def gen_alignment(seq_names = [], alignment = SeqGroup(), outputfile = "alignment.out"): """generate alignment from the input taxa name list - seq_name, and SeqGroup - alignment""" newalign = SeqGroup() for taxa in seq_names: seq = alignment.get_seq(taxa) newalign.set_seq(taxa, seq) newalign.write(outfile = outputfile) return outputfile, newalign
def __write_algn(self, fullpath):
"""
to write algn in paml format
"""
seq_group = SeqGroup()
for n in self:
seq_group.id2seq [n.node_id] = n.nt_sequence
seq_group.id2name [n.node_id] = n.name
seq_group.name2id [n.name ] = n.node_id
seq_group.write(outfile=fullpath, format='paml')
def __write_algn(self, fullpath):
"""
to write algn in paml format
"""
seq_group = SeqGroup()
for n in self:
seq_group.id2seq[n.node_id] = n.nt_sequence
seq_group.id2name[n.node_id] = n.name
seq_group.name2id[n.name] = n.node_id
seq_group.write(outfile=fullpath, format='paml')
def random_remove_taxa(falign, num_remove, num_repeat = 1): align = SeqGroup(sequences = falign) entrs = align.get_entries() numseq = len(entrs) index = range(numseq) namel = [] for i in range(num_repeat): newalign = SeqGroup() random.shuffle(index) idxs = index[num_remove:] for idx in idxs: newalign.set_seq(entrs[idx][0], entrs[idx][1]) newalign.write(outfile = falign + "_" + repr(num_remove)+ "_" + repr(i + 1) + ".afa") namel.append(falign + "_" + repr(num_remove)+ "_" + repr(i + 1) + ".afa") return namel
def extract_placement_crop(nfin_place, nfin_aln, nfout, min_lw = 0.5, logfile = "spcount.log"):
if os.path.exists(logfile):
os.remove(logfile)
if os.path.exists(nfout + "_inode_picked_otus.fasta"):
os.remove(nfout + "_inode_picked_otus.fasta")
jsondata = open (nfin_place)
align_orgin = SeqGroup(sequences = nfin_aln)
data = json.load(jsondata)
placements = data["placements"]
tree = data["tree"]
ete_tree = tree.replace("{", "[&&NHX:B=")
ete_tree = ete_tree.replace("}", "]")
root = Tree(ete_tree, format=1)
leaves = root.get_leaves()
allnodes = root.get_descendants()
allnodes.append(root)
"""get refseq"""
refseqset = []
for leaf in leaves:
refseqset.append(leaf.name)
refali = gen_alignment2(seq_names = refseqset, alignment = align_orgin)
placemap = {}
"""find how many edges are used for placement"""
for placement in placements:
edges = placement["p"]
curredge = edges[0][0]
lw = edges[0][2]
if lw >= min_lw:
placemap[curredge] = placemap.get(curredge, [])
"""placement quality control"""
discard_file = open(nfout+".discard.placement.txt", "w")
"""group taxa to edges"""
for placement in placements:
edges = placement["p"]
taxa_names = placement["n"]
curredge = edges[0][0]
lw = edges[0][2]
if lw >= min_lw:
a = placemap[curredge]
a.extend(taxa_names)
placemap[curredge] = a
else:
discard_file.write(repr(taxa_names) + "\n")
discard_file.close()
groups = placemap.items()
cnt_leaf = 0
cnt_inode = 0
"""check each edge"""
for i,item in enumerate(groups):
seqset_name = item[0]
seqset = item[1]
"""check if placed on leaf node and find the node being placed on"""
flag = False
place_node = None
for node in allnodes:
if str(node.B) == str(seqset_name):
place_node = node
if node.is_leaf():
flag = True
break
"""generate aligment"""
if flag:
"""process leaf node placement"""
cnt_leaf = cnt_leaf + 1
newalign = SeqGroup()
for taxa in seqset:
seq = align_orgin.get_seq(taxa)
newalign.set_seq(taxa, seq)
place_seq = align_orgin.get_seq(place_node.name)
newalign.set_seq("*R*" + place_node.name, place_seq) #set the reference sequence name
newalign.write(outfile = nfout + "_leaf_"+repr(cnt_leaf) + ".lfa")
else:
cnt_inode = cnt_inode + 1
newalign = SeqGroup()
for taxa in seqset:
seq = align_orgin.get_seq(taxa)
newalign.set_seq(taxa, seq)
if len(newalign.get_entries()) < 2:
count_and_pick_reads(align = newalign, outputfile = nfout + "_inode_picked_otus.fasta")
sp_log(sfout = logfile, logs="I the palcement is on an internal node \nD find new species\nK reads number: 1 \n")
else:
#for entr in refali.get_entries():
# sname = entr[0]
# seqe = entr[1]
# newalign.set_seq(sname, seq)
newalign.write(outfile = nfout + "_inode_"+repr(cnt_inode) + ".ifa")
def extract_placement(nfin_place, nfin_aln, nfout, min_lw = 0.5, logfile = "spcount.log"):
if os.path.exists(logfile):
os.remove(logfile)
if os.path.exists(nfout + "_inode_picked_otus.fasta"):
os.remove(nfout + "_inode_picked_otus.fasta")
jsondata = open (nfin_place)
align_orgin = SeqGroup(sequences = nfin_aln)
data = json.load(jsondata)
placements = data["placements"]
tree = data["tree"]
ete_tree = tree.replace("{", "[&&NHX:B=")
ete_tree = ete_tree.replace("}", "]")
root = Tree(ete_tree, format=1)
leaves = root.get_leaves()
allnodes = root.get_descendants()
allnodes.append(root)
"""get refseq"""
refseqset = []
for leaf in leaves:
refseqset.append(leaf.name)
refali = gen_alignment2(seq_names = refseqset, alignment = align_orgin)
placemap = {}
"""find how many edges are used for placement"""
for placement in placements:
edges = placement["p"]
curredge = edges[0][0]
lw = edges[0][2]
if lw >= min_lw:
placemap[curredge] = placemap.get(curredge, [])
"""placement quality control"""
discard_file = open(nfout+".discard.placement.txt", "w")
"""group taxa to edges"""
for placement in placements:
edges = placement["p"]
taxa_names = placement["n"]
curredge = edges[0][0]
lw = edges[0][2]
if lw >= min_lw:
a = placemap[curredge]
a.extend(taxa_names)
placemap[curredge] = a
else:
discard_file.write(repr(taxa_names) + "\n")
discard_file.close()
groups = placemap.items()
cnt_leaf = 0
cnt_inode = 0
"""check each edge"""
for i,item in enumerate(groups):
seqset_name = item[0]
seqset = item[1]
"""check if placed on leaf node and find the node being placed on"""
flag = False
place_node = None
for node in allnodes:
if str(node.B) == str(seqset_name):
place_node = node
if node.is_leaf():
flag = True
break
"""find the furthest leaf of the placement node"""
fnode = place_node.get_farthest_node()[0]
outgroup_name = fnode.name
"""find sister node"""
snode = place_node.get_sisters()[0]
if not snode.is_leaf():
snode = snode.get_closest_leaf()[0]
sister_name = snode.name
"""generate aligment"""
if flag:
"""process leaf node placement"""
cnt_leaf = cnt_leaf + 1
newalign = SeqGroup()
for taxa in seqset:
seq = align_orgin.get_seq(taxa)
newalign.set_seq(taxa, seq)
if len(newalign.get_entries()) < 2:
#count_and_pick_reads(align = newalign, outputfile = nfout + "_leaf_picked_otus.fasta")
og_seq = align_orgin.get_seq(outgroup_name)
sis_seq = align_orgin.get_seq(sister_name)
newalign.set_seq("sister", sis_seq) #set the sister seqeunce to make 4 taxa
newalign.set_seq("root_ref", og_seq) #set the outgroup name
place_seq = align_orgin.get_seq(place_node.name)
newalign.set_seq("*R*" + place_node.name, place_seq) #set the reference sequence name
newalign.write(outfile = nfout + "_leaf_"+repr(cnt_leaf) + ".lfa")
else:
og_seq = align_orgin.get_seq(outgroup_name)
newalign.set_seq("root_ref", og_seq) #set the outgroup name
place_seq = align_orgin.get_seq(place_node.name)
newalign.set_seq("*R*" + place_node.name, place_seq) #set the reference sequence name
newalign.write(outfile = nfout + "_leaf_"+repr(cnt_leaf) + ".lfa")
else:
"""genrate the newwick string to be inserted into the ref tree"""
rep = re.compile(r"\{[0-9]*\}")
multi_fcating = "("
for seqname in seqset:
multi_fcating = multi_fcating + seqname + ","
multi_fcating = multi_fcating[:-1]
multi_fcating = "{" + repr(seqset_name) + "}," + multi_fcating + ")"
mtfc_tree = tree.replace("{" + repr(seqset_name) + "}", multi_fcating)
mtfc_tree = rep.sub("", mtfc_tree)
cnt_inode = cnt_inode + 1
newalign = SeqGroup()
for taxa in seqset:
seq = align_orgin.get_seq(taxa)
newalign.set_seq(taxa, seq)
if len(newalign.get_entries()) < 2:
count_and_pick_reads(align = newalign, outputfile = nfout + "_inode_picked_otus.fasta")
sp_log(sfout = logfile, logs="I the palcement is on an internal node \nD find new species\nK reads number: 1 \n")
else:
#og_seq = align_orgin.get_seq(outgroup_name)
#newalign.set_seq("root_ref", og_seq)
for entr in refali.get_entries():
sname = entr[0]
seqe = entr[1]
newalign.set_seq(sname, seq)
newalign.write(outfile = nfout + "_inode_"+repr(cnt_inode) + ".ifa")
mtfc_out = open(nfout + "_inode_"+repr(cnt_inode) + ".mttree", "w")
mtfc_out.write(mtfc_tree)
mtfc_out.close()
