def pxrr(tree_file, output_dir):
"""Unroot the tree returned by treeshrink."""
unrooted = util.file_name(tree_file)
cmd = ' '.join([
'pxrr', '--unroot', '--treef {}'.format(tree_file),
'--outf {}'.format(unrooted)
])
with util.cd(output_dir):
subprocess.check_call(cmd, shell=True)
util.remove_files('phyx.logfile')
return unrooted
def raxml(fasta_file, output_dir, output_ext, seq_type, cpus, seed):
"""Build a tree with raxml."""
model = "PROTCATWAG" if seq_type == "aa" else "GTRCAT"
tree = util.file_name(fasta_file, output_ext)
cmd = ' '.join([
'raxml', '-T {}'.format(cpus), '-p {}'.format(seed),
'-m {}'.format(model), '-s {}'.format(fasta_file), '-n {}'.format(tree)
])
with util.cd(output_dir):
subprocess.check_call(cmd, shell=True)
tree_src = 'RAxML_bestTree.' + tree
move(tree_src, tree)
util.remove_files('RAxML_*')
return tree
def pasta(fasta_file, output_dir, output_ext, seq_type, cpus):
"""Align sequences."""
in_path = fasta_file
if seq_type == 'aa':
in_path = bio.adjust_aa_seqs(fasta_file, output_dir)
cmd = ' '.join([
which('run_pasta.py'),
'--datatype {}'.format('Protein' if seq_type == 'aa' else 'DNA'),
'--num-cpus {}'.format(cpus), "--input '{}'".format(in_path),
"--output-directory '{}'".format(abspath(output_dir))
])
with util.cd(output_dir):
subprocess.check_call(cmd, shell=True)
base_name = splitext(basename(fasta_file))[0]
temp_aligned = 'pastajob.marker001.' + base_name + EXT
aligned = base_name + output_ext
move(temp_aligned, aligned)
util.remove_files('pastajob*')
return aligned
def pxclsq(fasta_file, output_dir, output_ext, seq_type, min_occupancy,
min_len):
"""Filter aligned sequences for occupancy and length."""
ext = output_ext + EXT_PXCLSQ
temp_cleaned = util.file_name(fasta_file, ext)
cmd = ' '.join([
'pxclsq', '--aminoacid' if seq_type == 'aa' else '',
'--prop {}'.format(min_occupancy), '--seqf {}'.format(fasta_file),
'--outf {}'.format(basename(temp_cleaned))
])
cleaned = util.file_name(fasta_file, output_ext)
with util.cd(output_dir):
subprocess.check_call(cmd, shell=True)
with open(temp_cleaned) as in_file, open(cleaned, 'w') as out_file:
for header, seq in SimpleFastaParser(in_file):
if len(seq.replace('-', '')) >= min_len:
bio.write_fasta_record(out_file, header, seq)
util.remove_files('phyx.logfile')
return cleaned
def raxml_bs(fasta_file,
output_dir,
output_ext,
seq_type,
cpus,
seed,
replicates=100):
"""Build a bootstrapped tree with raxml."""
model = "PROTCATWAG" if seq_type == "aa" else "GTRCAT"
tree = util.file_name(fasta_file, output_ext)
cmd = ' '.join([
'raxml', '-T {}'.format(cpus), '-f a', '-x {}'.format(seed),
'-p {}'.format(seed), '-m {}'.format(model),
'-# {}'.format(replicates), '-s {}'.format(fasta_file),
'-n {}'.format(tree)
])
with util.cd(output_dir):
subprocess.check_call(cmd, shell=True)
tree_src = 'RAxML_bipartitions.' + tree
move(tree_src, tree)
util.remove_files('RAxML_*')
return tree