def run_kb_dram_annotate(self, ctx, params):
"""
This example function accepts any number of parameters and returns results in a KBaseReport
:param params: instance of mapping from String to unspecified object
:returns: instance of type "ReportResults" -> structure: parameter
"report_name" of String, parameter "report_ref" of String
"""
# ctx is the context object
# return variables are: output
#BEGIN run_kb_dram_annotate
# validate inputs
if not isinstance(params['assembly_input_ref'], str) or not len(
params['assembly_input_ref']):
raise ValueError('Pass in a valid assembly reference string')
if not isinstance(params['desc'], str) or not len(params['desc']):
raise ValueError('Pass in a valid genomeSet description')
if not isinstance(params['output_name'], str) or not len(
params['output_name']):
raise ValueError('Pass in a valid genomeSet output name')
if not isinstance(params['min_contig_size'],
int) or (params['min_contig_size'] < 0):
raise ValueError('Min contig size must be a non-negative integer')
# setup params
with open("/kb/module/kbase.yml", 'r') as stream:
data_loaded = yaml.load(stream)
version = str(data_loaded['module-version'])
is_metagenome = params['is_metagenome']
min_contig_size = params['min_contig_size']
trans_table = str(params['trans_table'])
bitscore = params['bitscore']
rbh_bitscore = params['rbh_bitscore']
output_dir = os.path.join(self.shared_folder, 'DRAM_annos')
output_objects = []
# create Util objects
wsClient = workspaceService(self.workspaceURL, token=ctx['token'])
assembly_util = AssemblyUtil(self.callback_url)
genome_util = GenomeFileUtil(self.callback_url)
# set DRAM database locations
print('DRAM version: %s' % dram_version)
import_config('/data/DRAM_databases/CONFIG')
# This is a hack to get around a bug in my database setup
set_database_paths(
description_db_loc='/data/DRAM_databases/description_db.sqlite')
print_database_locations()
# get files
assemblies = assembly_util.get_fastas(
{'ref_lst': [params['assembly_input_ref']]})
# would paths ever have more than one thing?
fasta_locs = [
assembly_data['paths'][0]
for assembly_ref, assembly_data in assemblies.items()
]
# get assembly refs from dram assigned genome names
assembly_ref_dict = {
os.path.splitext(
os.path.basename(
remove_suffix(assembly_data['paths'][0], '.gz')))[0]:
assembly_ref
for assembly_ref, assembly_data in assemblies.items()
}
# annotate and distill with DRAM
annotate_bins(fasta_locs,
output_dir,
min_contig_size,
trans_table=trans_table,
bit_score_threshold=bitscore,
rbh_bit_score_threshold=rbh_bitscore,
low_mem_mode=True,
rename_bins=False,
keep_tmp_dir=False,
threads=THREADS,
verbose=False)
output_files = get_annotation_files(output_dir)
distill_output_dir = os.path.join(output_dir, 'distilled')
summarize_genomes(output_files['annotations']['path'],
output_files['trnas']['path'],
output_files['rrnas']['path'],
output_dir=distill_output_dir,
groupby_column='fasta')
output_files = get_distill_files(distill_output_dir, output_files)
if is_metagenome:
pass # TODO: make annotated metagenome object
else:
# generate genome files
annotations = pd.read_csv(output_files['annotations']['path'],
sep='\t',
index_col=0)
genome_objects = generate_genomes(
annotations, output_files['genes_fna']['path'],
output_files['genes_faa']['path'], assembly_ref_dict,
assemblies, params["workspace_name"], ctx.provenance())
genome_ref_dict = dict()
genome_set_elements = dict()
for genome_object in genome_objects:
info = genome_util.save_one_genome(genome_object)["info"]
genome_ref = '%s/%s/%s' % (info[6], info[0], info[4])
genome_set_elements[genome_object["name"]] = {
'ref': genome_ref
}
output_objects.append({
"ref": genome_ref,
"description": 'Annotated Genome'
})
genome_ref_dict[genome_object["name"]] = genome_ref
# add ontology terms
anno_api = annotation_ontology_api(service_ver="beta")
ontology_events = add_ontology_terms(annotations, params['desc'],
version,
params['workspace_name'],
self.workspaceURL,
genome_ref_dict)
[
anno_api.add_annotation_ontology_events(i)
for i in ontology_events
]
# make genome set
# TODO: only make genome set if there is more than one genome
if 'provenance' in ctx:
provenance = ctx['provenance']
else:
provenance = [{}]
# add additional info to provenance here, in this case the input data object reference
provenance[0]['input_ws_objects'] = list(genome_ref_dict.values())
provenance[0]['service'] = 'kb_SetUtilities'
provenance[0]['method'] = 'KButil_Batch_Create_GenomeSet'
output_genomeSet_obj = {
'description': params['desc'],
'elements': genome_set_elements
}
output_genomeSet_name = params['output_name']
new_obj_info = wsClient.save_objects({
'workspace':
params['workspace_name'],
'objects': [{
'type': 'KBaseSearch.GenomeSet',
'data': output_genomeSet_obj,
'name': output_genomeSet_name,
'meta': {},
'provenance': provenance
}]
})[0]
genome_set_ref = '%s/%s/%s' % (new_obj_info[6], new_obj_info[0],
new_obj_info[4])
output_objects.append({
"ref": genome_set_ref,
"description": params['desc']
})
# generate report
product_html_loc = os.path.join(distill_output_dir, 'product.html')
report = generate_product_report(self.callback_url,
params['workspace_name'], output_dir,
product_html_loc, output_files,
output_objects)
output = {
'report_name': report['name'],
'report_ref': report['ref'],
}
#END run_kb_dram_annotate
# At some point might do deeper type checking...
if not isinstance(output, dict):
raise ValueError('Method run_kb_dram_annotate return value ' +
'output is not type dict as required.')
# return the results
return [output]
def test_what_is_fastas(self):
assembly_util = AssemblyUtil(self.callback_url)
fastas = assembly_util.get_fastas({'ref_lst': ['41343/11/3']})
print(fastas)
def run_kb_dramv_annotate(self, ctx, params):
"""
:param params: instance of mapping from String to unspecified object
:returns: instance of type "ReportResults" -> structure: parameter
"report_name" of String, parameter "report_ref" of String
"""
# ctx is the context object
# return variables are: output
#BEGIN run_kb_dramv_annotate
warnings.filterwarnings("ignore")
# setup
affi_contigs_shock_ids = params['affi_contigs_shock_id']
min_contig_size = params['min_contig_size']
trans_table = str(params['trans_table'])
bitscore = params['bitscore']
rbh_bitscore = params['rbh_bitscore']
assembly_util = AssemblyUtil(self.callback_url)
datafile_util = DataFileUtil(self.callback_url)
# get contigs and merge
assemblies = assembly_util.get_fastas(
{'ref_lst': [params['assembly_input_ref']]})
fasta = os.path.join(self.shared_folder, 'merged_contigs.fasta')
with open(fasta, 'w') as f:
for assembly_ref, assembly_data in assemblies.items():
fasta_path = assembly_data['paths'][0]
for line in open(fasta_path):
f.write(line)
# get affi contigs, read all and merge
affi_contigs_path = os.path.join(self.shared_folder,
'VIRSorter_affi-contigs.tab')
with open(affi_contigs_path, 'w') as f:
for affi_contigs_shock_id in affi_contigs_shock_ids:
temp_affi_contigs_path = os.path.join(
self.shared_folder, 'temp_VIRSorter_affi-contigs.tab')
temp_affi_contigs = datafile_util.shock_to_file({
'shock_id':
affi_contigs_shock_id,
'file_path':
temp_affi_contigs_path,
'unpack':
'unpack'
})['file_path']
for line in open(temp_affi_contigs):
f.write(line)
os.remove(temp_affi_contigs)
# set DRAM database locations
print('DRAM version: %s' % dram_version)
import_config('/data/DRAM_databases/CONFIG')
# This is a hack to get around a bug in my database setup
set_database_paths(
description_db_loc='/data/DRAM_databases/description_db.sqlite')
print_database_locations()
# clean affi contigs file
cleaned_fasta = os.path.join(
self.shared_folder, '%s.cleaned.fasta' % os.path.basename(fasta))
remove_bad_chars(input_fasta=fasta, output=cleaned_fasta)
cleaned_affi_contigs = os.path.join(
self.shared_folder, 'VIRSorter_affi-contigs.cleaned.tab')
remove_bad_chars(input_virsorter_affi_contigs=affi_contigs_path,
output=cleaned_affi_contigs)
# annotate and distill
output_dir = os.path.join(self.shared_folder, 'DRAM_annos')
annotate_vgfs(cleaned_fasta,
cleaned_affi_contigs,
output_dir,
min_contig_size,
trans_table=trans_table,
bit_score_threshold=bitscore,
rbh_bit_score_threshold=rbh_bitscore,
low_mem_mode=True,
keep_tmp_dir=False,
threads=THREADS,
verbose=False)
output_files = get_annotation_files(output_dir)
distill_output_dir = os.path.join(output_dir, 'distilled')
summarize_vgfs(output_files['annotations']['path'],
distill_output_dir,
groupby_column='scaffold')
output_files = get_viral_distill_files(distill_output_dir,
output_files)
# generate report
product_html_loc = os.path.join(distill_output_dir, 'product.html')
report = generate_product_report(self.callback_url,
params['workspace_name'], output_dir,
product_html_loc, output_files)
output = {
'report_name': report['name'],
'report_ref': report['ref'],
}
#END run_kb_dramv_annotate
# At some point might do deeper type checking...
if not isinstance(output, dict):
raise ValueError('Method run_kb_dramv_annotate return value ' +
'output is not type dict as required.')
# return the results
return [output]