def run_pal2nal(fname_aln, fname_nuc, fname_prot):
"""
Generate a codon alignment via PAL2NAL.
@param fname_aln:
MSA of protein sequences in CLUSTAL format (.aln)
@param fname_nuc:
Nucleotide sequences in FASTA format (.fasta)
@param fname_prot:
Protein sequences in FASTA format (.fasta)
@return:
Codon alignment in CLUSTAL format (.aln), suitable for codeml
1"""
sys.stderr.write("\nSTEP: run_pal2nal(%s, %s)\n" % (fname_aln, fname_nuc))
# Reorder fname_nuc according to the order of the proteins in fname_aln, which
# was reordered due to CLUSTALW2. Note that the first protein in each of
# these files remains the same as at the start, however; this first protein
# is our original query protein.
nuc_records = [record for record in SeqIO.parse(fname_nuc, "fasta")]
prot_records = [record for record in SeqIO.parse(fname_prot, "fasta")]
records_map = dict((pr.id, nr) for pr, nr in zip(prot_records, nuc_records))
fname_nuc2 = "homologs_ordered.dna.fasta"
with open(fname_nuc2, "w") as f:
for record in SeqIO.parse(fname_aln, "clustal"):
SeqIO.write(records_map[record.id], f, "fasta")
fname_codon = "homologs.codon.aln"
# TODO: use subprocess
os.system("%s/pal2nal.pl %s %s -output paml > %s" % (bin_dir(), fname_aln, fname_nuc2, fname_codon))
return fname_codon
def run_clustalw2(fname_prot):
"""
Generate a MSA of the amino acids (in fasta format) via clustalw.
@param fname_prot:
Protein sequences in FASTA format (.fasta)
@return:
MSA of protein sequences in CLUSTAL format (.aln)
"""
sys.stderr.write("\nSTEP: run_clustalw2(%s)\n" % fname_prot)
fname_aln = "homologs.aa.aln"
fname_log = "clustalw2.log"
sys.stderr.write("\tRunning clustalw2, please be patient (may take minutes)...\n")
proc = subprocess.Popen(
"%s/clustalw2 -INFILE=%s -OUTFILE=%s" % (bin_dir(), os.path.abspath(fname_prot), os.path.abspath(fname_aln)),
stdout=subprocess.PIPE,
stderr=subprocess.PIPE,
shell=True,
)
stdout, stderr = proc.communicate()
with open(fname_log, "w") as f_log:
f_log.write(stderr)
sys.stderr.write("\tclustalw2 run successful, log available at %s\n" % os.path.abspath(fname_log))
# Remove the extra file...
if "." in os.path.split(fname_prot)[-1]:
fname_dnd = ".".join(fname_prot.split(".")[:-1]) + ".dnd"
else:
fname_dnd = fname_prot + ".dnd"
os.remove(fname_dnd)
return fname_aln
def run_codeml(fname_ctl):
"""
XXX
Input:
fname_ctl
Output:
codeml output, currently
"""
os.system("%s/codeml %s" % (bin_dir(), fname_ctl))
def run_phyml(fname_aln, n_bootstrap):
"""
Generate a phylogenetic tree via PHYML.
@param fname_aln:
MSA of protein sequences in CLUSTAL format (.aln)
@return:
(tree_file, bootstrap_file) = File of phylo tree with clade confidences (_tree.txt),
file of bootstrapped phylo trees (_boot_trees.txt)
"""
sys.stderr.write("\nSTEP: run_phyml(%s, %s)\n" % (fname_aln, n_bootstrap))
fname_phy = "homologs.aa.phy"
with open(fname_aln, "rU") as f_in:
with open(fname_phy, "w") as f_out:
SeqIO.convert(f_in, "clustal", f_out, "phylip-relaxed")
current_dir = os.getcwd()
fname_tree = fname_phy + "_phyml_tree.txt"
if n_bootstrap > 1:
bootstrap_str = "-b %d" % n_bootstrap
fname_boot_trees = fname_phy + "_phyml_boot_trees.txt"
else:
bootstrap_str = ""
fname_boot_trees = None
fname_log = "phyml.log"
sys.stderr.write("\tRunning phyml, please be patient (may take minutes)...\n")
proc = subprocess.Popen(
"%s/phyml -i %s -d aa %s %s" % (bin_dir(), os.path.abspath(fname_phy), bootstrap_str, current_dir),
stdout=subprocess.PIPE,
stderr=subprocess.PIPE,
shell=True,
)
stdout, stderr = proc.communicate()
with open(fname_log, "w") as f_log:
f_log.write(stderr)
sys.stderr.write("\tclustalw2 run successful, log available at %s\n" % os.path.abspath(fname_log))
return fname_tree, fname_boot_trees
