def run_tool(args):
logging.debug('Arguments {}'.format(args))
# init a Patient
patient_data = Patient.Patient(indiv_name=args.indiv_id,
driver_genes_file=args.driver_genes_file)
# Patient load cluster and mut ccf files
parse_sif_file(args.sif, args.mutation_ccf_file, patient_data)
load_clustering_results(args.cluster_ccf_file, patient_data)
tree_edges = load_tree_edges_file(args.tree_tsv)
bt_engine = BuildTreeEngine(patient_data)
tree = Tree()
tree.init_tree_from_clustering(patient_data.ClusteringResults)
tree.set_new_edges(tree_edges)
patient_data.TopTree = tree
patient_data.TopTree = bt_engine.top_tree
patient_data.TreeEnsemble = bt_engine.trees
gk_engine = GrowthKineticsEngine(patient_data)
gk_engine.estimate_growth_rate(args.n_iter)
"""
def run_tool(args):
logging.debug('Arguments {}'.format(args))
import data.Patient as Patient
from .GrowthKineticsEngine import GrowthKineticsEngine
patient_data = Patient.Patient(indiv_name=args.indiv_id)
mcmc_trace_cell_abundance, num_itertaions = load_mcmc_trace_abundances(
args.abundance_mcmc_trace)
with open(args.sif, 'r') as f:
header = f.readline().strip('\n').split('\t')
sample_time_points = {}
for line in f:
fields = dict(zip(header, line.strip('\n').split('\t')))
try:
sample_time_points[fields['sample_id']] = float(
fields['timepoint'])
except ValueError:
raise ValueError('Sample time points in sif file are required')
wbc_time_points = np.array(args.time) if args.time is not None else None
gk_engine = GrowthKineticsEngine(patient_data, args.wbc)
gk_engine.estimate_growth_rate(mcmc_trace_cell_abundance,
wbc_time_points=wbc_time_points,
sample_time_points=sample_time_points,
n_iter=min(num_itertaions, args.n_iter))
# Output and visualization
import output.PhylogicOutput
phylogicoutput = output.PhylogicOutput.PhylogicOutput()
phylogicoutput.write_growth_rate_tsv(gk_engine.growth_rates, args.indiv_id)
phylogicoutput.plot_growth_rates(gk_engine.growth_rates, args.indiv_id)
def run_tool(args):
logging.debug('Arguments {}'.format(args))
import data.Patient as Patient
from .CellPopulationEngine import CellPopulationEngine
from .BuildTreeEngine import BuildTreeEngine
import output.PhylogicOutput
# init a Patient
patient_data = Patient.Patient(indiv_name=args.indiv_id,
driver_genes_file=args.driver_genes_file)
# Patient load cluster and mut ccf files
parse_sif_file(args.sif, args.mutation_ccf_file, patient_data)
load_clustering_results(args.cluster_ccf_file, patient_data,
args.blacklist_threshold, args.blacklist_cluster)
patient_data.preprocess_samples()
# Building Phylogenetic Tree
bt_engine = BuildTreeEngine(patient_data)
bt_engine.build_tree(n_iter=args.n_iter)
# Output and visualization
phylogicoutput = output.PhylogicOutput.PhylogicOutput()
# Assign Top tree to Patient
patient_data.TopTree = bt_engine.top_tree
patient_data.TreeEnsemble = bt_engine.mcmc_trace
phylogicoutput.write_tree_tsv(bt_engine.mcmc_trace, args.indiv_id)
# TODO: Fix this
# patient_data.TreeEnsemble = bt_engine.trees
# Computing Cell Population
cp_engine = CellPopulationEngine(patient_data)
constrained_ccf = cp_engine.compute_constrained_ccf(n_iter=args.n_iter)
cell_abundance = cp_engine.get_cell_abundance_across_samples(
constrained_ccf)
phylogicoutput.write_all_cell_abundances(
cp_engine.get_all_cell_abundances(), args.indiv_id)
cell_ancestry = bt_engine.get_cell_ancestry()
phylogicoutput.write_constrained_ccf_tsv(constrained_ccf, cell_ancestry,
args.indiv_id)
phylogicoutput.write_cell_abundances_tsv(cell_abundance, cell_ancestry,
args.indiv_id)
phylogicoutput.generate_html_from_tree(
args.mutation_ccf_file,
args.cluster_ccf_file,
args.indiv_id + '_build_tree_posteriors.tsv',
args.indiv_id + '_constrained_ccf.tsv',
sif=args.sif,
drivers=patient_data.driver_genes,
treatment_file=args.treatment_data,
tumor_sizes_file=args.tumor_size,
cnv_file=args.indiv_id + '.cnvs.txt')
def run_tool(args):
logging.debug('Arguments {}'.format(args))
import data.Patient as Patient
from .CellPopulationEngine import CellPopulationEngine
from .BuildTreeEngine import BuildTreeEngine
from .ClusteringResults import ClusteringResults
# init a Patient
patient_data = Patient.Patient(indiv_name=args.indiv_id,
driver_genes_file=args.driver_genes_file)
# Patient load cluster and mut ccf files
clustering_results = ClusteringResults(args.mutation_ccf_file,
args.cluster_ccf_file)
patient_data.ClusteringResults = clustering_results
bt_engine = BuildTreeEngine(patient_data)
bt_engine.build_tree(args.n_iter)
patient_data.TopTree = bt_engine.top_tree
patient_data.TreeEnsemble = bt_engine.trees
cp_engine = CellPopulationEngine(patient_data)
constrained_ccf = cp_engine.samples_average_constrained_ccf()
cell_ancestry = bt_engine.get_cell_ancestry()
cell_abundance = cp_engine.get_cell_abundance(constrained_ccf)
# Output and visualization
import output.PhylogicOutput
phylogicoutput = output.PhylogicOutput.PhylogicOutput()
phylogicoutput.write_tree_tsv(bt_engine.mcmc_trace, bt_engine.trees_ll,
args.indiv_id)
phylogicoutput.write_constrained_ccf_tsv(constrained_ccf, cell_ancestry,
args.indiv_id)
phylogicoutput.write_cell_abundances_tsv(cell_abundance, cell_ancestry,
args.indiv_id)
phylogicoutput.generate_html_from_tree(
args.mutation_ccf_file,
args.cluster_ccf_file,
args.indiv_id + '_build_tree_posteriors.tsv',
args.indiv_id + '_constrained_ccf.tsv',
sif=args.sif,
drivers=patient_data.driver_genes,
treatment_file=args.treatment_data,
tumor_sizes_file=args.tumor_size,
cnv_file=args.indiv_id + '.cnvs.txt')
def run_tool(args):
logging.debug('Arguments {}'.format(args))
import data.Patient as Patient
from .Tree import Tree
from .CellPopulationEngine import CellPopulationEngine
from .BuildTreeEngine import BuildTreeEngine
# init a Patient
patient_data = Patient.Patient(indiv_name=args.indiv_id,
driver_genes_file=args.driver_genes_file)
# try: # if sif file is specified
# Patient load cluster and mut ccf files
parse_sif_file(args.sif, args.mutation_ccf_file, patient_data)
load_clustering_results(args.cluster_ccf_file, patient_data)
tree_edges = load_tree_edges_file(args.tree_tsv)
bt_engine = BuildTreeEngine(patient_data)
tree = Tree()
tree.init_tree_from_clustering(patient_data.ClusteringResults)
tree.set_new_edges(tree_edges)
patient_data.TopTree = tree
# Computing Cell Population
cp_engine = CellPopulationEngine(patient_data)
constrained_ccf = cp_engine.compute_constrained_ccf()
cell_ancestry = bt_engine.get_cell_ancestry()
cell_abundance = cp_engine.get_cell_abundance(constrained_ccf)
# Output and visualization
import output.PhylogicOutput
phylogicoutput = output.PhylogicOutput.PhylogicOutput()
# TODO write cell population MCMC trace to file
phylogicoutput.write_all_cell_abundances(
cp_engine.get_all_cell_abundances(), args.indiv_id)
phylogicoutput.write_constrained_ccf_tsv(constrained_ccf, cell_ancestry,
args.indiv_id)
phylogicoutput.write_cell_abundances_tsv(cell_abundance, cell_ancestry,
args.indiv_id)
phylogicoutput.generate_html_from_tree(
args.mutation_ccf_file,
args.cluster_ccf_file,
args.indiv_id + '_build_tree_posteriors.tsv',
args.indiv_id + '_constrained_ccf.tsv',
sif=args.sif,
drivers=patient_data.driver_genes,
treatment_file=args.treatment_data,
tumor_sizes_file=args.tumor_size,
cnv_file=args.indiv_id + '.cnvs.txt')
def addPatient(self, name, ipp, MainWindow):
"""Method to add and save a new patient using the fields"""
#Confirmation message
qm = QtWidgets.QMessageBox()
reply = qm.question(
MainWindow, 'Continuer ?',
'Voulez vraiment ajouter ' + name + "(" + ipp + ") ?", qm.Yes,
qm.No)
if reply == qm.Yes:
new = data.deserialize()
new.getListPatient().append(Patient(name, ipp))
data.seralize(new)
self.syncData()
#Clear fields
self.lineEdit_name.setText("")
self.lineEdit_ipp.setText("")
def run_tool(args):
logging.debug('Arguments {}'.format(args))
import data.Patient as Patient
from .GrowthKineticsEngine import GrowthKineticsEngine
patient_data = Patient.Patient(indiv_name=args.indiv_id)
mcmc_trace_cell_abundance, num_itertaions = load_mcmc_trace_abundances(
args.abundance_mcmc_trace)
gk_engine = GrowthKineticsEngine(patient_data, args.wbc)
gk_engine.estimate_growth_rate(mcmc_trace_cell_abundance,
n_iter=min(num_itertaions, args.n_iter))
# Output and visualization
import output.PhylogicOutput
phylogicoutput = output.PhylogicOutput.PhylogicOutput()
phylogicoutput.write_growth_rate_tsv(gk_engine.growth_rates, args.indiv_id)
phylogicoutput.plot_growth_rates(gk_engine.growth_rates, args.indiv_id)
def run_tool(args):
print(args)
if not args.maf and not args.html:
print("No output type specified, specify any combination of --html and --maf")
# sys.path.append(args.phylogic_path)
# data storage objects
import data.Patient as Patient
import data.Sample as Sample
import ClusterEngine
import DpEngine
if not os.path.isfile(args.PoN):
logging.warning("PanelofNormals (PoN) inaccessible or not specified - not using PoN!")
PoN = False
else:
PoN = args.PoN
# init a Patient
patient_data = Patient.Patient(artifact_blacklist=args.artifact_blacklist,
PoN_file=PoN, indiv_name=args.indiv_id, artifact_whitelist=args.artifact_whitelist,
min_coverage=args.min_cov, use_indels=args.use_indels,
impute_missing=args.impute_missing,
driver_genes_file=args.driver_genes_file)
# delete_auto_bl=args.Delete_Blacklist,
# Load sample data
if args.sif: # if sif file is specified
sif_file = open(args.sif)
for file_idx, file_line in enumerate(sif_file):
##for now, assume input file order is of the type sample_id\tmaf_fn\tseg_fn\tpurity\ttimepoint
if not file_idx:
continue
if file_line.strip('\n').strip == "":
continue # empty rows
smpl_spec = file_line.strip('\n').split('\t')
sample_id = smpl_spec[0]
maf_fn = smpl_spec[1]
seg_fn = smpl_spec[2]
purity = float(smpl_spec[3])
timepoint = float(smpl_spec[4])
print(timepoint)
patient_data.addSample(maf_fn, sample_id, timepoint_value=timepoint, grid_size=args.grid_size,
_additional_muts=None, seg_file=seg_fn,
purity=purity, input_type=args.maf_input_type)
if len(patient_data.sample_list) == args.n_samples: # use only first N samples
break
else: # if sample names/files are specified directly on cmdline
# sort order on timepoint or order of entry on cmdline if not present
print(args.sample_data)
for idx, sample_entry in enumerate(args.sample_data):
##for now, assume input order is of the type sample_id\tmaf_fn\tseg_fn\tpurity\ttimepoint
smpl_spec = sample_entry.strip('\n').split(':')
sample_id = smpl_spec[0]
maf_fn = smpl_spec[1]
seg_fn = smpl_spec[2]
purity = float(smpl_spec[3])
timepoint = float(smpl_spec[4])
patient_data.addSample(maf_fn, sample_id, timepoint_value=timepoint, grid_size=args.grid_size,
_additional_muts=None,
seg_file=seg_fn,
purity=purity)
if len(patient_data.sample_list) == args.n_samples: # use only first N samples
break
patient_data.get_arm_level_cn_events()
patient_data.preprocess_samples()
# TODO: how 1D (one sample) is handeled
DP_Cluster = ClusterEngine.ClusterEngine(patient_data) # html_out=args.indiv_id + '.html')
DP_Cluster.run_DP_ND(N_iter=args.iter, PriorK={'r': args.Pi_k_r, 'mu': args.Pi_k_mu},
mode="maxpear", seed=args.seed)
patient_data.ClusteringResults = DP_Cluster.results
patient_data.cluster_temp_removed()
# sample_id = args.indiv_id
# Output and visualization
import output.PhylogicOutput
phylogicoutput = output.PhylogicOutput.PhylogicOutput()
cluster_ccfs = {i + 1: ccf_hists for i, ccf_hists in enumerate(patient_data.ClusteringResults.clust_CCF_dens)}
phylogicoutput.write_patient_cluster_ccfs(patient_data, cluster_ccfs)
phylogicoutput.write_patient_mut_ccfs(patient_data, cluster_ccfs)
phylogicoutput.write_patient_cnvs(patient_data, cluster_ccfs)
phylogicoutput.write_patient_unclustered_events(patient_data)
phylogicoutput.plot_1d_clusters('{}.cluster_ccfs.txt'.format(patient_data.indiv_name))
phylogicoutput.plot_1d_mutations('{}.mut_ccfs.txt'.format(patient_data.indiv_name))
if not args.buildtree: # run only Clustering tool
phylogicoutput.generate_html_from_clustering_results(patient_data.ClusteringResults, patient_data,
drivers=patient_data.driver_genes,
treatment_file=args.treatment_data)
else: # run build tree next
import BuildTree.BuildTree
args.cluster_ccf_file = '{}.cluster_ccfs.txt'.format(patient_data.indiv_name)
args.mutation_ccf_file = '{}.mut_ccfs.txt'.format(patient_data.indiv_name)
args.n_iter = args.iter
args.blacklist_cluster = None
BuildTree.BuildTree.run_tool(args)