def optimize_model(self, gtree, stree, gene2species):
"""Optimizes the model"""
CostModel.optimize_model(self, gtree, stree, gene2species)
# ensure gtree and stree are both rooted and binary
if not (treelib.is_rooted(gtree) and treelib.is_binary(gtree)):
raise Exception("gene tree must be rooted and binary")
if not (treelib.is_rooted(stree) and treelib.is_binary(stree)):
raise Exception("species tree must be rooted and binary")
try:
junk = phylo.reconcile(gtree, stree, gene2species)
except:
raise Exception("problem mapping gene tree to species tree")
treeout = StringIO.StringIO()
if not self.printed:
import pprint
treelib.draw_tree(gtree, out=treeout, minlen=5, maxlen=5)
print "gene tree:\n"
print(treeout.getvalue())
treelib.draw_tree(self.stree, out=treeout, minlen=5, maxlen=5)
print "spec tree:\n"
print(treeout.getvalue())
pprint.pprint(junk)
self.printed = True
def __init__(self, extra):
"""Initializes the model"""
CostModel.__init__(self, extra)
self.VERSION = "1.0.1"
self.mincost = 0
parser = optparse.OptionParser(prog="DupLossModel")
parser.add_option(
"-D",
"--dupcost",
dest="dupcost",
metavar="<dup cost>",
default=1.0,
type="float",
help="duplication cost (default: 1.0)",
)
parser.add_option(
"-L",
"--losscost",
dest="losscost",
metavar="<loss cost>",
default=1.0,
type="float",
help="loss cost (default: 1.0)",
)
self.parser = parser
CostModel._parse_args(self, extra)
def __init__(self, extra):
"""Initializes the model"""
CostModel.__init__(self, extra)
self.VERSION = "0.1.0"
self.mincost = 0
parser = optparse.OptionParser(prog="DTLModel")
parser.add_option("-D", "--dupcost", dest="dupcost",
metavar="<dup cost>",
default=2, type="int",
help="duplication cost (default: 2)")
parser.add_option("-T", "--transfercost", dest="transfercost",
metavar="<transfer cost>",
default=3, type="int",
help="transfer cost (default: 3)")
parser.add_option("-L", "--losscost", dest="losscost",
metavar="<loss cost>",
default=1, type="int",
help="loss cost (default: 1)")
self.parser = parser
CostModel._parse_args(self, extra)
# make temporary file
fd, self.treefile = tempfile.mkstemp()
os.close(fd)
def __init__(self, extra):
"""Initializes the model"""
CostModel.__init__(self, extra)
self.VERSION = "1.0.1"
self.mincost = 0
self.printed = False
def __init__(self, extra):
"""Initializes the model"""
CostModel.__init__(self, extra)
self.VERSION = "1.0.1"
self.mincost = 0
self.count = 0
self.log = open('matched.txt', 'w')
def __init__(self, extra):
"""Initializes the model"""
CostModel.__init__(self, extra)
self.VERSION = "0.1.0"
self.mincost = 0
CostModel._parse_args(self, extra)
# make temporary file
fd, self.treefile = tempfile.mkstemp()
os.close(fd)
def __init__(self, extra):
"""Initializes the model"""
CostModel.__init__(self, extra)
self.VERSION = "0.1.2"
self.mincost = 0
parser = optparse.OptionParser(prog="DTLModel")
parser.add_option("--cmd", dest="cmd",
metavar="<ranger-dtl-U command>",
default="ranger-dtl-U",
help="ranger-dtl-U command (default: \"ranger-dtl-U\")")
parser.add_option("-D", "--dupcost", dest="dupcost",
metavar="<duplication cost>",
default=2, type="int",
help="duplication cost, integer only (default: 2)")
parser.add_option("-T", "--transcost", "--transfercost", dest="transcost",
metavar="<transfer cost>",
default=3, type="int",
help="transfer cost, integer only (default: 3)")
parser.add_option("-L", "--losscost", dest="losscost",
metavar="<loss cost>",
default=1, type="int",
help="loss cost, integer only (default: 1)")
parser.add_option("--seed", dest="seed",
metavar="<seed>",
type="int",
help="user defined random number generator seed")
parser.add_option("--tmp", dest="tmp",
metavar="<tmp directory>",
help="directory for temporary files (must exist)")
self.parser = parser
CostModel._parse_args(self, extra)
# check temporary directory
if self.tmp:
if not os.path.exists(os.path.realpath(self.tmp)):
raise Exception("--tmp directory does not exist")
# make temporary file
fd, self.treefile = tempfile.mkstemp(dir=self.tmp)
os.close(fd)
# hack for cygwin (ranger-dtl-U cannot handle system files)
if sys.platform == 'cygwin':
cwd = os.getcwd()
if self.treefile.startswith(cwd):
# remove working path (and backslash)
self.treefile = self.treefile[len(cwd)+1:]
else:
raise Exception("--tmp must be a relative path when using cygwin")
def optimize_model(self, gtree, stree, gene2species):
"""Optimizes the model"""
CostModel.optimize_model(self, gtree, stree, gene2species)
# ensure gtree and stree are both rooted and binary
if not (treelib.is_rooted(gtree) and treelib.is_binary(gtree)):
raise Exception("gene tree must be rooted and binary")
if not (treelib.is_rooted(stree) and treelib.is_binary(stree)):
raise Exception("species tree must be rooted and binary")
try:
junk = phylo.reconcile(gtree, stree, gene2species)
except:
raise Exception("problem mapping gene tree to species tree")
def optimize_model(self, gtree, stree, gene2species):
"""Optimizes the model"""
CostModel.optimize_model(self, gtree, stree, gene2species)
#=============================
# read sequences
if not self.align:
self.parser.error("--align must be specified")
self.align = fasta.read_fasta(self.align)
#=============================
# read SPIDIR parameters
if not self.params:
self.parser.error("--param must be specified")
self.params = spidir.read_params(self.params)
#=============================
# determine background base frequency
if self.bgfreq:
# use supplied frequency
vals = map(float, self.bgfreq.split(","))
if len(vals) != 4:
self.parser.error("invalid --bgfreq: %s" % self.bgfreq)
self.bgfreq = vals
else:
# compute frequency from alignment
self.bgfreq = alignlib.compute_bgfreq(self.align)
#=============================
# branch lengths
if self.kappa >= 0:
# use supplied kappa
self.kappa = self.kappa
else:
# compute kappa from alignment
# from spidir.find_ml_kapp_hky
minkappa = 0.4; maxkappa = 5.0; stepkappa = 0.1
maxlk = -util.INF
maxk = minkappa
for k in util.frange(minkappa, maxkappa, stepkappa):
l = spidir.find_ml_branch_lengths_hky(gtree, self.align, self.bgfreq, k, maxiter=1,
parsinit=(k == minkappa))
if l > maxlk:
maxlk = l
maxk = k
self.kappa = maxk
def __init__(self, extra):
"""Initializes the model"""
CostModel.__init__(self, extra)
self.VERSION = "0.1.0"
self.mincost = -util.INF
parser = optparse.OptionParser(prog="SpimapModel")
parser.add_option("-a", "--align", dest="align",
metavar="<alignment fasta>",
help="sequence alignment in fasta format")
parser.add_option("-p", "--param", dest="params",
metavar="<params file>",
help="substitution rate parameters file")
parser.add_option("--simplereroot", dest="simplereroot",
metavar="<simple reroot>",
default=False, action="store_true",
help="set to reroot using dup/loss cost")
grp_seq = optparse.OptionGroup(parser, "Sequence evolution model")
grp_seq.add_option("-k", "--kappa", dest="kappa",
metavar="<transition/transversion estimate>",
default=-1.0, type="float",
help="used for HKY model (default: estimate)")
grp_seq.add_option("-f", "--bgfreq", dest="bgfreq",
metavar="<A freq>,<C freq>,<G freq>,<T freq>",
help="background frequencies (default: estimate)")
parser.add_option_group(grp_seq)
grp_duploss = optparse.OptionGroup(parser, "Dup/loss evolution model")
grp_duploss.add_option("-D", "--duprate", dest="duprate",
metavar="<duplication rate>",
default=0.1, type="float",
help="rate of gene duplication (default: 0.1)")
grp_duploss.add_option("-L", "--lossrate", dest="lossrate",
metavar="<loss rate>",
default=0.1, type="float",
help="rate of gene loss (default: 0.1)")
grp_duploss.add_option("-P", "--pretime", dest="pretime",
metavar="<pre-speciation time parameter>",
default=1.0, type="float",
help="lambda param of pre-speciation distribution (default: 1.0)")
parser.add_option_group(grp_duploss)
self.parser = parser
CostModel._parse_args(self, extra)
def optimize_model(self, gtree, stree, gene2species):
"""Optimizes the model"""
CostModel.optimize_model(self, gtree, stree, gene2species)
if self.dupcost < 0:
self.parser.error("-D/--dupcost must be >= 0")
if self.losscost < 0:
self.parser.error("-L/--losscost must be >= 0")
# ensure gtree and stree are both rooted and binary
if not (treelib.is_rooted(gtree) and treelib.is_binary(gtree)):
raise Exception("gene tree must be rooted and binary")
if not (treelib.is_rooted(stree) and treelib.is_binary(stree)):
raise Exception("species tree must be rooted and binary")
try:
junk = phylo.reconcile(gtree, stree, gene2species)
except:
raise Exception("problem mapping gene tree to species tree")
def __init__(self, extra):
"""Initializes the model"""
CostModel.__init__(self, extra)
self.VERSION = "0.1.0"
self.mincost = 0
parser = optparse.OptionParser(prog="DLCModel")
parser.add_option("-D", "--dupcost", dest="dupcost",
metavar="<dup cost>",
default=1.0, type="float",
help="duplication cost (default: 1.0)")
parser.add_option("-L", "--losscost", dest="losscost",
metavar="<loss cost>",
default=1.0, type="float",
help="loss cost (default: 1.0)")
parser.add_option("-C", "--coalcost", dest="coalcost",
metavar="<(deep) coalescence cost>",
default=1.0, type="float",
help="deep coalescence cost (default: 1.0)")
parser.add_option("-o", "--output", dest="output",
metavar="<locus tree output file>",
help="locus tree output file")
grp_search = optparse.OptionGroup(parser, "Search Options")
grp_search.add_option("--dcs", dest="dcs",
metavar="<DCS threshold>",
default=0.25, type="float",
help="duplication consistency score threshold: " +
"DCS >/<= thr resolves to DUP/ILS (default: 0)")
grp_search.add_option("--switch", dest="switch",
default=False, action="store_true",
help="switches equality assignment for DCS threshold, " +
"i.e. DCS >=/< thr resolves to DUP/ILS")
parser.add_option_group(grp_search)
self.parser = parser
CostModel._parse_args(self, extra)
self.locustree = None
def optimize_model(self, gtree, stree, gene2species):
"""Optimizes the model"""
CostModel.optimize_model(self, gtree, stree, gene2species)
# check arguments
if self.dupcost < 0:
self.parser.error("-D/--dupcost must be >= 0: " + str(self.dupcost))
if self.losscost < 0:
self.parser.error("-L/--losscost must be >= 0: " + str(self.losscost))
if self.coalcost < 0:
self.parser.error("-C/--coalcost must be >= 0: " + str(self.coalcost))
if self.output is None:
self.parser.error("-o/--output must be specified")
if self.dcs < 0 or self.dcs > 1:
self.parser.error("--dcs must be in [0,1]: " + str(self.dcs))
# copy over tree topology
ltree = treelib.Tree(nextname=gtree.nextname)
def walk(node):
# copy of node
newnode = treelib.TreeNode(node.name)
# recurse
for child in node.children:
ltree.add_child(newnode, walk(child))
return newnode
if gtree.root:
walk(gtree.root)
ltree.root = ltree.nodes[gtree.root.name]
else:
raise Exception("input gene tree must be rooted")
self.locustree = ltree
# search functions
self.search = phylo.TreeSearchMix(None)
self.search.add_proposer(phylo.TreeSearchNni(None), 0.5)
self.search.add_proposer(phylo.TreeSearchSpr(None), 0.5)
def recon_root(self, gtree, newCopy=True, returnCost=False):
"""
Reroots the tree by minimizing the duplication/loss cost
Note, may NOT minimize the cost function
"""
if self.simplereroot:
tree = phylo.recon_root(gtree, self.stree, self.gene2species,
newCopy = newCopy,
keepName = True, returnCost = False)
if returnCost:
return tree, self.compute_cost(tree)
else:
return tree
else:
return CostModel.recon_root(self, gtree, newCopy, returnCost)