def LowSimilarityRegionCodingBlockGraph2GeneTreeGraph(cbg):
"""
Convert LowSimilarityRegion 2 GeneTree
@attention: function just converts, error check is not performed here!
@type cbg: LowSimilarityRegion
@param cbg: LowSimilarityRegion instance
@rtype: GeneTreeGraph
@return: GeneTreeGraph instance
"""
gtg = GeneTreeGraph()
cbgnode2orgnode = {}
for node in cbg.get_nodes():
org = cbg._organism_from_node(node)
gtg.add_node(org)
# add node/org combi to mapping dict
cbgnode2orgnode[node] = org
# now add all the edges
omsr = cbg.overall_minimal_spanning_range()
for (n1, n2) in cbg.pairwisecrosscombinations_node():
if cbg.has_edge(n1, n2):
# get pacbp(orf) object
pacbps = cbg.get_pacbps_by_nodes(node1=n1, node2=n2)
if pacbps:
identityscore = pacbps[0].identityscore
else:
# this edge has no pacbp in the lsrCBG -> happens often
identityscore = 0.0
else:
# this edge is absent in the lsrCBG!
identityscore = 0.0
# get organism identifyers from node and add edge
o1, o2 = cbgnode2orgnode[n1], cbgnode2orgnode[n2]
gtg.add_edge(o1, o2, wt=identityscore)
# check if the graph is saturated (complete)
# if not (organism/node/orf missing), add this as a zero-wt edge
gtg.makecompletegraph(wt=0.0)
# and return this new genetree graph
return gtg
def LowSimilarityRegionCodingBlockGraph2GeneTreeGraph(cbg):
"""
Convert LowSimilarityRegion 2 GeneTree
@attention: function just converts, error check is not performed here!
@type cbg: LowSimilarityRegion
@param cbg: LowSimilarityRegion instance
@rtype: GeneTreeGraph
@return: GeneTreeGraph instance
"""
gtg = GeneTreeGraph()
cbgnode2orgnode = {}
for node in cbg.get_nodes():
org = cbg._organism_from_node(node)
gtg.add_node(org)
# add node/org combi to mapping dict
cbgnode2orgnode[ node ] = org
# now add all the edges
omsr = cbg.overall_minimal_spanning_range()
for (n1,n2) in cbg.pairwisecrosscombinations_node():
if cbg.has_edge(n1,n2):
# get pacbp(orf) object
pacbps = cbg.get_pacbps_by_nodes(node1=n1,node2=n2)
if pacbps:
identityscore = pacbps[0].identityscore
else:
# this edge has no pacbp in the lsrCBG -> happens often
identityscore = 0.0
else:
# this edge is absent in the lsrCBG!
identityscore = 0.0
# get organism identifyers from node and add edge
o1,o2 = cbgnode2orgnode[ n1 ], cbgnode2orgnode[ n2 ]
gtg.add_edge( o1, o2, wt=identityscore )
# check if the graph is saturated (complete)
# if not (organism/node/orf missing), add this as a zero-wt edge
gtg.makecompletegraph(wt=0.0)
# and return this new genetree graph
return gtg
def CodingBlockGraph2GeneTreeGraph(cbg):
"""
Convert CodingBlockGraph 2 GeneTree
@attention: function just converts, error check is not performed here!
@type cbg: CodingBlockGraph
@param cbg: CodingBlockGraph instance
@rtype: GeneTreeGraph
@return: GeneTreeGraph instance
"""
gtg = GeneTreeGraph()
cbgnode2orgnode = {}
for node in cbg.get_nodes():
org = cbg._organism_from_node(node)
gtg.add_node(org)
# add node/org combi to mapping dict
cbgnode2orgnode[ node ] = org
# now add all the edges
omsr = cbg.overall_minimal_spanning_range()
for (n1,n2) in cbg.pairwisecrosscombinations_node():
if cbg.has_edge(n1,n2):
# get pacbp(orf) object
thepacbp = cbg.get_pacbps_by_nodes(node1=n1,node2=n2)[0]
# get relative coordinates of the OMSR part of the alignment
omsrQs = thepacbp.alignmentposition_by_query_pos( min( omsr[n1] ) )
omsrQe = thepacbp.alignmentposition_by_query_pos( max( omsr[n1] ) )
# CHECK these coordinates; pacb.exceptions.CoordinateOutOfRange can occur
# in freaky cases. They shouldn't, but do without discovered reason.
# However, in the majority of cases, it is just a 1/few aa offset, which
# can be easily corrected here.
if str(omsrQs) == str(pacb.exceptions.CoordinateOutOfRange):
if thepacbp.__class__.__name__ == 'PacbP':
# solve by taking thepacbp.query_start
omsrQs = thepacbp.alignmentposition_by_query_pos( thepacbp.query_start )
else:
# thepacbp.__class__.__name__ in ['PacbPDNA','PacbPORF']:
# solve by taking orginal alignment position start
omsrQs = thepacbp.alignmentposition_by_query_pos(
thepacbp._get_original_alignment_pos_start().query_pos
)
###########################################################################
## print warning message(s)
#print "WARNING: pacb.exceptions.CoordinateOutOfRange (omsrQs, ",
#print "node %s in CodingBlockGraph2GeneTreeGraph" % ( str(n1) )
#print "WARNING: min(omsr(", min( omsr[n1] ), ")", min(omsr[n1]),
#print max(omsr[n1]), " taken ->", thepacbp.query_start, omsrQs
#print "WARNING: ", thepacbp
###########################################################################
if str(omsrQe) == str(pacb.exceptions.CoordinateOutOfRange):
if thepacbp.__class__.__name__ == 'PacbP':
# solve by taking thepacbp.query_end
omsrQe = thepacbp.alignmentposition_by_query_pos( thepacbp.query_end )
else:
# thepacbp.__class__.__name__ in ['PacbPDNA','PacbPORF']:
# solve by taking orginal alignment position end
omsrQe = thepacbp.alignmentposition_by_query_pos(
thepacbp._get_original_alignment_pos_end().query_pos
) + 1 # add +1 to create a python list range coordinate
###########################################################################
## print warning message(s)
#print "WARNING: pacb.exceptions.CoordinateOutOfRange (omsrQe, ",
#print node %s in CodingBlockGraph2GeneTreeGraph" % ( str(n1) )
#print "WARNING: max(omsr(", max( omsr[n1] ), ")", min(omsr[n1]),
#print max(omsr[n1]), " taken ->", thepacbp.query_end, omsrQe
#print "WARNING: ", thepacbp
###########################################################################
else:
# omsrQe was nicely an integer; add +1 because max(OMSR) is not a range coord
omsrQe += 1
# calculate identityscore
identityscore = pacb.calculate_identityscore( thepacbp.alignment[omsrQs:omsrQe] )
else:
# this edge is absent in the CBG!
# TODO -> this will cause a crash a few lines later
# by definition, a CBG MUST HAVE ALL EDGES at this stage!
print "about to crash!!!!"
print cbg
print cbg.node_count(), cbg.edge_count(), "missing:", (n1,n2)
identityscore = 0.0
# get organism identifyers from node and add edge
o1,o2 = cbgnode2orgnode[ n1 ], cbgnode2orgnode[ n2 ]
# Wt used is identityscore == Identity + 0.5* Similarity
gtg.add_edge( o1, o2, wt=identityscore )
# add additional statistics to gtg object. Wt used is
# identitypercentage is TRUE aa indentity %
identityperc = pacb.calculate_identity( thepacbp.alignment[omsrQs:omsrQe] )
gtg._aa_identity_percentages[(o1,o2)] = identityperc
gtg._aa_identity_percentages[(o2,o1)] = identityperc
# bitscoreratio is ratio of bits / max bits
bitscoreratio = pacb.calculate_bitscoreratio(
thepacbp.query[omsrQs:omsrQe],
thepacbp.sbjct[omsrQs:omsrQe],
matrix = thepacbp.MATRIX
)
gtg._bitscore_ratios[(o1,o2)] = bitscoreratio
gtg._bitscore_ratios[(o2,o1)] = bitscoreratio
# ntidentity is obviously nt identity%
dnaQseq, dnaSseq = thepacbp.get_unextended_aligned_dna_sequences()
ntidentity = sequence_identity_ratio(dnaQseq,dnaSseq)
gtg._nt_identity_percentages[(o1,o2)] = ntidentity
gtg._nt_identity_percentages[(o2,o1)] = ntidentity
# check if the graph is saturated (complete)
# if not (organism/node/orf missing), add this as a zero-wt edge
gtg.makecompletegraph(wt=0.0)
# and return this new genetree graph
return gtg
def CodingBlockGraph2GeneTreeGraph(cbg):
"""
Convert CodingBlockGraph 2 GeneTree
@attention: function just converts, error check is not performed here!
@type cbg: CodingBlockGraph
@param cbg: CodingBlockGraph instance
@rtype: GeneTreeGraph
@return: GeneTreeGraph instance
"""
gtg = GeneTreeGraph()
cbgnode2orgnode = {}
for node in cbg.get_nodes():
org = cbg._organism_from_node(node)
gtg.add_node(org)
# add node/org combi to mapping dict
cbgnode2orgnode[node] = org
# now add all the edges
omsr = cbg.overall_minimal_spanning_range()
for (n1, n2) in cbg.pairwisecrosscombinations_node():
if cbg.has_edge(n1, n2):
# get pacbp(orf) object
thepacbp = cbg.get_pacbps_by_nodes(node1=n1, node2=n2)[0]
# get relative coordinates of the OMSR part of the alignment
omsrQs = thepacbp.alignmentposition_by_query_pos(min(omsr[n1]))
omsrQe = thepacbp.alignmentposition_by_query_pos(max(omsr[n1]))
# CHECK these coordinates; pacb.exceptions.CoordinateOutOfRange can occur
# in freaky cases. They shouldn't, but do without discovered reason.
# However, in the majority of cases, it is just a 1/few aa offset, which
# can be easily corrected here.
if str(omsrQs) == str(pacb.exceptions.CoordinateOutOfRange):
if thepacbp.__class__.__name__ == 'PacbP':
# solve by taking thepacbp.query_start
omsrQs = thepacbp.alignmentposition_by_query_pos(
thepacbp.query_start)
else:
# thepacbp.__class__.__name__ in ['PacbPDNA','PacbPORF']:
# solve by taking orginal alignment position start
omsrQs = thepacbp.alignmentposition_by_query_pos(
thepacbp._get_original_alignment_pos_start().query_pos)
###########################################################################
## print warning message(s)
#print "WARNING: pacb.exceptions.CoordinateOutOfRange (omsrQs, ",
#print "node %s in CodingBlockGraph2GeneTreeGraph" % ( str(n1) )
#print "WARNING: min(omsr(", min( omsr[n1] ), ")", min(omsr[n1]),
#print max(omsr[n1]), " taken ->", thepacbp.query_start, omsrQs
#print "WARNING: ", thepacbp
###########################################################################
if str(omsrQe) == str(pacb.exceptions.CoordinateOutOfRange):
if thepacbp.__class__.__name__ == 'PacbP':
# solve by taking thepacbp.query_end
omsrQe = thepacbp.alignmentposition_by_query_pos(
thepacbp.query_end)
else:
# thepacbp.__class__.__name__ in ['PacbPDNA','PacbPORF']:
# solve by taking orginal alignment position end
omsrQe = thepacbp.alignmentposition_by_query_pos(
thepacbp._get_original_alignment_pos_end().query_pos
) + 1 # add +1 to create a python list range coordinate
###########################################################################
## print warning message(s)
#print "WARNING: pacb.exceptions.CoordinateOutOfRange (omsrQe, ",
#print node %s in CodingBlockGraph2GeneTreeGraph" % ( str(n1) )
#print "WARNING: max(omsr(", max( omsr[n1] ), ")", min(omsr[n1]),
#print max(omsr[n1]), " taken ->", thepacbp.query_end, omsrQe
#print "WARNING: ", thepacbp
###########################################################################
else:
# omsrQe was nicely an integer; add +1 because max(OMSR) is not a range coord
omsrQe += 1
# calculate identityscore
identityscore = pacb.calculate_identityscore(
thepacbp.alignment[omsrQs:omsrQe])
else:
# this edge is absent in the CBG!
# TODO -> this will cause a crash a few lines later
# by definition, a CBG MUST HAVE ALL EDGES at this stage!
print "about to crash!!!!"
print cbg
print cbg.node_count(), cbg.edge_count(), "missing:", (n1, n2)
identityscore = 0.0
# get organism identifyers from node and add edge
o1, o2 = cbgnode2orgnode[n1], cbgnode2orgnode[n2]
# Wt used is identityscore == Identity + 0.5* Similarity
gtg.add_edge(o1, o2, wt=identityscore)
# add additional statistics to gtg object. Wt used is
# identitypercentage is TRUE aa indentity %
identityperc = pacb.calculate_identity(
thepacbp.alignment[omsrQs:omsrQe])
gtg._aa_identity_percentages[(o1, o2)] = identityperc
gtg._aa_identity_percentages[(o2, o1)] = identityperc
# bitscoreratio is ratio of bits / max bits
bitscoreratio = pacb.calculate_bitscoreratio(
thepacbp.query[omsrQs:omsrQe],
thepacbp.sbjct[omsrQs:omsrQe],
matrix=thepacbp.MATRIX)
gtg._bitscore_ratios[(o1, o2)] = bitscoreratio
gtg._bitscore_ratios[(o2, o1)] = bitscoreratio
# ntidentity is obviously nt identity%
dnaQseq, dnaSseq = thepacbp.get_unextended_aligned_dna_sequences()
ntidentity = sequence_identity_ratio(dnaQseq, dnaSseq)
gtg._nt_identity_percentages[(o1, o2)] = ntidentity
gtg._nt_identity_percentages[(o2, o1)] = ntidentity
# check if the graph is saturated (complete)
# if not (organism/node/orf missing), add this as a zero-wt edge
gtg.makecompletegraph(wt=0.0)
# and return this new genetree graph
return gtg
