def write_vcf(chrom):
treefile = args.tree_file[chrom]
vcf = open(args.vcffile[chrom], "w")
mut_rate = args.mut_rate[chrom]
seed = seeds[chrom]
logfile.write("Simulating mutations on" + treefile + "\n")
ts = msprime.load(treefile)
rng = msprime.RandomGenerator(seed)
nodes = msprime.NodeTable()
edgesets = msprime.EdgesetTable()
sites = msprime.SiteTable()
mutations = msprime.MutationTable()
migrations = msprime.MigrationTable()
ts.dump_tables(nodes=nodes, edgesets=edgesets, migrations=migrations)
mutgen = msprime.MutationGenerator(rng, mut_rate)
mutgen.generate(nodes, edgesets, sites, mutations)
logfile.write("Saving to" + args.vcffile[chrom] + "\n")
mutated_ts = msprime.load_tables(nodes=nodes,
edgesets=edgesets,
sites=sites,
mutations=mutations)
mutated_ts.write_vcf(vcf, ploidy=1)
return True
def run(self, ngens):
nodes = msprime.NodeTable()
edges = msprime.EdgeTable()
migrations = msprime.MigrationTable()
sites = msprime.SiteTable()
mutations = msprime.MutationTable()
provenances = msprime.ProvenanceTable()
if self.deep_history:
# initial population
init_ts = msprime.simulate(self.N, recombination_rate=1.0)
init_ts.dump_tables(nodes=nodes, edges=edges)
nodes.set_columns(time=nodes.time + ngens, flags=nodes.flags)
else:
for _ in range(self.N):
nodes.add_row(time=ngens)
pop = list(range(self.N))
for t in range(ngens - 1, -1, -1):
if self.debug:
print("t:", t)
print("pop:", pop)
dead = [random.random() > self.survival for k in pop]
# sample these first so that all parents are from the previous gen
new_parents = [(random.choice(pop), random.choice(pop))
for k in range(sum(dead))]
k = 0
if self.debug:
print("Replacing", sum(dead), "individuals.")
for j in range(self.N):
if dead[j]:
# this is: offspring ID, lparent, rparent, breakpoint
offspring = nodes.num_rows
nodes.add_row(time=t)
lparent, rparent = new_parents[k]
k += 1
bp = self.random_breakpoint()
if self.debug:
print("--->", offspring, lparent, rparent, bp)
pop[j] = offspring
if bp > 0.0:
edges.add_row(left=0.0,
right=bp,
parent=lparent,
child=offspring)
if bp < 1.0:
edges.add_row(left=bp,
right=1.0,
parent=rparent,
child=offspring)
if self.debug:
print("Done! Final pop:")
print(pop)
flags = [(msprime.NODE_IS_SAMPLE if u in pop else 0)
for u in range(nodes.num_rows)]
nodes.set_columns(time=nodes.time, flags=flags)
if self.debug:
print("Done.")
print("Nodes:")
print(nodes)
print("Edges:")
print(edges)
return msprime.TableCollection(nodes, edges, migrations, sites,
mutations, provenances)
def _load_legacy_hdf5_v10(root, remove_duplicate_positions=False):
# We cannot have duplicate positions in v10, so this parameter is ignored
nodes_group = root["nodes"]
nodes = msprime.NodeTable()
metadata = None
metadata_offset = None
if "metadata" in nodes_group:
metadata = nodes_group["metadata"]
metadata_offset = nodes_group["metadata_offset"]
nodes.set_columns(flags=nodes_group["flags"],
population=nodes_group["population"],
time=nodes_group["time"],
metadata=metadata,
metadata_offset=metadata_offset)
edges_group = root["edges"]
edges = msprime.EdgeTable()
edges.set_columns(left=edges_group["left"],
right=edges_group["right"],
parent=edges_group["parent"],
child=edges_group["child"])
migrations_group = root["migrations"]
migrations = msprime.MigrationTable()
if "left" in migrations_group:
migrations.set_columns(left=migrations_group["left"],
right=migrations_group["right"],
node=migrations_group["node"],
source=migrations_group["source"],
dest=migrations_group["dest"],
time=migrations_group["time"])
sites_group = root["sites"]
sites = msprime.SiteTable()
if "position" in sites_group:
metadata = None
metadata_offset = None
if "metadata" in sites_group:
metadata = sites_group["metadata"]
metadata_offset = sites_group["metadata_offset"]
sites.set_columns(
position=sites_group["position"],
ancestral_state=sites_group["ancestral_state"],
ancestral_state_offset=sites_group["ancestral_state_offset"],
metadata=metadata,
metadata_offset=metadata_offset)
mutations_group = root["mutations"]
mutations = msprime.MutationTable()
if "site" in mutations_group:
metadata = None
metadata_offset = None
if "metadata" in mutations_group:
metadata = mutations_group["metadata"]
metadata_offset = mutations_group["metadata_offset"]
mutations.set_columns(
site=mutations_group["site"],
node=mutations_group["node"],
parent=mutations_group["parent"],
derived_state=mutations_group["derived_state"],
derived_state_offset=mutations_group["derived_state_offset"],
metadata=metadata,
metadata_offset=metadata_offset)
provenances_group = root["provenances"]
provenances = msprime.ProvenanceTable()
if "timestamp" in provenances_group:
timestamp = provenances_group["timestamp"]
timestamp_offset = provenances_group["timestamp_offset"]
if "record" in provenances_group:
record = provenances_group["record"]
record_offset = provenances_group["record_offset"]
else:
record = np.empty_like(timestamp)
record_offset = np.zeros_like(timestamp_offset)
provenances.set_columns(timestamp=timestamp,
timestamp_offset=timestamp_offset,
record=record,
record_offset=record_offset)
provenances.add_row(_get_upgrade_provenance(root))
return msprime.load_tables(nodes=nodes,
edges=edges,
migrations=migrations,
sites=sites,
mutations=mutations,
provenances=provenances)