def importIsoggSnps(self):
'import ISOGG SNPs'
SNP.setClassVariables(self)
self.readPreferredSNPnameSet()
self.readRepresentativeSNPnameSet()
self.readIsoggMultiAllelicPosSet()
self.readIsoggOmitSet()
self.readIsoggCorrectionsDict()
self.parseIsoggTable()
self.setDepthFirstNodeList()
self.sortSNPlistsAndSetRepresentatives()
self.writeIsoggCounts()
self.writeUniqueSNPtable()
self.writeNewick()
self.checkMultiAllelics()
def prioritySortSNPlistAndSetHgSNP(self):
'''
first, sorts snp list (or dropped marker list) by priority ranking.
then, sets reresentative-SNP-based label: self.hgSNP
the standard form incudes the truncated haplogroup label
and the label of a representative SNP, separated by a hyphen (e.g. R-V88).
'''
# root: no markers
if self.isRoot():
self.hgSNP = self.haplogroup
# normal case
elif self.snpList:
self.snpList = SNP.prioritySortMarkerList(self.snpList)
self.hgSNP = self.mostHighlyRankedSNP.hgSNP
# backup: use discared marker name
elif self.droppedMarkerList:
self.droppedMarkerList = SNP.prioritySortMarkerList(
self.droppedMarkerList)
markerName = self.mostHighlyRankedDroppedMarker.name
self.hgSNP = '%s-%s' % (self.hgTrunc, markerName)
# no markers to use
else:
if self.parent.hgSNP:
symbol = '*' if self.isLeaf() else '+'
self.hgSNP = self.parent.hgSNP + symbol
# uniquify if necessary
if self.hgSNP in Node.hgSNPset:
i = 1
hgSNPuniqe = '%s%d' % (self.hgSNP, i)
while hgSNPuniqe in Node.hgSNPset:
i += 1
hgSNPuniqe = '%s%d' % (self.hgSNP, i)
self.hgSNP = hgSNPuniqe
else:
Node.errAndLog('WARNING. Attempted to set star label, ' + \
'but parent.hgSNP not set yet: %s\n' % self.haplogroup)
self.hgSNP = self.haplogroup
Node.hgSNPset.add(self.hgSNP)
def snp(self, rsid):
"""Returns SNP with given integer-only RSID."""
try:
genotype, chromo, position = self._genome[rsid]
geno = map(Nucleotide, genotype)
return _to_snp("rs%d" % rsid, self._orientation, (geno, chromo,
position))
except KeyError:
return SNP([], "rs%d" % rsid, self._orientation, 0, 0)
def constructSNP(self, name, haplogroup, position, mutation):
'''
typically, instantiates a SNP and adds it to various containers.
note that when SNPs are instantiated, they are added to the tree,
and this process may entail growing the tree to include the corresponding node.
more specialized things occur if a SNP already exists at this position.
'''
if self.hg2nodeDict:
ancestral, derived = mutation[0], mutation[3]
snpKey = (haplogroup, position)
if snpKey in self.snpDict: # snp exists under an alias
snp = self.snpDict[snpKey]
if snp.isAncestral(ancestral) and snp.isDerived(derived):
snp.addName(name)
self.snpDict[name] = snp
else:
newSNP = SNP(name, haplogroup, position, ancestral,
derived)
sys.exit('\n\nERROR! Conlicting SNPs:\n%s\n%s\n' % \
(snp, newSNP))
else:
if position in self.snpDict: # another snp with same position
oldSNP = self.snpDict[position]
if ancestral not in oldSNP.alleleSet or \
derived not in oldSNP.alleleSet:
self.multiAllelicNewPosSet.add(position)
# typical behavior
snp = SNP(name, haplogroup, position, ancestral, derived)
self.snpDict[(haplogroup, position)] = snp
self.snpDict[name] = snp
self.snpDict[position] = snp
self.snpList.append(snp)
self.snpPosSet.add(position)
self.snpNameSet.add(name)
self.isoggCountsDict['unique'] += 1
def updateWithBranchAssessment(self, ancSNPlist, derSNPlist):
'''
extends derived SNP list, sets most derived SNP, and adds number of
ancestral alleles seen. also, manages tracking of whether or not path
has pushed through an (anc,der)==(1,0) branch
'''
numAncestral, numDerived = len(ancSNPlist), len(derSNPlist)
self.numAncestral += numAncestral
self.derSNPlist.extend(derSNPlist)
if derSNPlist:
self.mostDerivedSNP = SNP.mostHighlyRankedMarkerOnList(derSNPlist)
if self.hasPushedThrough:
self.updatePushThroughVars(numAncestral, numDerived)
elif (numAncestral, numDerived) == (1, 0):
self.setPushThroughVars()
def positions_from_csv(filename, min_allele_freq):
'''
Read SNP positions from a CSV file in following format:
chrom,pos,freq
1,1110,0.02631
1,2271,0.03125
1,2402,0.03125
1,4559,0.02777
'''
#log.info('Reading SNPs from file: %s', filename)
SNPs = []
with open(os.path.join(os.path.dirname(__file__), filename)) as f:
reader = csv.reader(f)
header = next(reader, None)
for chrom, pos, freq in reader:
snp = SNP(int(chrom), int(pos), float(freq))
if snp.freq > min_allele_freq:
SNPs.append(snp)
return SNPs
l_count = 0
snps = []
for variant_line in variant_lines:
# remove /n form end of line
# variant_line = variant_line.strip()
# remove " from lines
# variant_line[0] = variant_line[0].translate(None, '"')
# variant_line[-1] = variant_line[-1].translate(None, '"')
if len(variant_line) >= 16:
success_count += 1
context = variant_line[5].split(",")
consequences = variant_line[6].split(",")
snps.append(
SNP(l_count, variant_line[0], int(variant_line[1]),
variant_line[2], variant_line[3], variant_line[4], context,
consequences,
variant_line[7], variant_line[8], variant_line[9],
int(variant_line[10]), variant_line[11], variant_line[12],
variant_line[13], variant_line[14], variant_line[15]))
elif 16 > len(variant_line) > 4:
success_count += 1
# context = variant_line[5].split(",")
# consequences = variant_line[6].split(",")
snps.append(
SNP(l_count, variant_line[0], int(variant_line[1]),
variant_line[2], variant_line[3], variant_line[4],
variant_line[5], ".", ".", ".", ".", ".", ".", ".", ".",
".", "."))
else:
print "INVALID DATA (16 >= length > 4) in Line {}".format(l_count)
print variant_line
fail_count += 1
def mostHighlyRankedSNP(self):
'the most highly ranked SNP'
return SNP.mostHighlyRankedMarkerOnList(self.snpList)
def mostHighlyRankedDroppedMarker(self):
'the most highly ranked dropped marker'
return SNP.mostHighlyRankedMarkerOnList(self.droppedMarkerList)
def _to_snp(rsid, orientation, value):
genotype, chromo, position = value
return SNP(list(genotype), rsid, orientation, chromo, position)