def _get_provenance(command, attrs):
"""
Returns the V2 tree provenance attributes reformatted as a V3
provenance string.
"""
environment = json.loads(attrs["environment"])
parameters = json.loads(attrs["parameters"])
provenance = msprime.get_provenance_dict(command, parameters)
provenance["version"] = environment["msprime_version"]
provenance["environment"] = environment
return json.dumps(provenance)
def _get_provenance(command, attrs):
"""
Returns the V2 tree provenance attributes reformatted as a V3
provenance string.
"""
environment = json.loads(attrs["environment"])
parameters = json.loads(attrs["parameters"])
provenance = msprime.get_provenance_dict(command, parameters)
provenance["version"] = environment["msprime_version"]
provenance["environment"] = environment
return json.dumps(provenance)
def _load_legacy_hdf5(root):
if 'format_version' not in root.attrs:
raise ValueError("HDF5 file not in msprime format")
format_version = root.attrs['format_version']
if format_version[0] != 2:
raise ValueError("Only version 2.x are supported")
# Get the coalescence records
trees_group = root["trees"]
left = np.array(trees_group["left"])
right = np.array(trees_group["right"])
node = np.array(trees_group["node"])
children = np.array(trees_group["children"])
population = np.array(trees_group["population"])
time = np.array(trees_group["time"])
num_records = len(left)
records = num_records * [None]
for j in range(num_records):
records[j] = msprime.CoalescenceRecord(left=left[j],
right=right[j],
node=node[j],
children=tuple(children[j]),
time=time[j],
population=population[j])
# Get the samples (if present)
samples = None
if "samples" in root:
samples_group = root["samples"]
population = np.array(samples_group["population"])
time = None
if "time" in samples_group:
time = np.array(samples_group["time"])
sample_size = len(population)
samples = sample_size * [None]
for j in range(sample_size):
t = 0
if time is not None:
t = time[j]
samples[j] = msprime.Sample(population=population[j], time=t)
# Get the mutations (if present)
mutations = None
if "mutations" in root:
mutations_group = root["mutations"]
position = np.array(mutations_group["position"])
node = np.array(mutations_group["node"])
num_mutations = len(node)
mutations = num_mutations * [None]
for j in range(num_mutations):
mutations[j] = msprime.Mutation(position=position[j],
node=node[j],
index=j)
ll_ts = _msprime.TreeSequence()
if samples is None:
ll_ts.load_records(records)
else:
ll_ts.load_records(records, samples)
ll_ts.add_provenance_string(
_get_provenance("generate_trees", trees_group.attrs))
if mutations is not None:
ll_ts.set_mutations(mutations)
ll_ts.add_provenance_string(
_get_provenance("generate_mutations", mutations_group.attrs))
ll_ts.add_provenance_string(
json.dumps(msprime.get_provenance_dict("upgrade", {})))
return ll_ts
def _load_legacy_hdf5(root):
if 'format_version' not in root.attrs:
raise ValueError("HDF5 file not in msprime format")
format_version = root.attrs['format_version']
if format_version[0] != 2:
raise ValueError("Only version 2.x are supported")
# Get the coalescence records
trees_group = root["trees"]
left = np.array(trees_group["left"])
right = np.array(trees_group["right"])
node = np.array(trees_group["node"])
children = np.array(trees_group["children"])
population = np.array(trees_group["population"])
time = np.array(trees_group["time"])
num_records = len(left)
records = num_records * [None]
for j in range(num_records):
records[j] = msprime.CoalescenceRecord(
left=left[j], right=right[j], node=node[j],
children=tuple(children[j]), time=time[j],
population=population[j])
# Get the samples (if present)
samples = None
if "samples" in root:
samples_group = root["samples"]
population = np.array(samples_group["population"])
time = None
if "time" in samples_group:
time = np.array(samples_group["time"])
sample_size = len(population)
samples = sample_size * [None]
for j in range(sample_size):
t = 0
if time is not None:
t = time[j]
samples[j] = msprime.Sample(population=population[j], time=t)
# Get the mutations (if present)
mutations = None
if "mutations" in root:
mutations_group = root["mutations"]
position = np.array(mutations_group["position"])
node = np.array(mutations_group["node"])
num_mutations = len(node)
mutations = num_mutations * [None]
for j in range(num_mutations):
mutations[j] = msprime.Mutation(
position=position[j], node=node[j], index=j)
ll_ts = _msprime.TreeSequence()
if samples is None:
ll_ts.load_records(records)
else:
ll_ts.load_records(records, samples)
ll_ts.add_provenance_string(
_get_provenance("generate_trees", trees_group.attrs))
if mutations is not None:
ll_ts.set_mutations(mutations)
ll_ts.add_provenance_string(
_get_provenance("generate_mutations", mutations_group.attrs))
ll_ts.add_provenance_string(
json.dumps(msprime.get_provenance_dict("upgrade", {})))
return ll_ts
