def build_hmm_from_tree(base,tree_name,aln_name,msa_dir,hmm_dir):
'''
Reads tree and corresponding msa and create an MSA & HMM for each internal node.
'''
# Annotate internal nodes with name of corresponding HMM.
pt = PhyloTree(tree_name,alignment=aln_name,alg_format="fasta")
i_node = 0
for node in pt.traverse():
if not node.is_leaf():
node_name = 'node%s' % (str(i_node))
node.add_features(hmm=node_name)
i_node += 1
# make msa for node
msa_string = []
for leaf in node.iter_leaves():
msa_string.append(">%s" % leaf.name)
msa_string.append(str(leaf.sequence))
msa_string = '\n'.join(msa_string)
msa = open('%s%s.aln' % (msa_dir, node_name),'w'); msa.write(msa_string); msa.close()
# build HMM for node
check_call(['python', 'build_hmmer3_hmm_from_alignment.py', '--name',
'%s%s' % (hmm_dir, node_name),
'%s%s.aln' % (msa_dir, node_name)])
#concatenate HMMs into one file for Hmmscan
os.system('cat %s*.hmm > %s%s_concat.hmm' %
(hmm_dir, hmm_dir, base))
return pt
def integrate_pwids_into_tree(tree, alignment):
'''Takes a tree and an alignment and returns a new tree with values of pwid added to each
node in the tree as node.pwid.'''
pt = PhyloTree(tree,alignment=alignment,alg_format="fasta")
for ind, node in enumerate(pt.traverse()):
node.node_kerf_name = 'node%s' % str(ind).zfill(3)
# For later kerf and sh functions
node.kerf_pass = False
node.sh_pass = False
if not node.is_leaf():
node.min_pwid = get_min_pwid_of_leaves(node.get_leaves())
else:
node.min_pwid = 1.0
return pt
def build_hmm_from_tree(tree_name, aln_name, msa_dir, hmm_dir):
'''
Reads tree and corresponding msa and create an MSA & HMM for each internal node.
'''
# Annotate internal nodes with name of corresponding HMM.
pt = PhyloTree(tree_name, alignment=aln_name, alg_format="fasta")
i_node = 0
for node in pt.traverse():
if not node.is_leaf():
node_name = 'node%s' % (str(i_node))
#print node_name
#print node
node.add_features(hmm=node_name)
i_node += 1
# make msa for node
msa_string = []
for leaf in node.iter_leaves():
msa_string.append(">%s" % leaf.name)
msa_string.append(str(leaf.sequence))
msa_string = '\n'.join(msa_string)
msa = open('%s%s.aln' % (msa_dir, node_name), 'w')
msa.write(msa_string)
msa.close()
# build HMM for node
check_call([
'build_hmmer3_hmm_from_alignment.py', '--name',
'%s%s' % (hmm_dir, node_name),
'%s%s.aln' % (msa_dir, node_name)
])
#concatenate HMMs into one file for Hmmscan
os.system('cat %s*.hmm > %sconcat.hmm' % (hmm_dir, hmm_dir))
return pt
#########################################################################################
print "Reading complete ancestral sequence data generated through R"
rtree=PhyloTree(intree) # Tree for "R" generated patterns
tree=ape.read_tree(intree)
rlist=[]
ropf = open(ressurect_file, "r") # rdata.dat is a rgp.R generated output file
for tab in ropf.readlines():
tab=tab.rstrip()
rlist.append(tab.split(" "))
ropf.close()
ori=np.array(rlist)
for node in rtree.traverse("postorder"): # Patterns are being linked to their corresponding nodes
if node.is_leaf():
node.add_features(data=[None for i in range(len(rlist[0])-1)]) # Its rlist[0]-1, because nucleotides begins with name of species
for i in range(len(ori[:,0])):
if '"'+node.name+'"' == ori[:,0][i] :
node.add_features(rtoken=i+1)
else :
node.add_features(data=[None for i in range(len(rlist[0])-1)])
node.add_features(rtoken=None)
for node in rtree.traverse("postorder"):
if node.is_leaf():
node.data=map(lambda x: x, rlist[node.rtoken-1][1:]) # Its the sequence after name
node.up.rtoken=int(ph.Ancestors(tree, node.rtoken, "parent")[0])
else:
try:
time.time() - start_time)
start_time = time.time()
rtree = PhyloTree(intree) # Tree for "R" generated patterns
tree = ape.read_tree(intree)
rlist = []
ropf = open(ressurect_file, "r") # rdata.dat is a rgp.R generated output file
for tab in ropf.readlines():
tab = tab.rstrip()
rlist.append(tab.split(" "))
ropf.close()
ori = np.array(rlist)
for node in rtree.traverse(
"postorder"): # Patterns are being linked to their corresponding nodes
if node.is_leaf():
node.add_features(data=[
None for i in range(len(rlist[0]) - 1)
]) # Its rlist[0]-1, because nucleotides begins with name of species
for i in range(len(ori[:, 0])):
if '"' + node.name + '"' == ori[:, 0][i]:
node.add_features(rtoken=i + 1)
else:
node.add_features(data=[None for i in range(len(rlist[0]) - 1)])
node.add_features(rtoken=None)
for node in rtree.traverse("postorder"):
if node.is_leaf():
node.data = map(lambda x: x,
rlist[node.rtoken -
