain it, which means removing those elements
# TODO
del self.entries[entry.id]
def remove_reaction(self, reaction):
"""Remove a Reaction element from the pathway."""
assert _is_int_or_long(reaction.id), \
"Node ID must be an integer, got %s (%s)" % (type(reaction.id),
reaction.id)
# We need to remove the reaction from any other elements that may
# contain it, which means removing those elements
# TODO
del self._reactions[reaction.id]
def add_reaction(self, reaction):
"""Add a Reaction element to the pathway."""
# We insist that the node ID is an integer and corresponds to an entry
assert _is_int_or_long(reaction.id), \
"Node ID must be an integer, got %s (%s)" % (type(reaction.id),
reaction.id)
assert reaction.id in self.entries, \
"Reaction ID %d has no corresponding entry" % reaction.id
reaction._pathway = self # Let the reaction know about the pathway
self._reactions[reaction.id] = reaction
def add_reaction(self, reaction):
"""Add a Reaction element to the pathway."""
# We insist that the node ID is an integer and corresponds to an entry
assert _is_int_or_long(reaction.id), \
"Node ID must be an integer, got %s (%s)" % (type(reaction.id),
reaction.id)
assert reaction.id in self.entries, \
"Reaction ID %d has no corresponding entry" % reaction.id
reaction._pathway = self # Let the reaction know about the pathway
self._reactions[reaction.id] = reaction
def _get_taxon_id_from_ncbi_lineage(self, taxonomic_lineage):
"""This is recursive! (PRIVATE).
taxonomic_lineage - list of taxonomy dictionaries from Bio.Entrez
First dictionary in list is the taxonomy root, highest would be the species.
Each dictionary includes:
- TaxID (string, NCBI taxon id)
- Rank (string, e.g. "species", "genus", ..., "phylum", ...)
- ScientificName (string)
(and that is all at the time of writing)
This method will record all the lineage given, returning the taxon id
(database key, not NCBI taxon id) of the final entry (the species).
"""
ncbi_taxon_id = taxonomic_lineage[-1]["TaxId"]
# Is this in the database already? Check the taxon table...
taxon_id = self.adaptor.execute_and_fetch_col0(
"SELECT taxon_id FROM taxon"
" WHERE ncbi_taxon_id=%s" % ncbi_taxon_id)
if taxon_id:
# we could verify that the Scientific Name etc in the database
# is the same and update it or print a warning if not...
if isinstance(taxon_id, list):
assert len(taxon_id) == 1
return taxon_id[0]
else:
return taxon_id
# We have to record this.
if len(taxonomic_lineage) > 1:
# Use recursion to find out the taxon id (database key) of the
# parent.
parent_taxon_id = self._get_taxon_id_from_ncbi_lineage(
taxonomic_lineage[:-1])
assert _is_int_or_long(parent_taxon_id), repr(parent_taxon_id)
else:
parent_taxon_id = None
# INSERT new taxon
rank = taxonomic_lineage[-1].get("Rank", None)
self.adaptor.execute(
"INSERT INTO taxon(ncbi_taxon_id, parent_taxon_id, node_rank)"
" VALUES (%s, %s, %s)", (ncbi_taxon_id, parent_taxon_id, rank))
taxon_id = self.adaptor.last_id("taxon")
assert isinstance(taxon_id, (int, long)), repr(taxon_id)
# ... and its name in taxon_name
scientific_name = taxonomic_lineage[-1].get("ScientificName", None)
if scientific_name:
self.adaptor.execute(
"INSERT INTO taxon_name(taxon_id, name, name_class)"
" VALUES (%s, %s, 'scientific name')", (taxon_id,
scientific_name[:255]))
return taxon_id
def _get_taxon_id_from_ncbi_lineage(self, taxonomic_lineage):
"""This is recursive! (PRIVATE).
taxonomic_lineage - list of taxonomy dictionaries from Bio.Entrez
First dictionary in list is the taxonomy root, highest would be the species.
Each dictionary includes:
- TaxID (string, NCBI taxon id)
- Rank (string, e.g. "species", "genus", ..., "phylum", ...)
- ScientificName (string)
(and that is all at the time of writing)
This method will record all the lineage given, returning the taxon id
(database key, not NCBI taxon id) of the final entry (the species).
"""
ncbi_taxon_id = taxonomic_lineage[-1]["TaxId"]
# Is this in the database already? Check the taxon table...
taxon_id = self.adaptor.execute_and_fetch_col0(
"SELECT taxon_id FROM taxon"
" WHERE ncbi_taxon_id=%s" % ncbi_taxon_id)
if taxon_id:
# we could verify that the Scientific Name etc in the database
# is the same and update it or print a warning if not...
if isinstance(taxon_id, list):
assert len(taxon_id) == 1
return taxon_id[0]
else:
return taxon_id
# We have to record this.
if len(taxonomic_lineage) > 1:
# Use recursion to find out the taxon id (database key) of the
# parent.
parent_taxon_id = self._get_taxon_id_from_ncbi_lineage(
taxonomic_lineage[:-1])
assert _is_int_or_long(parent_taxon_id), repr(parent_taxon_id)
else:
parent_taxon_id = None
# INSERT new taxon
rank = taxonomic_lineage[-1].get("Rank", None)
self.adaptor.execute(
"INSERT INTO taxon(ncbi_taxon_id, parent_taxon_id, node_rank)"
" VALUES (%s, %s, %s)", (ncbi_taxon_id, parent_taxon_id, rank))
taxon_id = self.adaptor.last_id("taxon")
assert isinstance(taxon_id, (int, long)), repr(taxon_id)
# ... and its name in taxon_name
scientific_name = taxonomic_lineage[-1].get("ScientificName", None)
if scientific_name:
self.adaptor.execute(
"INSERT INTO taxon_name(taxon_id, name, name_class)"
" VALUES (%s, %s, 'scientific name')", (taxon_id,
scientific_name[:255]))
return taxon_id
def add_reaction(self, reaction):
"""Add a Reaction element to the pathway."""
# We insist that the node ID is an integer and corresponds to an entry
if not _is_int_or_long(reaction.id):
raise ValueError("Node ID must be an integer, got %s (%s)" %
(type(reaction.id), reaction.id))
if reaction.id not in self.entries:
raise ValueError("Reaction ID %d has no corresponding entry" %
reaction.id)
reaction._pathway = self # Let the reaction know about the pathway
self._reactions[reaction.id] = reaction
def __getitem__(self, time):
"""Return a subset of signals or a single signal."""
if isinstance(time, slice):
# Fix the missing values in the slice
if time.start is None:
start = 0
else:
start = time.start
if time.stop is None:
stop = max(self.get_times())
else:
stop = time.stop
time = np.arange(start, stop, time.step)
return list(self._interpolate(time))
elif _is_int_or_long(time) or isinstance(time, float):
return self._interpolate(time)
raise ValueError('Invalid index')
def __getitem__(self, time):
"""Return a subset of signals or a single signal."""
if isinstance(time, slice):
# Fix the missing values in the slice
if time.start is None:
start = 0
else:
start = time.start
if time.stop is None:
stop = max(self.get_times())
else:
stop = time.stop
time = np.arange(start, stop, time.step)
return list(self._interpolate(time))
elif _is_int_or_long(time) or isinstance(time, float):
return self._interpolate(time)
raise ValueError('Invalid index')
def _get_taxon_id_from_ncbi_lineage(self, taxonomic_lineage):
"""This is recursive! (PRIVATE).
taxonomic_lineage - list of taxonomy dictionaries from Bio.Entrez
First dictionary in list is the taxonomy root, highest would be the species.
Each dictionary includes:
- TaxID (string, NCBI taxon id)
- Rank (string, e.g. "species", "genus", ..., "phylum", ...)
- ScientificName (string)
(and that is all at the time of writing)
This method will record all the lineage given, returning the taxon id
(database key, not NCBI taxon id) of the final entry (the species).
"""
ncbi_taxon_id = int(taxonomic_lineage[-1]["TaxId"])
left_value = None
right_value = None
parent_left_value = None
parent_right_value = None
# Is this in the database already? Check the taxon table...
rows = self.adaptor.execute_and_fetchall(
"SELECT taxon_id, left_value, right_value FROM taxon"
" WHERE ncbi_taxon_id=%s" % ncbi_taxon_id)
if rows:
# we could verify that the Scientific Name etc in the database
# is the same and update it or print a warning if not...
assert len(rows) == 1
return rows[0]
# We have to record this.
if len(taxonomic_lineage) > 1:
# Use recursion to find out the taxon id (database key) of the
# parent.
parent_taxon_id, parent_left_value, parent_right_value = self._get_taxon_id_from_ncbi_lineage(
taxonomic_lineage[:-1])
left_value = parent_right_value
right_value = parent_right_value + 1
assert _is_int_or_long(parent_taxon_id), repr(parent_taxon_id)
else:
# we have reached the top of the lineage but no current taxonomy
# id has been found
parent_taxon_id = None
left_value = self.adaptor.execute_one(
"SELECT MAX(left_value) FROM taxon")[0]
if not left_value:
left_value = 0
right_value = left_value + 1
self._update_left_right_taxon_values(left_value)
# INSERT new taxon
rank = str(taxonomic_lineage[-1].get("Rank"))
self.adaptor.execute(
"INSERT INTO taxon(ncbi_taxon_id, parent_taxon_id, node_rank, left_value, right_value)"
" VALUES (%s, %s, %s, %s, %s)",
(ncbi_taxon_id, parent_taxon_id, rank, left_value, right_value))
taxon_id = self.adaptor.last_id("taxon")
# assert isinstance(taxon_id, int), repr(taxon_id)
# ... and its name in taxon_name
scientific_name = taxonomic_lineage[-1].get("ScientificName")
if scientific_name:
self.adaptor.execute(
"INSERT INTO taxon_name(taxon_id, name, name_class)"
" VALUES (%s, %s, 'scientific name')",
(taxon_id, scientific_name[:255]))
return taxon_id, left_value, right_value
def _is_numeric(x):
return _py3k._is_int_or_long(x) or isinstance(x, (float, complex))
def _is_numeric(x):
"""Return True if is numeric."""
return _py3k._is_int_or_long(x) or isinstance(x, (float, complex))
def _is_numeric(x):
return _py3k._is_int_or_long(x) or isinstance(x, (float, complex))
def _get_taxon_id_from_ncbi_lineage(self, taxonomic_lineage):
"""This is recursive! (PRIVATE).
taxonomic_lineage - list of taxonomy dictionaries from Bio.Entrez
First dictionary in list is the taxonomy root, highest would be the species.
Each dictionary includes:
- TaxID (string, NCBI taxon id)
- Rank (string, e.g. "species", "genus", ..., "phylum", ...)
- ScientificName (string)
(and that is all at the time of writing)
This method will record all the lineage given, returning the taxon id
(database key, not NCBI taxon id) of the final entry (the species).
"""
ncbi_taxon_id = int(taxonomic_lineage[-1]["TaxId"])
left_value = None
right_value = None
parent_left_value = None
parent_right_value = None
# Is this in the database already? Check the taxon table...
rows = self.adaptor.execute_and_fetchall(
"SELECT taxon_id, left_value, right_value FROM taxon" " WHERE ncbi_taxon_id=%s" % ncbi_taxon_id
)
if rows:
# we could verify that the Scientific Name etc in the database
# is the same and update it or print a warning if not...
assert len(rows) == 1
return rows[0]
# We have to record this.
if len(taxonomic_lineage) > 1:
# Use recursion to find out the taxon id (database key) of the
# parent.
parent_taxon_id, parent_left_value, parent_right_value = self._get_taxon_id_from_ncbi_lineage(
taxonomic_lineage[:-1]
)
left_value = parent_right_value
right_value = parent_right_value + 1
assert _is_int_or_long(parent_taxon_id), repr(parent_taxon_id)
else:
# we have reached the top of the lineage but no current taxonomy
# id has been found
parent_taxon_id = None
left_value = self.adaptor.execute_one("SELECT MAX(left_value) FROM taxon")[0]
if not left_value:
left_value = 0
right_value = left_value + 1
self._update_left_right_taxon_values(left_value)
# INSERT new taxon
rank = str(taxonomic_lineage[-1].get("Rank"))
self.adaptor.execute(
"INSERT INTO taxon(ncbi_taxon_id, parent_taxon_id, node_rank, left_value, right_value)"
" VALUES (%s, %s, %s, %s, %s)",
(ncbi_taxon_id, parent_taxon_id, rank, left_value, right_value),
)
taxon_id = self.adaptor.last_id("taxon")
# assert isinstance(taxon_id, int), repr(taxon_id)
# ... and its name in taxon_name
scientific_name = taxonomic_lineage[-1].get("ScientificName")
if scientific_name:
self.adaptor.execute(
"INSERT INTO taxon_name(taxon_id, name, name_class)" " VALUES (%s, %s, 'scientific name')",
(taxon_id, scientific_name[:255]),
)
return taxon_id, left_value, right_value
def _is_numeric(x):
"""Return True if is numeric."""
return _py3k._is_int_or_long(x) or isinstance(x, (float, complex))
