def simsam_range(table, tree, simulated_sample_sizes, dissimilarities, mapping_f=None):
"""Applies sim_otu_table over a range of parameters
table: the input table to simulate samples from
tree: tree related OTUs in input table
simulated_sample_sizes: a list of ints defining how many
output samples should be create per input sample
dissimilarities: a list of floats containing the
dissimilarities to use in simulating tables
mapping_f: file handle for metadata mapping file, if
a mapping file should be created with the samples from
each simulated table
This function will yield tuples with the following form:
(output table, output mapping lines, simulated_sample_size, dissimilarity)
If the user does not provide mapping_f, the tuples will look like:
(output table, None, simulated_sample_size, dissimilarity)
"""
if mapping_f is not None:
# if the user provided a mapping file, load it into
# a list for repeated use, and define the function for
# processing the mapping file
mapping_lines = list(mapping_f)
process_map = create_replicated_mapping_file
else:
# otherwise create a dummy function for processing the
# mapping file so we don't have to check whether it
# exists on every iteration
mapping_lines = None
def process_map(mapping_lines, simulated_sample_size, sample_ids):
return None
for simulated_sample_size in simulated_sample_sizes:
# create the output mapping file data
output_mapping_lines = process_map(mapping_lines, simulated_sample_size, table.ids())
for dissimilarity in dissimilarities:
# create the simulated otu table
output_sample_ids, output_otu_ids, output_data, output_metadata = sim_otu_table(
table.ids(),
table.ids(axis="observation").tolist(),
table.iter(),
table.metadata(axis="observation"),
tree,
simulated_sample_size,
dissimilarity,
)
output_table = Table(
output_data,
output_otu_ids,
output_sample_ids,
observation_metadata=output_metadata,
generated_by=get_generated_by_for_biom_tables(),
create_date=datetime.now().isoformat(),
)
yield (output_table, output_mapping_lines, simulated_sample_size, dissimilarity)
def make_otu_table(otu_map_f, otu_to_taxonomy=None, delim='_', table_id=None,
otu_ids_to_exclude=None, sample_metadata=None, seq_counts=None):
"""Generate a BIOM table from an OTU map
Parameters
----------
otu_map_f : file-like object
The OTU map. Jagged tab-separated file where the first column contains
the OTU ID and subsequent columns contain sequence IDs belonging to
that OTU
otu_to_taxonomy : dict, optional
Defaults to ``None``. If supplied, the dict maps OTU IDs to taxonomies
delim : str, optional
Defaults to "_". The delimiter that is used in the sequence IDs to join
the sample ID to the sequence number
table_id : object, optional
Defaults to ``None``. The identifier that will be given to the
generated BIOM table
otu_ids_to_exclude : iterable, optional
Defaults to ``None``. If present, these OTUs will not be added to the
OTU table from the OTU map
sample_metadata : dict of dicts, optional
Defaults to ``None``. If supplied, keys in the outer dict should be
sample IDs, and keys in the inner dicts should be column names.
seq_counts : dict, optional
Defaults to ``None``. If supplied, the dict maps seq ids to seq counts
"""
data, sample_ids, otu_ids = parse_otu_map(
otu_map_f, delim=delim, otu_ids_to_exclude=otu_ids_to_exclude,seq_counts=seq_counts)
if otu_to_taxonomy is not None:
otu_metadata = []
for o in otu_ids:
otu_metadata.append({'taxonomy': otu_to_taxonomy.get(o, ["None"])})
else:
otu_metadata = None
# if sample_metadata is supplied, put in index-order with the OTU map's
# sample_ids, and do not include samples that were in the mapping file
# but NOT in the OTU map
if sample_metadata is not None:
try:
sample_metadata = [sample_metadata[sample_id]
for sample_id in sample_ids]
except KeyError:
raise KeyError("Sample IDs found in OTU map without sample "
"metadata")
try:
return Table(data, otu_ids, sample_ids,
observation_metadata=otu_metadata,
sample_metadata=sample_metadata, table_id=table_id,
generated_by=get_generated_by_for_biom_tables(),
create_date=datetime.now().isoformat())
except ValueError as e:
raise ValueError("Couldn't create OTU table. Is your OTU map empty?"
" Original error message: %s" % (str(e)))
def make_otu_table(otu_map_f, otu_to_taxonomy=None, delim='_', table_id=None,
otu_ids_to_exclude=None, sample_metadata=None):
"""Generate a BIOM table from an OTU map
Parameters
----------
otu_map_f : file-like object
The OTU map. Jagged tab-separated file where the first column contains
the OTU ID and subsequent columns contain sequence IDs belonging to
that OTU
otu_to_taxonomy : dict, optional
Defaults to ``None``. If supplied, the dict maps OTU IDs to taxonomies
delim : str, optional
Defaults to "_". The delimiter that is used in the sequence IDs to join
the sample ID to the sequence number
table_id : object, optional
Defaults to ``None``. The identifier that will be given to the
generated BIOM table
otu_ids_to_exclude : iterable, optional
Defaults to ``None``. If present, these OTUs will not be added to the
OTU table from the OTU map
sample_metadata : iterable of dicts, optional
Defaults to ``None``.
"""
data, sample_ids, otu_ids = parse_otu_map(
otu_map_f, delim=delim, otu_ids_to_exclude=otu_ids_to_exclude)
if otu_to_taxonomy is not None:
otu_metadata = []
for o in otu_ids:
otu_metadata.append({'taxonomy': otu_to_taxonomy.get(o, ["None"])})
else:
otu_metadata = None
# if sample_metadata is not None:
# raise NotImplementedError("Passing of sample metadata to "
# "make_otu_table is not currently supported.")
try:
return Table(data, otu_ids, sample_ids,
observation_metadata=otu_metadata,
sample_metadata=sample_metadata, table_id=table_id,
generated_by=get_generated_by_for_biom_tables(),
create_date=datetime.now().isoformat())
except ValueError as e:
raise ValueError("Couldn't create OTU table. Is your OTU map empty?"
" Original error message: %s" % (str(e)))
def make_otu_table(otu_map_f,
otu_to_taxonomy=None,
delim='_',
table_id=None,
otu_ids_to_exclude=None,
sample_metadata=None):
"""Generate a BIOM table from an OTU map
Parameters
----------
otu_map_f : file-like object
The OTU map. Jagged tab-separated file where the first column contains
the OTU ID and subsequent columns contain sequence IDs belonging to
that OTU
otu_to_taxonomy : dict, optional
Defaults to ``None``. If supplied, the dict maps OTU IDs to taxonomies
delim : str, optional
Defaults to "_". The delimiter that is used in the sequence IDs to join
the sample ID to the sequence number
table_id : object, optional
Defaults to ``None``. The identifier that will be given to the
generated BIOM table
otu_ids_to_exclude : iterable, optional
Defaults to ``None``. If present, these OTUs will not be added to the
OTU table from the OTU map
sample_metadata : dict of dicts, optional
Defaults to ``None``. If supplied, keys in the outer dict should be
sample IDs, and keys in the inner dicts should be column names.
"""
data, sample_ids, otu_ids = parse_otu_map(
otu_map_f, delim=delim, otu_ids_to_exclude=otu_ids_to_exclude)
if otu_to_taxonomy is not None:
otu_metadata = []
for o in otu_ids:
otu_metadata.append({'taxonomy': otu_to_taxonomy.get(o, ["None"])})
else:
otu_metadata = None
# if sample_metadata is supplied, put in index-order with the OTU map's
# sample_ids, and do not include samples that were in the mapping file
# but NOT in the OTU map
if sample_metadata is not None:
try:
sample_metadata = [
sample_metadata[sample_id] for sample_id in sample_ids
]
except KeyError:
raise KeyError("Sample IDs found in OTU map without sample "
"metadata")
try:
return Table(data,
otu_ids,
sample_ids,
observation_metadata=otu_metadata,
sample_metadata=sample_metadata,
table_id=table_id,
generated_by=get_generated_by_for_biom_tables(),
create_date=datetime.now().isoformat())
except ValueError as e:
raise ValueError("Couldn't create OTU table. Is your OTU map empty?"
" Original error message: %s" % (str(e)))
def simsam_range(table,
tree,
simulated_sample_sizes,
dissimilarities,
mapping_f=None):
"""Applies sim_otu_table over a range of parameters
table: the input table to simulate samples from
tree: tree related OTUs in input table
simulated_sample_sizes: a list of ints defining how many
output samples should be create per input sample
dissimilarities: a list of floats containing the
dissimilarities to use in simulating tables
mapping_f: file handle for metadata mapping file, if
a mapping file should be created with the samples from
each simulated table
This function will yield tuples with the following form:
(output table, output mapping lines, simulated_sample_size, dissimilarity)
If the user does not provide mapping_f, the tuples will look like:
(output table, None, simulated_sample_size, dissimilarity)
"""
if mapping_f is not None:
# if the user provided a mapping file, load it into
# a list for repeated use, and define the function for
# processing the mapping file
mapping_lines = list(mapping_f)
process_map = create_replicated_mapping_file
else:
# otherwise create a dummy function for processing the
# mapping file so we don't have to check whether it
# exists on every iteration
mapping_lines = None
def process_map(mapping_lines, simulated_sample_size, sample_ids):
return None
for simulated_sample_size in simulated_sample_sizes:
# create the output mapping file data
output_mapping_lines = \
process_map(mapping_lines, simulated_sample_size, table.ids())
for dissimilarity in dissimilarities:
# create the simulated otu table
output_sample_ids, output_otu_ids, output_data, output_metadata = \
sim_otu_table(table.ids(),
table.ids(axis='observation').tolist(),
table.iter(),
table.metadata(axis='observation'),
tree,
simulated_sample_size,
dissimilarity)
output_table = Table(
output_data,
output_otu_ids,
output_sample_ids,
observation_metadata=output_metadata,
generated_by=get_generated_by_for_biom_tables(),
create_date=datetime.now().isoformat())
yield (output_table, output_mapping_lines, simulated_sample_size,
dissimilarity)