def main(argv):
treefile = ''
species_list = argv[2:]
try:
opts, args = getopt.getopt(argv[0:2],"ht:o:",["tree=","alignment=","species=", "consensus=","log="])
except getopt.GetoptError:
print 'GetClades.py -t <treefile> -o <outfile> <species>'
sys.exit(2)
for opt, arg in opts:
if opt == '-h':
print 'GetClades.py -t <treefile> -o <outfile> <species>'
sys.exit()
elif opt in ("-t", "--tree"):
treefile = arg
#create outfilename from treefile
outfilename = treefile.replace(".nwk", "_clades.txt")
#skip tree files with a size of zero (usually caused by too few sequences to make a tree)
if path.getsize(treefile) == 0:
sys.exit()
#read in tree, root by midpoint, and add info about target species
t = Tree(treefile)
if len(t) > 3:
R = t.get_midpoint_outgroup() #midpoint root tree unless there are only three taxa, which was causing an error
t.set_outgroup(R)
t = add_species([t] + species_list)
#extract cluster name from treefile (remove path and extension)
cluster_name = path.split(treefile)[1].split(".")[0] + "_"
#make dictionary with clade numbers as keys and arrays of species as values
clade_dict = {}
clade_num = 0
for node in t.get_monophyletic(values=["ingroup"], target_attr="species"):
clade_num = clade_num +1
clade_dict[clade_num] = [""] * len(species_list) #create array of empty values with same length as species_list
for leaf in node:
for i in range(len(species_list)):
if species_list[i] in leaf.name:
clade_dict[clade_num][i] = cluster_name + str(clade_num)
#open outfile and write results
outfile_handle = open(outfilename, "w")
for clade in clade_dict.values():
outfile_handle.write("\t".join(clade) + "\n")
outfile_handle.close()
def main(argv):
treefile = ''
alnmentfile = ''
consfilename = ''
logfilename = ''
species_name = ''
try:
opts, args = getopt.getopt(argv,"ht:a:c:l:s:",["tree=","alignment=","species=", "consensus=","log="])
except getopt.GetoptError:
print 'MakeConsensus.py -t <treefile> -a <alignmentfile> -s <species> -c <consensusfile> -l <logfile>'
sys.exit(2)
for opt, arg in opts:
if opt == '-h':
print 'MakeConsensus.py -t <treefile> -a <alignmentfile> -s <species> -c <consensusfile> -l <logfile>'
sys.exit()
elif opt in ("-t", "--tree"):
treefile = arg
elif opt in ("-a", "--alignment"):
alnfile = arg
elif opt in ("-c", "--consensus"):
consfilename = arg
elif opt in ("-l", "--log"):
logfilename = arg
elif opt in ("-s", "--species"):
species_name = arg
#make a list of names to ensure that there are no duplicates.
#if saved list exists, use it instead
namefile = "names.p"
if path.exists(namefile):
name_list = pickle.load( open( namefile, "rb" ) )
else:
name_list = {}
name_num = 1
#read in tree and root by midpoint
t = Tree(treefile)
R = t.get_midpoint_outgroup()
t.set_outgroup(R)
#read in alignment and create index of sequence ids to access individual sequences by id
aln = AlignIO.read(alnfile, "phylip-relaxed")
aln_index = {}
seq_num = 0
for seq in aln:
aln_index[seq.id] = seq_num
seq_num = seq_num +1
#Add "species" feature to leaves on tree for easy lookup
t = add_species(t)
#cycle through monophyletic groups and build consensus sequences
for node in t.get_monophyletic(values=[species_name], target_attr="species"):
species_seqs = []
for leaf in node:
if 'kraussiana' not in leaf.name and 'willdenowii' not in leaf.name: #exclude reference sequence from consensus
species_seqs.append(aln[aln_index[leaf.name]])
if len(species_seqs) > 1: #create consensus sequence from sequence list
species_aln = MultipleSeqAlignment(species_seqs)
consensus = smart_consensus(MultipleSeqAlignment(pad_ends(species_aln))) #use pad_ends to convert end gaps to ambiguous
consensus = pad_ends([consensus], '-', 'N')[0] #convert back to gap characters (need to convert seq to a list)
name = pick_name(species_seqs) + "_cons"
while name in name_list:
name = name.split("_")[0] + "_" + str(name_num) + "_cons"
name_num = name_num + 1
name_list[name] = 1
name_num = 1
consensus = SeqRecord(Seq(str(consensus).replace('-','')),
id=name,
description=name)
elif len(species_seqs) == 1: #for singletons, just write ungapped version of original
consensus = species_seqs[0]
consensus.seq = consensus.seq.ungap("-")
name = consensus.id
else:
continue
#write consensus sequence to consensus file
consfile = open(consfilename, "a")
consfile.write(consensus.format("fasta"))
consfile.close()
#write summary of seqs represented by cons to log file
cluster_num = path.split(alnfile)[1].split(".")[0] #get cluster number
logfile = open(logfilename, "a")
for seq in species_seqs:
logfile.write("%s\n" % ", ".join([seq.id, name, cluster_num]))
logfile.close()
#save list of names for future uses of script
pickle.dump( name_list, open( namefile, "wb" ) )