def test_replacing_hgt(self):
N = 20
S = te.simulate_species_tree(N, model='innovation')
# true gene tree (with losses)
TGT = te.simulate_dated_gene_tree(
S,
dupl_rate=0.0,
loss_rate=0.0,
hgt_rate=1.0,
prohibit_extinction='per_species',
replace_prob=1.0,
)
# observable gene tree
OGT = te.observable_tree(TGT)
leaves = [v for v in OGT.leaves()]
colors = {v.color for v in leaves}
# print(TGT.to_newick())
# print(OGT.to_newick())
self.assertTrue(len(colors) == N and len(leaves) == N)
def test_no_extinction(self):
N = 10
repeats = 20
for _ in range(repeats):
species_tree = te.simulate_species_tree(N,
model='innovation',
non_binary=0.2)
gene_tree = te.simulate_dated_gene_tree(
species_tree,
dupl_rate=1.0,
loss_rate=1.0,
hgt_rate=0.5,
prohibit_extinction='per_species')
# check that there is no extinction in any species
color_dict = {
l.label: []
for l in species_tree.preorder()
if not l.children and l.event != 'L'
}
for v in gene_tree.preorder():
if not v.children and v.event != 'L':
color_dict[v.color].append(v.label)
for leaf_list in color_dict.values():
self.assertTrue(leaf_list)
gene_tree2 = te.simulate_dated_gene_tree(
species_tree,
dupl_rate=1.0,
loss_rate=1.0,
hgt_rate=0.5,
prohibit_extinction='per_family')
# check that there is no extinction in all species
self.assertTrue([l for l in gene_tree2.leaves()])
def test_rs_edges(self):
S = te.simulate_species_tree(10)
TGT = te.simulate_dated_gene_tree(S,
dupl_rate=1.0,
loss_rate=0.5,
hgt_rate=0.5)
OGT = te.observable_tree(TGT)
transf1 = analysis.true_transfer_edges(OGT)
transf2 = analysis.rs_transfer_edges(OGT, S)
self.assertTrue(transf1.issuperset(transf2))
def generate_solutions_unique_species(n, i_p=0.5, d_p=0.5):
done = False
count = 0
while not done:
S = te.simulate_species_tree(10, model='innovation')
TGT = te.simulate_dated_gene_tree(S,
dupl_rate=0.5,
loss_rate=0.5,
hgt_rate=0.5,
prohibit_extinction="per_family",
replace_prob=0.0)
OGT = te.observable_tree(TGT)
ldt = ldt_graph(OGT, S)
if len(ldt.nodes()) == n:
IG = InvestigateGraph(ldt)
IG.perturb_graph(i_p, d_p)
solver = LDTEditor(IG._G_perturbed)
solver.build_model()
solver.optimize(time_limit=None)
sol_graph, sol_distance = solver.get_solution()
properly_colored = is_properly_colored(sol_graph)
cograph = is_cograph(sol_graph)
compatible = is_compatible(sol_graph)
edit_dist = gt.symmetric_diff(IG._G_perturbed, sol_graph)
print("Runtime: {}".format(solver.get_solve_time()))
if properly_colored and cograph and compatible:
print("Saving data...")
solver._save_ILP_data(
IG._G_perturbed,
sol_graph,
solver.get_solve_time(),
edit_dist,
only_add=False,
only_delete=False,
filename="{}nodes/LDTEdit_exact_solution".format(n))
else:
print("No solution found!")
count += 1
if count == 100:
done = True
def generate_trees(n=100,
m=10,
model='innovation',
dupl_rate=0.5,
loss_rate=0.5,
hgt_rate=0.5,
prohibit_extinction="per_family",
replace_prob=0.0,
size=10):
i = 0
dirName = 'exact_solutions/trees/{}trees'.format(size)
# create folder if it doesnt exist
if not os.path.exists(dirName):
os.makedirs(dirName)
ID = 0
else:
ID = find_next_ID('exact_solutions/trees/{}trees/'.format(size))
while i < n:
S = te.simulate_species_tree(m, model=model)
TGT = te.simulate_dated_gene_tree(
S,
dupl_rate=dupl_rate,
loss_rate=loss_rate,
hgt_rate=hgt_rate,
prohibit_extinction=prohibit_extinction,
replace_prob=replace_prob)
OGT = te.observable_tree(TGT)
ldt = ldt_graph(OGT, S)
amount_nodes = len(ldt.nodes())
if amount_nodes == size:
# save trees
filename_species = 'exact_solutions/trees/{}trees/species_{}_{}_{}.json'.format(
size, m, model, ID)
filename_gene = 'exact_solutions/trees/{}trees/gene_{}_{}_{}_{}_{}_{}.json'.format(
size, dupl_rate, loss_rate, hgt_rate, prohibit_extinction,
replace_prob, ID)
S.serialize(filename_species)
TGT.serialize(filename_gene)
ID += 1
i += 1
def test_ldt_fitch(self):
S = te.simulate_species_tree(20, model='innovation')
# true gene tree (with losses)
TGT = te.simulate_dated_gene_tree(S,
dupl_rate=1.0,
loss_rate=0.5,
hgt_rate=0.2)
# observable gene tree
OGT = te.observable_tree(TGT)
# finally we can extract the LDT and Fitch graph
ldt = analysis.ldt_graph(OGT, S)
transfer_edges = analysis.rs_transfer_edges(OGT, S)
fitch = analysis.undirected_fitch(OGT, transfer_edges)
cotree = to_cotree(ldt)
self.assertTrue(gt.is_subgraph(ldt, fitch) and cotree)
# -*- coding: utf-8 -*-
import tralda.tools.GraphTools as gt
import asymmetree.treeevolve as te
from asymmetree.analysis import (undirected_fitch,
rs_transfer_edges,
below_equal_above,
ldt_graph,
RsScenarioConstructor,)
from asymmetree.tools.PhyloTreeTools import (to_newick,)
S = te.simulate_species_tree(10)
TGT = te.simulate_dated_gene_tree(S, dupl_rate=1.0, loss_rate=0.5,
hgt_rate=0.5)
OGT = te.observable_tree(TGT)
print('--- S ---\n', to_newick(S))
print(to_newick(S, distance=False, label_inner=False))
print('--- OGT ---\n', to_newick(OGT))
ldt, above, equal = below_equal_above(OGT, S)
fitch = undirected_fitch(OGT, rs_transfer_edges(OGT, S))
n = ldt.order()
print('Genes:', n, 'Total relations:', int(n * (n-1) / 2))
print('< {}\n= {}\n> {}'.format(ldt.size(), equal.size(), above.size()))
rs_scen_constr = RsScenarioConstructor(ldt)
result = rs_scen_constr.run()
if result:
# build loop
len(parameter_Df.index)
for ind in range(len(parameter_Df.index)-1):
# species tree of type ’PhyloTree’
s = te.simulate_species_tree(int(parameter_Df.loc[ind, 'num_of_leaves']),
model = parameter_Df.loc[ind, 'model'],
non_binary_prob = parameter_Df.loc[ind, 'non_binary_prob'],
planted = parameter_Df.loc[ind, 'planted'],
remove_extinct = parameter_Df.loc[ind, 'remove_extinct'],
rescale_to_height = parameter_Df.loc[ind, 'rescale_to_height']
)
# true gene tree (contains losses) of type ’PhyloTree’
tgt = te.simulate_dated_gene_tree(s,
dupl_rate = parameter_Df.loc[ind, 'dupl_rate'],
loss_rate = parameter_Df.loc[ind, 'loss_rate'],
hgt_rate = parameter_Df.loc[ind, 'hgt_rate'],
dupl_polytomy = 0.0,
prohibit_extinction= parameter_Df.loc[ind, 'prohibit_extinction'],
replace_prob = parameter_Df.loc[ind, 'replace_prob']
)
# serialization
s.serialize(wk_dir / '01_Data' / str(parameter_Df.loc[ind, 'ID'] + '_species_tree.pickle'))
tgt.serialize(wk_dir / '01_Data' / str(parameter_Df.loc[ind, 'ID'] + '_gene_tree.pickle'))
print('Simulating Tree :', ind)
ogt = te.observable_tree(tgt)
from tools.GraphTools import *
from tools.plotTools import *
import networkx as nx
import asymmetree.treeevolve as te
from asymmetree.datastructures import PhyloTree
from asymmetree.hgt import ldt_graph
from tools.LDT_ILP import LDTEditor
import asymmetree.tools.GraphTools as gt
import os
S = te.simulate_species_tree(20, model='innovation')
TGT = te.simulate_dated_gene_tree(S,
dupl_rate=0.5,
loss_rate=0.5,
hgt_rate=0.5,
prohibit_extinction="per_family",
replace_prob=0.0)
OGT = te.observable_tree(TGT)
ldt = ldt_graph(OGT, S)
colors = gt.sort_by_colors(ldt)
#print("edges of G: \n{}".format(G._G.edges()))
#a, b, c = get_P3_data(G._G)
#print("\nThe regions of P3s: {}".format(a))
#print("\nThe amounts in the regions: {}".format(b))
#print("\nThe distance between regions: {}\n".format(c))
print("Amount of nodes: {}".format(len(ldt.nodes())))
print("Amount of colors: {}".format(len(colors)))
print("Amount of edges: {}".format(len(ldt.edges())))
