def generate_sequences(self,
species_name,
samples_per_pop=10,
seq_len=2000,
use_seq_gen=True):
self.generate_pop_tree(species_name=species_name, samples_per_pop=samples_per_pop)
self.generate_gene_tree(species_name=species_name, samples_per_pop=samples_per_pop)
d = dendropy.DataSet(self.mutation_tree.taxon_set)
if SEQGEN and use_seq_gen:
sg = seqgen.SeqGen()
sg.seqgen_path = self.seqgen_path
sg.num_replicates = 1
sg.quiet = True
sg.rng = self.rng
sg.seq_len = seq_len
sg.char_model = 'HKY'
sg.ti_tv = float(self.kappa) / 2
sg.state_freqs = self.base_freqs
sg.trees = [self.mutation_tree]
d = sg.generate_dataset(dataset=d)
return d
else:
return seqsim.generate_hky_dataset(seq_len=seq_len,
tree_model=self.mutation_tree,
mutation_rate=1.0,
kappa=1.0,
base_freqs=[0.25, 0.25, 0.25, 0.25],
root_states=None,
dataset=d,
rng=self.rng)
def estimate_params(self,
seq_len=10000,
kappa=1.0,
base_freqs=[0.25, 0.25, 0.25, 0.25],
unequal_base_freqs=True,
gamma_rates=False,
prop_invar=False):
output_ds = seqsim.generate_hky_dataset(seq_len,
tree_model=self.tree_model,
kappa=kappa,
base_freqs=base_freqs)
self.tree_model.reindex_taxa(output_ds.char_matrices[0].taxon_set)
est_tree, mle = paup.estimate_model(char_matrix=output_ds.char_matrices[0],
tree_model=self.tree_model,
num_states=2,
unequal_base_freqs=unequal_base_freqs,
gamma_rates=gamma_rates,
prop_invar=prop_invar,
tree_est_criterion="likelihood",
tree_user_brlens=True,
paup_path='paup')
return mle
