def process( in_biom, out_biom, out_metadata ):
ordered_blast_keys = ["taxonomy", "subject", "evalue", "perc_identity", "perc_query_coverage", "aln_length"] # Keys in blast_affiliations metadata
taxonomy_depth = 0
unclassified_observations = list()
FH_metadata = open( out_metadata, "w" )
FH_metadata.write( "#OTUID\t" + "\t".join([item for item in ordered_blast_keys]) + "\n" )
biom = BiomIO.from_json( in_biom )
for observation in biom.get_observations():
for metadata_key in observation["metadata"].keys():
if metadata_key == "blast_affiliations": # Extract blast_affiliations metadata in metadata_file
if observation["metadata"][metadata_key] is not None:
for current_affi in observation["metadata"][metadata_key]:
if isinstance(current_affi["taxonomy"], list) or isinstance(current_affi["taxonomy"], tuple):
current_affi["taxonomy"] = ";".join( current_affi["taxonomy"] )
FH_metadata.write( observation["id"] + "\t" + "\t".join([str(current_affi[item]) for item in ordered_blast_keys]) + "\n" )
del observation["metadata"][metadata_key]
elif observation["metadata"][metadata_key] is not None: # All list are transformed in string
if isinstance(observation["metadata"][metadata_key], list) or isinstance(observation["metadata"][metadata_key], tuple):
observation["metadata"][metadata_key] = ";".join( map(str, observation["metadata"][metadata_key]) )
if observation["metadata"].has_key( "blast_taxonomy" ):
if observation["metadata"]["blast_taxonomy"] is None:
unclassified_observations.append( observation["id"] )
observation["metadata"]["taxonomy"] = list()
else:
taxonomy_depth = len(observation["metadata"]["blast_taxonomy"].split(";"))
observation["metadata"]["taxonomy"] = observation["metadata"]["blast_taxonomy"].split(";")
# Add "Unclassified" ranks in unclassified observations
if taxonomy_depth > 0:
for observation_id in unclassified_observations:
observation_metadata = biom.get_observation_metadata(observation_id)
observation_metadata["taxonomy"] = ["Unclassified"] * taxonomy_depth
BiomIO.write( out_biom, biom )
def remove_observations( removed_observations, input_biom, output_biom ):
"""
@summary: Removes the specified list of observations.
@param removed_observations: [list] The names of the observations to remove.
@param input_biom: [str] The path to the input BIOM.
@param output_biom: [str] The path to the output BIOM.
"""
biom = BiomIO.from_json( input_biom )
biom.remove_observations( removed_observations )
BiomIO.write( output_biom, biom )
def process(in_biom, out_biom, out_metadata):
ordered_blast_keys = [
"taxonomy", "subject", "evalue", "perc_identity",
"perc_query_coverage", "aln_length"
] # Keys in blast_affiliations metadata
taxonomy_depth = 0
unclassified_observations = list()
FH_metadata = open(out_metadata, "w")
FH_metadata.write("#OTUID\t" +
"\t".join([item for item in ordered_blast_keys]) + "\n")
biom = BiomIO.from_json(in_biom)
for observation in biom.get_observations():
for metadata_key in observation["metadata"].keys():
if metadata_key == "blast_affiliations": # Extract blast_affiliations metadata in metadata_file
if observation["metadata"][metadata_key] is not None:
for current_affi in observation["metadata"][metadata_key]:
if isinstance(current_affi["taxonomy"],
list) or isinstance(
current_affi["taxonomy"], tuple):
current_affi["taxonomy"] = ";".join(
current_affi["taxonomy"])
FH_metadata.write(observation["id"] + "\t" +
"\t".join([
str(current_affi[item])
for item in ordered_blast_keys
]) + "\n")
del observation["metadata"][metadata_key]
elif observation["metadata"][
metadata_key] is not None: # All list are transformed in string
if isinstance(observation["metadata"][metadata_key],
list) or isinstance(
observation["metadata"][metadata_key],
tuple):
observation["metadata"][metadata_key] = ";".join(
map(str, observation["metadata"][metadata_key]))
if observation["metadata"].has_key("blast_taxonomy"):
if observation["metadata"]["blast_taxonomy"] is None:
unclassified_observations.append(observation["id"])
observation["metadata"]["taxonomy"] = list()
else:
taxonomy_depth = len(
observation["metadata"]["blast_taxonomy"].split(";"))
observation["metadata"]["taxonomy"] = observation["metadata"][
"blast_taxonomy"].split(";")
# Add "Unclassified" ranks in unclassified observations
if taxonomy_depth > 0:
for observation_id in unclassified_observations:
observation_metadata = biom.get_observation_metadata(
observation_id)
observation_metadata["taxonomy"] = ["Unclassified"
] * taxonomy_depth
BiomIO.write(out_biom, biom)
def process( args ):
tmp_files = TmpFiles( os.path.split(args.output_file)[0] )
try:
# Add temp taxonomy if multiple and without consensus
tmp_biom = args.input_biom
used_taxonomy_tag = args.taxonomy_tag
if args.multiple_tag is not None:
used_taxonomy_tag = args.tax_consensus_tag
if args.tax_consensus_tag is None:
used_taxonomy_tag = "Used_taxonomy_FROGS-affi"
tmp_biom = tmp_files.add( "tax.biom" )
biom = BiomIO.from_json( args.input_biom )
for observation in biom.get_observations():
metadata = observation["metadata"]
if len(metadata[args.multiple_tag]) > 0:
metadata[used_taxonomy_tag] = metadata[args.multiple_tag][0][args.taxonomy_tag]
BiomIO.write( tmp_biom, biom )
del biom
# Rarefaction
tax_depth = [args.taxonomic_ranks.index(rank) for rank in args.rarefaction_ranks]
rarefaction_cmd = Rarefaction(tmp_biom, tmp_files, used_taxonomy_tag, tax_depth)
rarefaction_cmd.submit( args.log_file )
rarefaction_files = rarefaction_cmd.output_files
# Taxonomy tree
tree_count_file = tmp_files.add( "taxCount.enewick" )
tree_ids_file = tmp_files.add( "taxCount_ids.tsv" )
TaxonomyTree(tmp_biom, used_taxonomy_tag, tree_count_file, tree_ids_file).submit( args.log_file )
# Writes summary
write_summary( args.output_file, args.input_biom, tree_count_file, tree_ids_file, rarefaction_files, args )
finally:
if not args.debug:
tmp_files.deleteAll()
