def reconstruct(tree_file, align_file, csv_file="temp-align.csv"):
if align_file.endswith('.fasta'):
"Convert fasta to csv"
traits = fasta2csv(align_file, csv_file) # convert fasta to csv
data = csv_file # Path to the table containing tip/node annotations, in csv or tab format
else:
df = pd.read_csv(align_file, index_col=0)
traits = [column for column in df.columns]
data = align_file
tree = tree_file # Path to the tree in newick format
# Columns present in the annotation table,
# for which we want to reconstruct ancestral states
columns = traits #['Country']
# Path to the output compressed map visualisation
html_compressed = "tree-000_map.html"
# (Optional) path to the output tree visualisation
html = "tree-000_tree.html"
pastml_pipeline(data=data,
data_sep=',',
columns=columns,
tree=tree,
verbose=True)
parser.add_argument(
'--verbose',
action='store_true',
help="print information on the progress of the analysis")
params = parser.parse_args()
for tree, html in zip(params.trees, params.htmls):
work_dir = tempfile.mkdtemp(
) if not params.work_dir else params.work_dir
pastml_pipeline(data=params.data,
tree=tree,
html_compressed=html,
prediction_method=params.prediction_method,
model=params.model,
columns=params.columns,
name_column=params.name_column,
date_column=params.date_column,
tip_size_threshold=params.threshold,
parameters=params.parameters,
out_data=params.out_data,
work_dir=work_dir,
verbose=params.verbose)
if params.out_parameters:
for column, out_parameters in zip(params.columns,
params.out_parameters):
pastml_out_pars = \
get_pastml_parameter_file(method=params.prediction_method, model=params.model, column=column)
if pastml_out_pars:
copyfile(os.path.join(work_dir, pastml_out_pars),
out_parameters)
if not params.work_dir:
from pastml.acr import pastml_pipeline, COPY
from pastml.ml import MPPA, MAP, JOINT, ALL, ML
from pastml.models.f81_like import EFT, F81, JC
from pastml.parsimony import ACCTRAN, DELTRAN, DOWNPASS, MP
DATA_DIR = os.path.join(os.path.abspath(os.path.dirname(__file__)), 'data')
TREE_NWK = os.path.join(DATA_DIR, 'Albanian.tree.152tax.tre')
STATES_INPUT = os.path.join(DATA_DIR, 'data.txt')
STATES_COPY = os.path.join(DATA_DIR, 'copy_states.csv')
if '__main__' == __name__:
# The initial tree without ACR
pastml_pipeline(data=STATES_INPUT,
tree=TREE_NWK,
html=os.path.join(DATA_DIR, 'trees',
'Albanian_tree_initial.html'),
data_sep=',',
verbose=True,
prediction_method=COPY,
work_dir=os.path.join(DATA_DIR, 'pastml', 'initial'))
# Copy states
pastml_pipeline(data=STATES_COPY,
tree=TREE_NWK,
html=os.path.join(DATA_DIR, 'trees',
'Albanian_tree_{}.html'.format(COPY)),
html_compressed=os.path.join(
DATA_DIR, 'maps', 'Albanian_map_{}.html'.format(COPY)),
data_sep=',',
verbose=True,
prediction_method=COPY,
columns='Country',
work_dir=os.path.join(DATA_DIR, 'pastml', COPY))
import os
from pastml.ml import ML
from pastml.acr import pastml_pipeline
from pastml.visualisation.cytoscape_manager import TIMELINE_LTT
DATA_DIR = os.path.join(os.path.abspath(os.path.dirname(__file__)), 'data')
TREE_NWK = os.path.join(DATA_DIR, 'real', 'raxml_tree.dated.nwk')
STATES_INPUT = os.path.join(DATA_DIR, 'real', 'metadata.tab')
if '__main__' == __name__:
character = 'Location'
pastml_pipeline(data=STATES_INPUT,
tree=TREE_NWK,
prediction_method=ML,
html_compressed=os.path.join(
DATA_DIR, 'maps', 'map_{}.html'.format(character)),
html=os.path.join(DATA_DIR, 'maps',
'tree_{}.html'.format(character)),
timeline_type=TIMELINE_LTT,
verbose=True,
columns=[character],
upload_to_itol=True,
tip_size_threshold=25)
from pastml.acr import pastml_pipeline
dados = "dados.csv"
arvore = "arvore.tre"
#nome e número de colunas deve ser o mesmo do arquivo de dados
colunas = ['Coluna1', 'Coluna2','Coluna3','Coluna4','Coluna5','Coluna6']
for nome_coluna in colunas:
arv_compl = "arv_compl_"+nome_coluna+".html"
arv_compr = "arv_compr_"+nome_coluna+".html"
pastml_pipeline(data = dados,
data_sep = ",",
columns = colunas,
name_column = nome_coluna,
tree = arvore,
html_compressed = arv_compr,
html = arv_compl,
verbose = True)
print("coluna " + nome_coluna+" pronta!")
print("Uhuuul! Terminou :D")
#To run: python3 pastMLGeneReconstruction.py -g $GenePresenceAbsence.csv -t $Tree
import argparse
from pastml.acr import pastml_pipeline
if __name__ == "__main__":
parser = argparse.ArgumentParser()
parser.add_argument(
"-g",
help="Gene presence absence file produced by convertPanarooPastML.py")
parser.add_argument(
"-t",
help=
"Rooted tree containing all and only samples used in the pangenome reconstruction, with exactly the same sample names"
)
args = parser.parse_args()
genePresenceAbsence = open(args.g).readlines()
html_compressed = "~/Documents/abscessus/manchester.samples/squeaky/abscessus/cluster3/compressed.html"
html = "~/Documents/abscessus/manchester.samples/squeaky/abscessus/cluster3/full.html"
columns = genePresenceAbsence[0].strip().split(",")[
1:] #The columns to be reconstructed
pastml_pipeline(data=args.g,
data_sep=",",
columns=columns,
tree=args.t,
html_compressed=html_compressed,
html=html,
verbose=True)
import os
from pastml.acr import pastml_pipeline
from pastml.ml import ML
DATA_DIR = os.path.join(os.path.abspath(os.path.dirname(__file__)), 'data')
TREE_NWK = os.path.join(DATA_DIR, 'best', 'pastml_phyml_tree.nwk')
STATES_INPUT = os.path.join(DATA_DIR, 'metadata.tab')
if '__main__' == __name__:
mutations = ['RT:M184V']
pastml_pipeline(data=STATES_INPUT,
tree=TREE_NWK, prediction_method=ML,
html_compressed=os.path.join(DATA_DIR, 'maps', 'map_{}.html'.format('_'.join(mutations))),
verbose=True, columns=mutations, upload_to_itol=True)
def apply_pastml(id, data, tree, data_sep, id_index, columns, root_date, model,
prediction_method, name_column, html_compressed, html, email,
title, url, work_dir, no_trimming, timeline_type):
try:
from pastml.acr import pastml_pipeline
import os
import shutil
pastml_pipeline(tree=tree,
data=data,
data_sep=data_sep,
id_index=id_index,
columns=columns,
root_date=root_date,
model=model,
prediction_method=prediction_method,
name_column=name_column,
html_compressed=html_compressed,
html=html,
verbose=True,
work_dir=work_dir,
upload_to_itol=True,
itol_id='ZxuhG2okfKLQnsgd5xAEGQ',
itol_project='pastmlweb',
itol_tree_name=id,
tip_size_threshold=np.inf
if no_trimming else REASONABLE_NUMBER_OF_TIPS,
timeline_type=timeline_type)
itol_id = None
itol_id_file = os.path.join(work_dir, 'iTOL_tree_id.txt')
if os.path.exists(itol_id_file):
with open(itol_id_file, 'r') as f:
itol_id = f.readline().strip('\n')
copyfile(
itol_id_file,
os.path.join(work_dir, '..', 'pastml_{}_itol.txt'.format(id)))
if os.path.exists(html_compressed):
copyfile(
html_compressed,
os.path.join(work_dir, 'pastml_compressed_visualisation.html'))
if html and os.path.exists(html):
copyfile(
html,
os.path.join(work_dir, 'pastml_full_tree_visualisation.html'))
shutil.make_archive(
os.path.join(work_dir, '..', 'pastml_{}'.format(id)), 'zip',
work_dir)
try:
shutil.rmtree(work_dir)
except:
pass
if email:
send_analysis_email.delay(email, url, id, title, columns, model,
prediction_method, itol_id, None)
except Exception as e:
e_str = str(e)
with open(html_compressed, 'w+') as f:
f.write('<p>Could not reconstruct the states...<br/>{}</p>'.format(
e_str))
if email:
send_analysis_email.delay(email, url, id, title, columns, model,
prediction_method, None, e_str)
else:
send_analysis_email.delay('*****@*****.**', url, id,
title, columns, model, prediction_method,
None, e_str)
raise e