def GetPTDPrimerInfo(self, primer_info): #{
# [event_id, chimera.id, chimera.left, chimera.right, gene.id,
# gene.left, gene.right]
# event_id
event_id = primer_info.EventID()
# contig_id
contig = PrimerSeqCls(primer_info.ctg_id)
# contig_left
contig.left = min(primer_info.c2t_aligns[0].ctg_end,
primer_info.c2t_aligns[1].ctg_start)
# contig_right
contig.right = max(primer_info.c2t_aligns[0].ctg_end,
primer_info.c2t_aligns[1].ctg_start)
# gene_id
gene = PrimerSeqCls(primer_info.c2t_aligns[0].transcript)
# gene.left
gene.left = min(primer_info.c2t_aligns[0].transcript_end,
primer_info.c2t_aligns[1].transcript_start)
# gene.right
gene.right = max(primer_info.c2t_aligns[0].transcript_end,
primer_info.c2t_aligns[1].transcript_start)
# dup_length
dup_length = (gene.right - gene.left) + 1
# construct inferred chimera sequence
ctg_overlap = primer_info.C2TAlignOverlap()
extra = ""
if (0 > ctg_overlap): #{
extra = self.contigs[primer_info.ctg_id][contig_left+1:contig_right]
#} end if
if ("-" == primer_info.c2t_aligns[0].strand): #{
extra = ReverseComplement(extra)
#} end if
# chimera_id
chimera = PrimerSeqCls("G%i_%s_%s" % (primer_info.group_id,
primer_info.ctg_id, gene.id))
chimera.seq = (self.transcripts[gene.id][:gene.right] + extra +
self.transcripts[gene.id][gene.left-1:])
if (chimera.id in self.transcripts and
self.transcripts[chimera.id] != chimera.seq): #{
raise PrimerDesignerError("conflicting chimeric PTD sequences: %s" %
chimera.id)
#} end if
self.transcripts[chimera.id] = chimera.seq
# chimera_left
chimera.left = gene.left
# chimera_right
chimera.right = gene.right + len(extra) + dup_length
DebugMsg(self, "Overlap: %i, Extra: \"%s\", WT_Length: %i, Dup_Length: "
"%i, CT_Length: %i, Calc_Length: %i" % (ctg_overlap, extra,
len(self.transcripts[gene.id]), dup_length, len(chimera.seq),
len(self.transcripts[gene.id])+len(extra)+dup_length))
return map(str, [event_id, chimera.id, chimera.left, chimera.right,
gene.id, gene.left, gene.right])
def GetFusionPrimerInfo(self, primer_info): #{
# [event_id, contig_id, contig_left, contig_right, geneA_id,
# geneA_left, geneA_right, geneB_id, geneB_left, geneB_right]
# event_id
event_id = primer_info.EventID()
# contig_id
contig = PrimerSeqCls(primer_info.ctg_id)
# contig_left
contig.left = min(primer_info.c2t_aligns[0].ctg_end,
primer_info.c2t_aligns[1].ctg_start)
# contig_right
contig.right = max(primer_info.c2t_aligns[0].ctg_end,
primer_info.c2t_aligns[1].ctg_start)
# contig_overlap
ctg_overlap = primer_info.C2TAlignOverlap()
# ensure that geneA is the 5' gene, if possible
if ("+" == primer_info.c2t_aligns[0].strand): #{
alignA = primer_info.c2t_aligns[0]
alignB = primer_info.c2t_aligns[1]
geneA = self.GeneInfoFromAlign0(alignA, ctg_overlap)
geneB = self.GeneInfoFromAlign1(alignB, ctg_overlap)
else:
alignA = primer_info.c2t_aligns[1]
alignB = primer_info.c2t_aligns[0]
geneA = self.GeneInfoFromAlign1(alignA, ctg_overlap)
geneB = self.GeneInfoFromAlign0(alignB, ctg_overlap)
#} end if
if (MIN_CTG_LEN > len(self.contigs[contig.id])): #{
DebugMsg(self, "Contig %s too short (%i): attempting to construct "
"synthetic chimera" % (contig.id, len(self.contigs[contig.id])))
chimera = self.ConstructFusionSeq(primer_info, contig, ctg_overlap)
if (chimera.id in self.transcripts and
self.transcripts[chimera.id] != chimera.seq): #{
raise PrimerDesignerError("conflicting chimeric fusion sequences: %s" %
chimera.id)
#} end if
if (len(chimera.seq) > len(self.contigs[contig.id])): #{
DebugMsg(self, "Using synthetic chimera sequence")
self.transcripts[chimera.id] = chimera.seq
contig = chimera
#} end if
#} end if
return map(str, [event_id, contig.id, contig.left, contig.right,
geneA.id, geneA.left, geneA.right, geneB.id, geneB.left, geneB.right])
def ConstructFusionSeq(self, primer_info, contig, ctg_overlap): #{
# chimera_id
chimera = PrimerSeqCls("G%i_%s_%s/%s" % (primer_info.group_id,
primer_info.ctg_id, primer_info.c2t_aligns[0].transcript,
primer_info.c2t_aligns[1].transcript))
left_seq = self.RefSeqFromAlign0(primer_info.c2t_aligns[0])
right_seq = self.RefSeqFromAlign1(primer_info.c2t_aligns[1])
extra = ""
if (0 > ctg_overlap): #{
extra = self.contigs[primer_info.ctg_id][contig.left:contig.right-1]
else:
right_seq = right_seq[ctg_overlap:]
#} end if
chimera.seq = (left_seq + extra + right_seq)
# chimera_left
chimera.left = min(len(left_seq), len(left_seq)+ctg_overlap)
# chimera_right
chimera.right = max(len(left_seq), len(left_seq)+ctg_overlap)
DebugMsg(self, "Left: %s\nExtra: %s\nRight: %s\nOverlap: %i, "
"WT_Length0: %i, WT_Length1: %i, Extra_len: %i, CT_Length: %i, "
"Calc_Length: %i" % (left_seq, extra, right_seq, ctg_overlap,
len(left_seq), len(right_seq)+max(0,ctg_overlap), len(extra),
len(chimera.seq), len(left_seq)+len(right_seq)+len(extra)))
return chimera
def GetITDPrimerInfo(self, primer_info): #{
# [event_id, chimera.id, chimera.left, chimera.right, gene.id,
# gene.left, gene.right]
DebugMsg(self, "Getting Primer Info for %s %s" %
(primer_info.EventID(), primer_info.ctg_id))
# event_id
event_id = primer_info.EventID()
# contig_id
contig = PrimerSeqCls(primer_info.ctg_id)
# contig_left
contig.left = min(primer_info.c2g_aligns[1][0],
primer_info.c2g_aligns[2][0])
# contig_right
contig.right = max(primer_info.c2g_aligns[1][1],
primer_info.c2g_aligns[2][1])
# alignment to transcript
align = primer_info.c2t_aligns[0]
# ensure that there are some blocks
if (None == align.query_blocks): #{
align.query_blocks =[[align.ctg_start,align.ctg_end]]
#} end if
if (None == align.target_blocks): #{
align.target_blocks =[
[align.transcript_start,align.transcript_end]]
#} end if
#align = self.AdjustITDAlign(primer_info)
#DebugMsg(self, "Adjusted: %s" % align.ToString())
#cdup_coords = primer_info.c2g_aligns[1]
cdup_coords = self.GetContigDuplicationCoords(primer_info.c2g_aligns,
align)
DebugMsg(self, "CDup: %i-%i" % (cdup_coords[0], cdup_coords[1]))
# gene_id
gene = PrimerSeqCls(align.transcript)
num_blocks = len(align.query_blocks)
for index in range(num_blocks): #{
DebugMsg(self, "Block: %i-%i::%i-%i" % (align.query_blocks[index][0],
align.query_blocks[index][1], align.target_blocks[index][0],
align.target_blocks[index][1]))
if (align.query_blocks[index][0] <= cdup_coords[0] and
cdup_coords[1] <= align.query_blocks[index][1]): #{
DebugMsg(self, "Found block %i" % index)
break
#} end if
#} end for
if (num_blocks <= index): #{
raise PrimerDesignerError("could not find block containing duplication!")
#} end if
start_offset = (cdup_coords[0] - align.query_blocks[index][0])
if ("-" == align.strand): #{
start_offset = -start_offset
#} end if
tdup_start = align.target_blocks[index][0] + start_offset
#for index in range(index, num_blocks): #{
# if (align.query_blocks[index][1] > cdup_coords[1]): #{
# break
# #} end if
#} end for
end_offset = (cdup_coords[1] - align.query_blocks[index][0])
if ("-" == align.strand): #{
end_offset = -end_offset
#} end if
tdup_end = align.target_blocks[index][0] + end_offset
DebugMsg(self, "In Block: %i-%i::%i-%i, offsets:%i,%i, tdup:%i-%i" % (
align.query_blocks[index][0], align.query_blocks[index][1],
align.target_blocks[index][0], align.target_blocks[index][1],
start_offset, end_offset, tdup_start, tdup_end))
# gene.left
gene.left = min(tdup_start, tdup_end)
# gene.right
gene.right = max(tdup_start, tdup_end)
cdup_seq = self.contigs[contig.id][cdup_coords[0]-1:cdup_coords[1]]
tdup_seq = self.transcripts[gene.id][gene.left-1:gene.right]
if ("-" == align.strand): #{
tdup_seq = ReverseComplement(tdup_seq)
#} end if
DebugMsg(self, "Align: %s\nDup-coords: %s-%s;%s-%s\n %s\n %s" % (
align.ToString(), cdup_coords[0], cdup_coords[1], tdup_start,
tdup_end, cdup_seq, tdup_seq))
if (cdup_seq.upper() != tdup_seq.upper()): #{
diffs = 0
for i in range(len(cdup_seq)): #{
if (cdup_seq[i].upper() != tdup_seq[i].upper()): #{
diffs += 1
#} end if
#} end for
if (1 < diffs): #{
raise PrimerDesignerError("Contig duplicated-sequence is not the "
"same as transcript duplicated-sequence!")
else:
LogMsg(self, "single-base mismatch from ref in duplicated seq")
#} end if
#} end if
return map(str, [event_id, contig.id, contig.left, contig.right,
gene.id, gene.left, gene.right])