def pe_functional_profiling_pipeline(params):
"""Function calling functional profiling for fastq files"""
work_dir = os.path.join(params['work_dir'], str(uuid.uuid4()))
os.mkdir(work_dir)
config_file = write_config_file(work_dir)
project_file = write_project_file(params['input_reads'],
params['reference'], work_dir,
params['is_paired_end'])
project = Project(config_file=config_file, project_file=project_file)
if params['is_paired_end'] == "1":
project = fastq_pe_pipeline(project)
elif params['is_paired_end'] == "0":
project = fastq_se_pipeline(project)
else:
raise ValueError('Wrong values of is_paired_end parameter',
params['is_paired_end'])
out_dir = os.path.join(work_dir, 'out')
os.mkdir(out_dir)
# export_reads
output = {}
output['krona_charts'] = {}
out_fwd_fastq = os.path.join(work_dir, 'out_fwd.fastq')
sample_id = project.list_samples()[0]
if params['is_paired_end'] == "1":
out_rev_fastq = os.path.join(work_dir, 'out_rev.fastq')
else:
out_rev_fastq = ''
# write filtered fastq
write_filtered_fastq(out_fwd_fastq, out_rev_fastq, project)
output['fwd_reads'] = out_fwd_fastq
if params['is_paired_end'] == "1":
output['rev_reads'] = out_rev_fastq
# Generate output
out_report = os.path.join(out_dir, 'fama_report.html')
generate_html_report(out_report, project, params['name2ref'])
with zipfile.ZipFile(out_report + '.zip',
'w',
zipfile.ZIP_DEFLATED,
allowZip64=True) as zip_file:
zip_file.write(out_report, 'fama_report.html')
output['html_report'] = out_report + '.zip'
# TODO: Krona charts generate_functions_chart(parser_fwd)
report_files = {}
if params['is_paired_end'] == "1":
metric = 'efpkg'
rawcount_flag = False
for sample_id in project.list_samples():
if project.samples[sample_id].rpkg_scaling_factor == 0.0:
rawcount_flag = True
if rawcount_flag:
metric = 'fragmentcount'
else:
metric = 'erpkg'
rawcount_flag = False
for sample_id in project.list_samples():
if project.samples[sample_id].rpkg_scaling_factor == 0.0:
rawcount_flag = True
if rawcount_flag:
metric = 'readcount'
# Create TraitMatrix object
output['trait_matrix_ref'] = write_trait_matrix(project, params)
output['functional_profile_ref'] = write_functional_profile(
project, params, output['trait_matrix_ref'])
report_files[out_report] = 'fama_report.html'
project_xlsx_report = sanitize_file_name(
os.path.join(
project.options.work_dir,
project.options.project_name + '_' + metric + '_functions.xlsx'))
if os.path.exists(project_xlsx_report):
report_files[project_xlsx_report] = 'Functional_profiles_combined.xlsx'
else:
print('Project XLSX file not found:', project_xlsx_report)
project_xlsx_report = sanitize_file_name(
os.path.join(
project.options.work_dir, project.options.project_name + '_' +
metric + '_functions_taxonomy.xlsx'))
if os.path.exists(project_xlsx_report):
report_files[
project_xlsx_report] = 'Function_taxonomy_profiles_combined.xlsx'
else:
print('Project XLSX file not found:', project_xlsx_report)
for sample_id in project.list_samples():
sample_xlsx_report = sanitize_file_name(
os.path.join(project.options.work_dir, sample_id + '_' + metric +
'_functions_taxonomy.xlsx'))
if os.path.exists(sample_xlsx_report):
report_files[sample_xlsx_report] = sanitize_file_name(
sample_id + ' function taxonomy profile long.xlsx')
else:
print('Sample XLSX file not found:', sample_xlsx_report)
krona_file = sanitize_file_name(
os.path.join(
project.options.work_dir, sample_id + '_' + metric +
'_functional_taxonomy_profile.xml.html'))
if os.path.exists(krona_file):
krona_output = \
sanitize_file_name(os.path.join(out_dir, sample_id +
'_function_taxonomy_profile_chart.html'))
shutil.copy2(krona_file, krona_output)
with zipfile.ZipFile(krona_output + '.zip',
'w',
zipfile.ZIP_DEFLATED,
allowZip64=True) as zip_file:
zip_file.write(
krona_output,
sanitize_file_name(
sample_id + '_function_taxonomy_profile_chart.html'))
report_files[krona_output] = \
sanitize_file_name(sample_id + '_function_taxonomy_profile_chart.html')
output['krona_charts'][krona_output + '.zip'] = \
(sanitize_file_name(sample_id + '_function_taxonomy_profile_chart.html'),
sample_id + ' function taxonomy chart')
else:
print('Krona diagram file not found:', krona_file)
output_files = list()
result_file = os.path.join(project.options.work_dir, 'Fama_result.zip')
with zipfile.ZipFile(result_file,
'w',
zipfile.ZIP_DEFLATED,
allowZip64=True) as zip_file:
for filename in report_files:
zip_file.write(filename, report_files[filename])
output_files.append({
'path': result_file,
'name': os.path.basename(result_file),
'label': os.path.basename(result_file),
'description': 'Files generated by Fama App'
})
output['report_files'] = output_files
return output
def protein_functional_profiling_pipeline(params):
"""Function calling functional profiling for protein fasta file
params = {'input_proteins': input_proteins,
'work_dir': self.shared_folder,
'reference': fama_reference,
'ws_name': params['workspace_name'],
'ws_client': ws_client,
'featureset_name': params['output_feature_set_name'],
'annotation_prefix': params['output_annotation_name'],
'name2ref' : name2ref
}
"""
work_dir = os.path.join(params['work_dir'], str(uuid.uuid4()))
os.mkdir(work_dir)
config_file = write_config_file(work_dir)
project_file = write_project_file(params['input_proteins'],
params['reference'], work_dir)
project = Project(config_file=config_file, project_file=project_file)
# Run Fama
project = protein_pipeline(project)
out_dir = os.path.join(work_dir, 'out')
os.mkdir(out_dir)
# export_reads
output = {}
output['krona_charts'] = {}
# Generate output
report_files = {}
metric = 'proteincount'
project_xlsx_report = sanitize_file_name(
os.path.join(
project.options.work_dir,
project.options.project_name + '_' + metric + '_functions.xlsx'))
if os.path.exists(project_xlsx_report):
report_files[project_xlsx_report] = 'Functional_profiles_combined.xlsx'
else:
print('Project XLSX file not found:', project_xlsx_report)
project_xlsx_report = sanitize_file_name(
os.path.join(
project.options.work_dir, project.options.project_name + '_' +
metric + '_functions_taxonomy.xlsx'))
if os.path.exists(project_xlsx_report):
report_files[
project_xlsx_report] = 'Function_taxonomy_profiles_combined.xlsx'
else:
print('Project XLSX file not found:', project_xlsx_report)
project_text_report = sanitize_file_name(
os.path.join(project.options.work_dir, 'all_proteins.list.txt'))
if os.path.exists(project_text_report):
report_files[project_text_report] = 'proteins_list.txt'
else:
print('Proteins list not found:', project_text_report)
featureset_elements = {}
featureset_element_ordering = []
objects_created = []
genome_names = {}
# Get Domain Model Set reference
dms_ref = get_dms(params['reference'],
project.config.get_functions_file(project.collection),
params['ws_name'], params['ws_client'])
for sample_id in project.list_samples():
annotation_obj_ref, feature_ids, genome_name = \
save_domain_annotations(project, dms_ref, params['ws_name'],
params['ws_client'], params['annotation_prefix'],
sample_id, params['name2ref'][sample_id])
genome_names[sample_id] = genome_name
objects_created.append({
'ref':
annotation_obj_ref,
'description':
'Functional annotations for genome ' +
project.samples[sample_id].sample_name
})
for feature_id in feature_ids:
if feature_id not in featureset_elements:
featureset_elements[feature_id] = []
featureset_elements[feature_id].append(
params['name2ref'][sample_id])
featureset_element_ordering.append(feature_id)
sample_xlsx_report = sanitize_file_name(
os.path.join(project.options.work_dir, sample_id + '_' + metric +
'_functions_taxonomy.xlsx'))
if os.path.exists(sample_xlsx_report):
report_files[sample_xlsx_report] = \
sanitize_file_name(genome_name + '_function_taxonomy_profile_long.xlsx')
else:
print('Sample XLSX file not found:', sample_xlsx_report)
krona_file = sanitize_file_name(
os.path.join(
project.options.work_dir, sample_id + '_' + metric +
'_functional_taxonomy_profile.xml.html'))
if os.path.exists(krona_file):
krona_output = \
sanitize_file_name(os.path.join(out_dir, genome_name +
'_function_taxonomy_profile_chart.html'))
shutil.copy2(krona_file, krona_output)
with zipfile.ZipFile(krona_output + '.zip',
'w',
zipfile.ZIP_DEFLATED,
allowZip64=True) as zip_file:
zip_file.write(
krona_output,
sanitize_file_name(
genome_name + '_function_taxonomy_profile_chart.html'))
report_files[krona_output] = \
sanitize_file_name(genome_name + '_function_taxonomy_profile_chart.html')
output['krona_charts'][krona_output + '.zip'] = \
(sanitize_file_name(genome_name + '_function_taxonomy_profile_chart.html'),
sample_id + ' function taxonomy chart')
else:
print('Krona diagram file not found:', krona_file)
feature_set_data = {
'description': 'FeatureSet generated by Fama protein profiling',
'element_ordering': featureset_element_ordering,
'elements': featureset_elements
}
out_report = os.path.join(out_dir, 'fama_report.html')
generate_protein_html_report(out_report, project, params['name2ref'])
with zipfile.ZipFile(out_report + '.zip',
'w',
zipfile.ZIP_DEFLATED,
allowZip64=True) as zip_file:
zip_file.write(out_report, 'fama_report.html')
output['html_report'] = out_report + '.zip'
report_files[out_report] = 'fama_report.html'
output_files = list()
result_file = os.path.join(project.options.work_dir, 'Fama_result.zip')
with zipfile.ZipFile(result_file,
'w',
zipfile.ZIP_DEFLATED,
allowZip64=True) as zip_file:
for filename in report_files:
zip_file.write(filename, report_files[filename])
output_files.append({
'path': result_file,
'name': os.path.basename(result_file),
'label': os.path.basename(result_file),
'description': 'Files generated by Fama App'
})
output['report_files'] = output_files
output['project'] = project
output['feature_set_data'] = feature_set_data
output['objects_created'] = objects_created
return output
def make_sample_tax_func_xlsx(project,
scores,
metric,
function_id=None,
rank=None):
"""Generates XLSX file for taxa scores for one or all functions in all samples.
Args:
project (:obj:'Project'): Project object that stores all annotated reads
scores (dict[str, dict[str, dict[str, float]]]): outer key is function
identifier, middle-level key is sample identifier,
inner key is metric, value id float
metric (str, optional): acceptable values are 'readcount', 'erpk', 'rpkm',
'fragmentcount', 'fpk', 'efpk', 'fpkm', 'erpkm', 'efpkm',
'fpkg', 'rpkg', 'erpkg', 'efpkg', 'proteincount'
function_id (str, optional): function identifier. If function_id is None, all
functions will be included into workbook.
rank (str, optional): taxonomic rank. if rank parameter is not None, the
resulting XLSX file will contain only entries for this rank.
"""
if function_id is None:
if rank is None:
xlsxfile = sanitize_file_name(
os.path.join(
project.options.work_dir, project.options.project_name +
'_' + metric + '_samples_taxonomy.xlsx'))
else:
xlsxfile = sanitize_file_name(
os.path.join(
project.options.work_dir, project.options.project_name +
'_' + metric + '_samples_' + rank + '_taxonomy.xlsx'))
else:
if rank is None:
xlsxfile = sanitize_file_name(
os.path.join(
project.options.work_dir,
function_id + '_' + metric + '_samples_taxonomy.xlsx'))
else:
xlsxfile = sanitize_file_name(
os.path.join(
project.options.work_dir, function_id + '_' + metric +
'_samples_' + rank + '_taxonomy.xlsx'))
print('Writing', xlsxfile)
writer = pd.ExcelWriter(xlsxfile, engine='xlsxwriter')
for function in sorted(project.ref_data.functions_dict.keys()):
if function_id is not None and function != function_id:
continue
# Subsetting scores
sample_scores = autovivify(3, float)
for taxonomy_id in scores.keys():
if function in scores[taxonomy_id].keys():
for sample in project.list_samples():
if sample in scores[taxonomy_id][function]:
for key, val in scores[taxonomy_id][function][
sample].items():
sample_scores[taxonomy_id][sample][key] = val
else:
sample_scores[taxonomy_id][sample][metric] = 0.0
tax_profile = TaxonomyProfile()
tax_profile.make_function_taxonomy_profile(project.taxonomy_data,
sample_scores)
taxonomy_df = tax_profile.convert_profile_into_score_df(metric=metric)
if rank is None:
taxonomy_df.to_excel(writer,
sheet_name=function,
merge_cells=False)
else:
filtered_df = taxonomy_df[taxonomy_df[('', 'Rank')] == rank]
filtered_df.to_excel(writer,
sheet_name=function,
merge_cells=False)
format_taxonomy_worksheet(writer, function)
# Make 'Average' sheet
if function_id is None:
sample_scores = autovivify(3, float)
for taxonomy_id in scores:
for function in sorted(project.ref_data.functions_dict.keys()):
if function in scores[taxonomy_id]:
for sample in project.list_samples():
if sample in scores[taxonomy_id][function]:
for key, val in scores[taxonomy_id][function][
sample].items():
sample_scores[taxonomy_id][sample][key] += val
else:
sample_scores[taxonomy_id][sample][metric] += 0.0
for taxonomy_id in sample_scores:
for sample in sample_scores[taxonomy_id]:
sample_scores[taxonomy_id][sample][metric] = \
sample_scores[taxonomy_id][sample][metric] \
/ len(project.ref_data.functions_dict.keys())
tax_profile = TaxonomyProfile()
tax_profile.make_function_taxonomy_profile(project.taxonomy_data,
sample_scores)
taxonomy_df = tax_profile.convert_profile_into_score_df(metric=metric)
if rank is None:
taxonomy_df.to_excel(writer,
sheet_name='Average',
merge_cells=False)
else:
filtered_df = taxonomy_df[taxonomy_df[('', 'Rank')] == rank]
filtered_df.to_excel(writer,
sheet_name='Average',
merge_cells=False)
format_taxonomy_worksheet(writer, 'Average')
writer.save()
def make_assembly_xlsx(assembler):
"""Generates XLSX file for assembly.
Args:
assembler (:obj:'GeneAssembler'): gene assembler object
"""
xlsxfile = sanitize_file_name(
os.path.join(assembler.project.options.assembly_dir, 'out',
assembler.project.options.project_name +
'_assembly.xlsx'))
xlsxfile = xlsxfile.replace(' ', '_')
xlsxfile = xlsxfile.replace("'", "")
xlsxfile = xlsxfile.replace('"', '')
workbook = xlsxwriter.Workbook(xlsxfile)
bold = workbook.add_format({'bold': True})
cell_numformat0 = workbook.add_format()
cell_numformat0.set_num_format('0')
cell_numformat1 = workbook.add_format()
cell_numformat1.set_num_format('0.0')
cell_numformat5 = workbook.add_format()
cell_numformat5.set_num_format('0.00000')
functions_list = set()
samples_list = sorted(assembler.project.list_samples())
function_read_counts = autovivify(
2, float) # function_read_counts[function][sample]
gene_rpkm = autovivify(3, float) # gene_rpkm[function][gene][sample],
# parameters are RPKM, coverage, identity
# count reads per function, per sample
for function in assembler.assembly.reads:
functions_list.add(function)
for read in assembler.assembly.reads[function]:
function_read_counts[function][assembler.assembly.reads[function]
[read]] += 1
# collect RPKM scores for contigs per function, per sample (for contigs? for genes?)
# calculate total read count
total_read_count = 0
for sample in samples_list:
total_read_count += assembler.project.options.get_fastq1_readcount(
sample)
total_read_count += assembler.project.options.get_fastq2_readcount(
sample)
# generate output
# make worksheet for read counts per function
reads_worksheet = workbook.add_worksheet('Functions read count')
row = 0
col = 0
reads_worksheet.write(row, col, 'Function', bold)
for sample in samples_list:
col += 1
reads_worksheet.write(row, col, sample, bold)
col += 1
reads_worksheet.write(row, col, 'All samples', bold)
col += 1
reads_worksheet.write(row, col, 'Assembled reads', bold)
col += 1
reads_worksheet.write(row, col, 'Unassembled reads', bold)
col += 1
reads_worksheet.write(row, col, 'Definition', bold)
for function in sorted(functions_list):
row += 1
col = 0
reads_worksheet.write(row, col, function, bold)
for sample in samples_list:
col += 1
if sample in function_read_counts[function]:
reads_worksheet.write(
row, col, function_read_counts[function][sample] * 2,
cell_numformat0)
else:
reads_worksheet.write(row, col, 0, cell_numformat0)
col += 1
all_reads = sum(function_read_counts[function].values()) * 2
reads_worksheet.write(row, col, all_reads, cell_numformat0)
col += 1
assembled_reads = 0
if function in assembler.assembly.contigs:
assembled_reads = sum([
len(c.reads)
for c in assembler.assembly.contigs[function].values()
])
reads_worksheet.write(row, col, assembled_reads, cell_numformat0)
col += 1
reads_worksheet.write(row, col, all_reads - assembled_reads,
cell_numformat0)
col += 1
reads_worksheet.write(
row, col,
assembler.project.ref_data.lookup_function_name(function))
# adjust column width
reads_worksheet.set_column(0, 0, 10)
reads_worksheet.set_column(col, col, 50)
# make worksheet with contig data
contigs_worksheet = workbook.add_worksheet('Contigs')
row = 0
col = 0
contigs_worksheet.write(row, col, 'Contig', bold)
col += 1
contigs_worksheet.write(row, col, 'Function', bold)
col += 1
contigs_worksheet.write(row, col, 'Length', bold)
col += 1
contigs_worksheet.write(row, col, 'Read count', bold)
col += 1
contigs_worksheet.write(row, col, 'RPKM', bold)
col += 1
contigs_worksheet.write(row, col, 'Coverage', bold)
col += 1
contigs_worksheet.write(row, col, 'Number of genes', bold)
for sample in samples_list:
col += 1
contigs_worksheet.write(row, col, sample, bold)
col += 1
contigs_worksheet.write(row, col, sample, bold)
col += 1
contigs_worksheet.write(row, col, sample, bold)
col += 1
contigs_worksheet.write(row, col, 'Definition', bold)
row += 1
col = 6
for sample in samples_list:
col += 1
contigs_worksheet.write(row, col, 'Read count', bold)
col += 1
contigs_worksheet.write(row, col, 'RPKM', bold)
col += 1
contigs_worksheet.write(row, col, 'Coverage', bold)
for function in sorted(functions_list):
if function in assembler.assembly.contigs:
for contig in sorted(assembler.assembly.contigs[function].keys()):
row += 1
col = 0
contigs_worksheet.write(row, col, contig, bold)
col += 1
contigs_worksheet.write(row, col, function)
col += 1
contigs_worksheet.write(
row, col,
len(assembler.assembly.contigs[function][contig].sequence))
col += 1
contigs_worksheet.write(
row, col, assembler.assembly.contigs[function]
[contig].get_read_count())
col += 1
contigs_worksheet.write(
row, col, assembler.assembly.contigs[function]
[contig].get_rpkm(total_read_count), cell_numformat5)
col += 1
contigs_worksheet.write(
row, col, assembler.assembly.contigs[function]
[contig].get_coverage(), cell_numformat1)
col += 1
contigs_worksheet.write(
row, col,
len(assembler.assembly.contigs[function][contig].genes))
col += 1
for sample in samples_list:
contigs_worksheet.write(
row, col, assembler.assembly.contigs[function]
[contig].get_read_count(sample))
col += 1
contigs_worksheet.write(
row, col,
assembler.assembly.contigs[function][contig].get_rpkm(
assembler.project.options.get_fastq1_readcount(
sample), sample), cell_numformat5)
col += 1
contigs_worksheet.write(
row, col, assembler.assembly.contigs[function]
[contig].get_coverage(sample), cell_numformat1)
col += 1
contigs_worksheet.write(
row, col,
assembler.project.ref_data.lookup_function_name(function))
# adjust column width
contigs_worksheet.set_column(0, 1, 10)
contigs_worksheet.set_column(col, col, 50)
# make worksheet for genes
genes_worksheet = workbook.add_worksheet('Genes')
row = 0
col = 0
genes_worksheet.write(row, col, 'Gene', bold)
col += 1
genes_worksheet.write(row, col, 'Reads function', bold)
col += 1
genes_worksheet.write(row, col, 'Contig', bold)
col += 1
genes_worksheet.write(row, col, 'Gene start', bold)
col += 1
genes_worksheet.write(row, col, 'Gene end', bold)
col += 1
genes_worksheet.write(row, col, 'Gene length', bold)
col += 1
genes_worksheet.write(row, col, 'Gene strand', bold)
col += 1
genes_worksheet.write(row, col, 'Read count', bold)
col += 1
genes_worksheet.write(row, col, 'RPKM', bold)
col += 1
genes_worksheet.write(row, col, 'Coverage', bold)
col += 1
genes_worksheet.write(row, col, 'Fama gene status', bold)
col += 1
genes_worksheet.write(row, col, 'Fama function', bold)
col += 1
genes_worksheet.write(row, col, 'Fama identity', bold)
col += 1
genes_worksheet.write(row, col, 'CDS completeness', bold)
col += 1
genes_worksheet.write(row, col, 'Fama best hit', bold)
col += 1
genes_worksheet.write(row, col, 'Fama best hit taxonomy ID', bold)
col += 1
genes_worksheet.write(row, col, 'Fama best hit organism', bold)
col += 1
genes_worksheet.write(row, col, 'Fama best hit taxonomy', bold)
col += 1
genes_worksheet.write(row, col, 'Fama LCA taxonomy ID', bold)
col += 1
genes_worksheet.write(row, col, 'Fama LCA organism', bold)
col += 1
genes_worksheet.write(row, col, 'Fama LCA taxonomy', bold)
for sample in samples_list:
col += 1
genes_worksheet.write(row, col, sample, bold)
col += 1
genes_worksheet.write(row, col, sample, bold)
col += 1
genes_worksheet.write(row, col, sample, bold)
col += 1
genes_worksheet.write(row, col, 'Definition', bold)
row += 1
col = 20
for sample in samples_list:
col += 1
genes_worksheet.write(row, col, 'Read count', bold)
col += 1
genes_worksheet.write(row, col, 'RPKM', bold)
col += 1
genes_worksheet.write(row, col, 'Coverage', bold)
for function in sorted(functions_list):
if function not in assembler.assembly.contigs:
continue
for contig in sorted(assembler.assembly.contigs[function].keys()):
for gene_id in sorted(
assembler.assembly.contigs[function][contig].genes.keys()):
gene = assembler.assembly.contigs[function][contig].genes[
gene_id]
row += 1
col = 0
# Write Gene ID
genes_worksheet.write(row, col, gene_id)
col += 1
# Write Gene function from read mapping
genes_worksheet.write(row, col, function)
col += 1
# Write Contig ID
genes_worksheet.write(row, col, contig)
col += 1
# Write gene start
genes_worksheet.write(row, col, int(gene.start))
col += 1
# Write gene end
genes_worksheet.write(row, col, int(gene.end))
col += 1
# Write gene length
gene_length = int(gene.end) - int(gene.start) + 1
genes_worksheet.write(row, col, gene_length)
col += 1
# Write gene strand
genes_worksheet.write(row, col, gene.strand)
col += 1
# Write read count (calculated from read count of contig,
# adjusted by gene length)
gene_read_count = assembler.assembly.contigs[function][contig].get_read_count()\
* gene_length \
/ len(assembler.assembly.contigs[function][contig].sequence)
genes_worksheet.write(row, col, gene_read_count,
cell_numformat1)
col += 1
# Write RPKM
gene_rpkm = assembler.assembly.contigs[function][
contig].get_rpkm(total_read_count)
genes_worksheet.write(row, col, gene_rpkm, cell_numformat5)
col += 1
# Write coverage
genes_worksheet.write(
row, col, assembler.assembly.contigs[function]
[contig].get_coverage(), cell_numformat1)
col += 1
# Write FAMA gene status
genes_worksheet.write(row, col, gene.status)
col += 1
if gene.status == STATUS_GOOD:
# Write FAMA predicted functions
gene_functions = set(
[y for x in gene.hit_list.hits for y in x.functions])
genes_worksheet.write(row, col, ','.join(gene_functions))
col += 1
# Write FAMA identity
gene_identity = [x.identity for x in gene.hit_list.hits]
genes_worksheet.write(
row, col,
sum(gene_identity) / len(gene_identity),
cell_numformat1)
col += 1
# Write CDS completeness
ref_lengths = [x.s_len for x in gene.hit_list.hits]
genes_worksheet.write(
row, col,
len(gene.protein_sequence) * 100 * len(ref_lengths) /
sum(ref_lengths), cell_numformat1)
col += 1
# Write FAMA best hits
fama_hits = [
cleanup_protein_id(x.subject_id)
for x in gene.hit_list.hits
]
genes_worksheet.write(row, col, ','.join(fama_hits))
col += 1
# Write FAMA taxonomy ID
gene_taxonomy = [
assembler.project.ref_data.lookup_protein_tax(
cleanup_protein_id(x.subject_id))
for x in gene.hit_list.hits
]
genes_worksheet.write(row, col, ','.join(gene_taxonomy))
col += 1
# Write Fama best hit organism
gene_organism = [
assembler.project.taxonomy_data.get_name(x)
for x in gene_taxonomy
]
genes_worksheet.write(row, col, ','.join(gene_organism))
col += 1
# Write Fama best hit taxonomy
best_hit_taxonomy = [
assembler.project.taxonomy_data.get_taxonomy_lineage(x)
for x in gene_taxonomy
]
genes_worksheet.write(row, col,
'|'.join(best_hit_taxonomy))
col += 1
# Write Fama LCA taxonomy ID
lca_taxonomy_id = gene.taxonomy
genes_worksheet.write(row, col, lca_taxonomy_id)
col += 1
# Write Fama LCA organism
lca_organism = assembler.project.taxonomy_data.get_name(
lca_taxonomy_id)
genes_worksheet.write(row, col, lca_organism)
col += 1
# Write Fama LCA taxonomy
lca_taxonomy = assembler.project.taxonomy_data.get_taxonomy_lineage(
lca_taxonomy_id)
genes_worksheet.write(row, col, lca_taxonomy)
else:
for _ in range(0, 10):
genes_worksheet.write(row, col, 'N/A')
col += 1
for sample in samples_list:
col += 1
gene_read_count = assembler.assembly.contigs[function][
contig].get_read_count(sample) * len(
gene.protein_sequence) * 3 / len(
assembler.assembly.contigs[function]
[contig].sequence)
genes_worksheet.write(row, col, gene_read_count,
cell_numformat1)
col += 1
gene_rpkm = assembler.assembly.contigs[function][
contig].get_rpkm(
assembler.project.options.get_fastq1_readcount(
sample), sample)
genes_worksheet.write(row, col, gene_rpkm, cell_numformat5)
col += 1
genes_worksheet.write(
row, col, assembler.assembly.contigs[function]
[contig].get_coverage(sample), cell_numformat1)
col += 1
genes_worksheet.write(
row, col,
assembler.project.ref_data.lookup_function_name(function))
# adjust column width
genes_worksheet.set_column(0, 0, 20)
genes_worksheet.set_column(1, 1, 10)
genes_worksheet.set_column(7, 9, 15)
genes_worksheet.set_column(col, col, 50)
workbook.close()
def make_function_sample_xlsx(project, scores, metric, sample_id=None):
"""Generates XLSX file for function scores for one or more samples.
Args:
project (:obj:'Project'): Project object that stores all annotated reads
scores (dict[str, dict[str, dict[str, float]]]): outer key is function
identifier, middle-level key is sample identifier,
inner key is metric, value id float
metric (str, optional): acceptable values are 'readcount', 'erpk', 'rpkm',
'fragmentcount', 'fpk', 'efpk', 'fpkm', 'erpkm', 'efpkm',
'fpkg', 'rpkg', 'erpkg', 'efpkg', 'proteincount'
sample_id (str, optional): sample identifier
"""
if sample_id is None:
xlsxfile = sanitize_file_name(
os.path.join(
project.options.work_dir, project.options.project_name + '_' +
metric + '_functions.xlsx'))
else:
xlsxfile = sanitize_file_name(
os.path.join(project.options.work_dir,
sample_id + '_' + metric + '_functions.xlsx'))
print('Writing', xlsxfile)
workbook = xlsxwriter.Workbook(xlsxfile)
bold = workbook.add_format({'bold': True})
functions_list = sorted(project.ref_data.functions_dict.keys())
categories_list = sorted(
list(
set([
project.ref_data.functions_dict[x]['group']
for x in project.ref_data.functions_dict.keys()
])))
scores_cat = autovivify(2, float)
# generate tables for functions
scores_worksheet = workbook.add_worksheet('Functions ' + metric)
row = 0
col = 0
scores_worksheet.write(row, col, 'Function', bold)
for sample in project.list_samples():
if sample_id is not None and sample != sample_id:
continue
col += 1
scores_worksheet.write(row, col, sample, bold)
col += 1
scores_worksheet.write(row, col, 'Definition', bold)
for function in functions_list:
category = project.ref_data.lookup_function_group(function)
row += 1
col = 0
scores_worksheet.write(row, col, function, bold)
for sample in project.list_samples():
if sample_id is not None and sample != sample_id:
continue
col += 1
if function in scores and sample in scores[function]:
scores_worksheet.write(row, col,
scores[function][sample][metric])
scores_cat[category][sample] += scores[function][sample][
metric]
else:
scores_worksheet.write(row, col, 0.0)
col += 1
scores_worksheet.write(row, col,
project.ref_data.lookup_function_name(function))
# adjust column width
scores_worksheet.set_column(0, 0, 10)
scores_worksheet.set_column(col, col, 50)
# Write worksheet for categories
scores_cat_worksheet = workbook.add_worksheet('Categories ' + metric)
row = 0
col = 0
scores_cat_worksheet.write(row, col, 'Categories', bold)
for sample in project.list_samples():
if sample_id is not None and sample != sample_id:
continue
col += 1
scores_cat_worksheet.write(row, col, sample, bold)
for category in categories_list:
row += 1
col = 0
scores_cat_worksheet.write(row, col, category, bold)
for sample in project.list_samples():
if sample_id is not None and sample != sample_id:
continue
col += 1
if category in scores_cat and sample in scores_cat[category]:
scores_cat_worksheet.write(row, col,
scores_cat[category][sample])
else:
scores_cat_worksheet.write(row, col, 0.0)
# adjust column width
scores_cat_worksheet.set_column(0, 0, 50)
workbook.close()