def from_newick(cls, newick_file, config={}, **kwargs):
"""Read tree from newick file."""
try:
import phylopandas as pd
df = pd.read_newick(newick_file)
data = df.to_dict(orient='records')
return cls(data=data, config={}, **kwargs)
except ImportError:
Exception("DendroPy and Phylopandas must be installed.")
def compute_tree(
self,
sequence_col='sequence',
datatype='aa',
bootstrap='-1',
model='LG',
frequencies='e',
**kwargs):
"""Compute tree."""
fname = "compute_tree.phy"
path = os.path.join(self.project_dir, fname)
# Write out path
self.data.phylo.to_phylip(filename=path)
# Prepare options for PhyML.
options = {
'input':path,
'datatype':datatype,
'bootstrap':bootstrap,
'model':model,
'frequencies':frequencies,
}
# Update with any kwargs manually set by users.
options.update(**kwargs)
# ----- Flexibility here to use different ext apps -----
# Build command line arguments for PhyML.
cml = Bio.Phylo.Applications.PhymlCommandline(**options)
cml_args = str(cml).split()
output = subprocess.run(cml_args)
# Get path (catch variations in file extension generated by phyml)
outfile = "compute_tree.phy_phyml_tree"
outpath = os.path.join(self.project_dir, outfile)
# Old versions of phyml have .txt at end.
if not os.path.exists(outpath):
outpath += '.txt'
# ------------------------------------------------------
# Update dataframe
tree_data = ph.read_newick(outpath)
# Swap ids and uids for leaf nodes
tree_data.loc[tree_data.type == 'leaf', 'uid'] = tree_data.id
# Add to main dataframe
self.data = self.data.phylo.combine(tree_data, on='uid')
def show(self):
"""Show rough interactive tree.
"""
# Write a dummy tree
tree_path = os.path.join(self.project_dir, 'vega-tree.newick')
self.data['tree'].write(path=tree_path, schema='newick')
tree_df = phylopandas.read_newick(tree_path)
# Create vega tree
vegatree = phylovega.VegaTree(tree_df)
# Write vega tree.
vega_path = os.path.join(self.project_dir, 'vega-tree.vg.json')
vegatree.to_json(vega_path)
# Display vega tree
vegatree.display()
def run(df, project_dir, program="phyml", keep_tmp=False, **kwargs):
tree_functions = {"phyml": _phyml}
# Figure out which alignment function to use
try:
tree_function = tree_functions[program]
except KeyError:
err = "Tree building program '{}' not recognized.\n\n".format(program)
err += "Should be one of:\n"
programs = list(tree_functions.keys())
programs.sort()
for p in programs:
err += " {}\n".format(p)
raise ValueError(err)
input_file = "_"
os.chdir(project_dir)
ph.seqio.write.to_phylip(df, id_col="uid", filename="_")
tree_function(input_file, **kwargs)
# Parse phyml output
new_tree = ph.read_newick("__phyml_tree.txt")
new_tree.loc[new_tree.type == 'leaf', 'uid'] = new_tree.id
# Add to main DataFrame
output = df.phylo.combine(new_tree, on='uid')
if not keep_tmp:
os.remove(input_file)
os.chdir("..")
return output