def runwindow(input, outgroup):
print "Running sliding-window method, using %i bp flanking each SNP." % args.window
tmpdir = 'window_tmp'
if os.path.isdir(tmpdir):
shutil.rmtree(tmpdir)
os.makedirs(tmpdir)
snpdb = {}
count = 0
with open(input+'.tab', 'rU') as data:
next(data) #skip header
for line in data:
count += 1
cols = line.split('\t')
scaffold = cols[0]
pos = cols[1]
name = 'snp'+str(count)
if not name in snpdb:
snpdb[name] = [scaffold, int(pos) - args.window-1, int(pos) + args.window]
sequencedb = {}
for num, id in enumerate(samples):
if id == 'Reference':
fastafile = os.path.join(samples[num+1], 'reference', 'ref.fa')
else:
fastafile = os.path.join(id, 'snps.consensus.subs.fa')
sequence = []
with open(fastafile, 'rU') as fasta:
sequence_dict = SeqIO.to_dict(SeqIO.parse(fasta, "fasta"))
for k,v in natsorted(snpdb.iteritems()):
rec = sequence_dict.get(v[0])
sequence.append(str(rec.seq[v[1]:v[2]]))
if id not in sequencedb:
sequencedb[id] = ''.join(sequence)
with open(os.path.join(tmpdir, 'windows.fa'), 'w') as output:
for k,v in sequencedb.iteritems():
output.write('>%s\n%s\n' % (k,v))
#run alignment on output
with open(os.path.join(tmpdir, 'windows.mafft.fa'), 'w') as output:
subprocess.call(['mafft', '--thread', str(args.cpus), '--auto', os.path.join(tmpdir, 'windows.fa')], stdout = output, stderr = FNULL)
#now trim alignment
subprocess.call(['trimal', '-in', os.path.join(tmpdir, 'windows.mafft.fa'), '-out', os.path.join(tmpdir, 'windows.trimal.fa'), '-automated1'])
if not outgroup:
subprocess.call(['raxmlHPC-PTHREADS', '-T', str(args.cpus), '-f', 'a', '-m', 'GTRGAMMA', '-p', '12345', '-x', '12345', '-#', str(args.bootstrap), '-s', 'windows.trimal.fa', '-n', 'windows.nwk'], cwd = tmpdir, stdout = FNULL, stderr = FNULL)
else:
subprocess.call(['raxmlHPC-PTHREADS', '-T', str(args.cpus), '-f', 'a', '-m', 'GTRGAMMA', '-p', '12345', '-x', '12345', '-#', str(args.bootstrap), '-s', 'windows.trimal.fa', '-o', outgroup, '-n', 'windows.nwk'], cwd = tmpdir, stdout = FNULL, stderr = FNULL)
#parse with biopython and draw
trees = list(Phylo.parse(os.path.join(tmpdir, 'RAxML_bootstrap.windows.nwk'), 'newick'))
best = Phylo.read(os.path.join(tmpdir,'RAxML_bestTree.windows.nwk'), 'newick')
support_tree = get_support(best, trees)
Phylo.write(support_tree, args.input+'.windows.phylogeny.nwk', 'newick')
Phylo.draw(support_tree, do_show=False)
pylab.axis('off')
pylab.savefig(args.input+'.windows.phylogeny.pdf', format='pdf', bbox_inches='tight', dpi=1000)
def runbinary(input, outgroup):
print "Runnning binary (presence/absence) method."
tmpdir = 'binary_tmp'
if os.path.isdir(tmpdir):
shutil.rmtree(tmpdir)
os.makedirs(tmpdir)
datadb = []
with open(input+'.binary.fa', 'w') as output:
with open(input+'.vcf', 'rU') as vcf:
for line in vcf:
line = line.replace('\n', '')
if line.startswith('##'):
continue
data = line.split('\t')[9:]
if '3' in data:
data1 = ['0' if x=='3' else x for x in data]
data1 = ['0' if x=='2' else x for x in data1]
datadb.append(data1)
data2 = ['0' if x=='1' else x for x in data]
data2 = ['1' if x=='2' else x for x in data2]
data2 = ['0' if x=='3' else x for x in data2]
data3 = ['0' if x=='1' else x for x in data]
data3 = ['0' if x=='2' else x for x in data3]
data3 = ['1' if x=='3' else x for x in data3]
datadb.append(data3)
elif '2' in data:
data1 = ['0' if x=='2' else x for x in data]
datadb.append(data1)
data2 = ['0' if x=='1' else x for x in data]
data2 = ['1' if x=='2' else x for x in data2]
datadb.append(data2)
else:
datadb.append(data)
binarydata = [list(x) for x in zip(*datadb)]
for i in binarydata:
output.write(">%s\n%s\n" % (i[0], ''.join(i[1:])))
if not outgroup:
subprocess.call(['raxmlHPC-PTHREADS', '-T', str(args.cpus), '-f', 'a', '-m', 'BINGAMMA', '-p', '12345', '-x', '12345', '-#', str(args.bootstrap), '-s', os.path.abspath(input+'.binary.fa'), '-n', 'binary.nwk'], cwd = tmpdir, stdout = FNULL, stderr = FNULL)
else:
subprocess.call(['raxmlHPC-PTHREADS', '-T', str(args.cpus), '-f', 'a', '-m', 'BINGAMMA', '-p', '12345', '-x', '12345', '-#', str(args.bootstrap), '-s', os.path.abspath(input+'.binary.fa'), '-o', outgroup, '-n', 'binary.nwk'], cwd = tmpdir, stdout = FNULL, stderr = FNULL)
#parse with biopython and draw
trees = list(Phylo.parse(os.path.join(tmpdir, 'RAxML_bootstrap.binary.nwk'), 'newick'))
best = Phylo.read(os.path.join(tmpdir,'RAxML_bestTree.binary.nwk'), 'newick')
support_tree = get_support(best, trees)
Phylo.write(support_tree, args.input+'.binary.phylogeny.nwk', 'newick')
Phylo.draw(support_tree, do_show=False)
pylab.axis('off')
pylab.savefig(args.input+'.binary.phylogeny.pdf', format='pdf', bbox_inches='tight', dpi=1000)
def main():
args = parser.parse_args()
table = args.table
if not table:
print("No input table provided. This argument is required\n")
parser.print_help()
exit(0)
df = pd.read_csv(table, sep=args.delim, quotechar='"', index_col=False)
for c in df.columns:
if c.lower() == "sample id":
df.set_index(c)
else:
df[c] = pd.to_numeric(df[c], errors="coerce", downcast="integer")
df.dropna(axis=1, how="all", inplace=True)
data = df.values.astype(int)
names = df.index.tolist()
constructor = DistanceTreeConstructor()
if args.upgma:
const = constructor.upgma
else:
const = constructor.nj
if args.tree:
main_tree = Phylo.read(args.tree, "newick")
else:
main_tree = const(get_distance_matrix(data, names))
trees = [
const(get_distance_matrix(sub, names))
for sub in bootstrap(data, args.times)
]
supp_tree = get_support(main_tree, trees)
# remove labels of inner clades
for clade in supp_tree.get_nonterminals():
clade.name = ""
Phylo.write(supp_tree, args.output, "newick")
def runsnps(input, outgroup):
print "Running concatenated SNP method."
file_in = input+'.aln'
tmpdir = 'snps_tmp'
if os.path.isdir(tmpdir):
shutil.rmtree(tmpdir)
os.makedirs(tmpdir)
if not outgroup:
subprocess.call(['raxmlHPC-PTHREADS', '-T', str(args.cpus), '-f', 'a', '-m', 'GTRGAMMA', '-p', '12345', '-x', '12345', '-#', str(args.bootstrap), '-s', os.path.abspath(file_in), '-n', 'snps.nwk'], cwd = tmpdir, stdout = FNULL, stderr = FNULL)
else:
subprocess.call(['raxmlHPC-PTHREADS', '-T', str(args.cpus), '-f', 'a', '-m', 'GTRGAMMA', '-p', '12345', '-x', '12345', '-#', str(args.bootstrap), '-s', os.path.abspath(file_in), '-o', outgroup, '-n', 'snps.nwk'], cwd = tmpdir, stdout = FNULL, stderr = FNULL)
#parse with biopython and draw
trees = list(Phylo.parse(os.path.join(tmpdir, 'RAxML_bootstrap.snps.nwk'), 'newick'))
best = Phylo.read(os.path.join(tmpdir,'RAxML_bestTree.snps.nwk'), 'newick')
support_tree = get_support(best, trees)
Phylo.write(support_tree, args.input+'.snps.phylogeny.nwk', 'newick')
Phylo.draw(support_tree, do_show=False)
pylab.axis('off')
pylab.savefig(args.input+'.snps.phylogeny.pdf', format='pdf', bbox_inches='tight', dpi=1000)
subprocess.call(['trimal', '-in', mafft_out, '-out', trimal_out, '-automated1', '-phylip'], stderr = FNULL, stdout = FNULL)
print "Running RAxML"
if args.method == 'prot':
method = 'PROTGAMMAAUTO'
elif args.method == 'nucl':
method = 'GTRGAMMA'
if args.cpus == 1:
if not args.outgroup:
subprocess.call(['raxmlHPC-PTHREADS', '-f', 'a', '-m', method, '-p', '12345', '-x', '12345', '-#', str(args.bootstrap), '-s', trimal_out, '-n', 'nwk'], cwd = tmp)
else:
subprocess.call(['raxmlHPC-PTHREADS', '-f', 'a', '-m', method, '-o', args.outgroup, '-p', '12345', '-x', '12345', '-#', str(args.bootstrap), '-s', trimal_out, '-n', 'nwk'], cwd = tmp)
else:
if not args.outgroup:
subprocess.call(['raxmlHPC-PTHREADS', '-T', str(args.cpus), '-f', 'a', '-m', method, '-p', '12345', '-x', '12345', '-#', str(args.bootstrap), '-s', trimal_out, '-n', 'nwk'], cwd = tmp)
else:
subprocess.call(['raxmlHPC-PTHREADS', '-T', str(args.cpus), '-f', 'a', '-m', method, '-o', args.outgroup, '-p', '12345', '-x', '12345', '-#', str(args.bootstrap), '-s', trimal_out, '-n', 'nwk'], cwd = tmp)
print "Drawing tree inferred from RAxML"
#parse with biopython and draw
trees = list(Phylo.parse(os.path.join(tmp, 'RAxML_bootstrap.nwk'), 'newick'))
best = Phylo.read(os.path.join(tmp, 'RAxML_bestTree.nwk'), 'newick')
support_tree = get_support(best, trees)
Phylo.write(support_tree, args.out.split('.')[0]+'.nwk', 'newick')
Phylo.draw(support_tree, do_show=False)
pylab.axis('off')
pylab.savefig(args.out, format='pdf', bbox_inches='tight', dpi=1000)
if not args.debug:
shutil.rmtree(tmp)