def test_prepare_nextflow_input_files(self):
"""test _prepare_nextflow_input_files"""
# Contents of the files is checked elsewhere.
# We'll just check that the files exist
ref_fasta = "tmp.prepare_nextflow_input_files.in.ref.fa"
outdir = "tmp.prepare_nextflow_input_files.outdir"
data_tsv = "tmp.prepare_nextflow_input_files.in.tsv"
vcf_file = "tmp.prepare_nextflow_input_files.in.vcf"
reads_file = "tmp.prepare_nextflow_input_files.in.reads"
with open(ref_fasta, "w") as f:
pass
with open(data_tsv, "w") as f:
print(vcf_file, reads_file, sep="\t", file=f)
with open(vcf_file, "w"), open(reads_file, "w"):
pass
if os.path.exists(outdir):
shutil.rmtree(outdir)
pipeline = multi_sample_pipeline.MultiSamplePipeline(
ref_fasta, data_tsv, outdir)
pipeline._make_output_dir()
pipeline._prepare_nextflow_input_files()
self.assertTrue(os.path.exists(outdir))
self.assertTrue(os.path.exists(pipeline.nextflow_input_tsv))
shutil.rmtree(outdir)
os.unlink(ref_fasta)
os.unlink(data_tsv)
os.unlink(vcf_file)
os.unlink(reads_file)
def run(options):
pipeline = multi_sample_pipeline.MultiSamplePipeline(
options.ref_fasta,
options.data_tsv,
options.outdir,
max_alleles_per_cluster=options.max_alleles_per_cluster,
min_large_ref_length=options.min_large_ref_length,
gramtools_max_read_length=options.max_read_length,
gramtools_kmer_size=options.gramtools_kmer_size,
gramtools_build_threads=options.gramtools_build_threads,
nextflow_config_file=options.nextflow_config_file,
nextflow_work_dir=options.nextflow_work_dir,
force=options.force,
no_run=options.no_run,
clean=not options.no_clean,
variants_per_split=options.variants_per_split,
alleles_per_split=options.alleles_per_split,
total_splits=options.total_splits,
nf_ram_cluster_small_vars=options.nf_ram_cluster_small_vars,
nf_ram_gramtools_build_small=options.nf_ram_gramtools_build_small,
nf_ram_minos_small_vars=options.nf_ram_minos_small_vars,
nf_ram_merge_small_vars=options.nf_ram_merge_small_vars,
testing=options.testing,
use_unmapped_reads=options.use_unmapped_reads,
)
pipeline.run()
def test_prepare_nextflow_input_files(self):
'''test _prepare_nextflow_input_files'''
# Contents of the files is checked elsewhere.
# We'll just check that the files exist
ref_fasta = 'tmp.prepare_nextflow_input_files.in.ref.fa'
outdir = 'tmp.prepare_nextflow_input_files.outdir'
data_tsv = 'tmp.prepare_nextflow_input_files.in.tsv'
vcf_file = 'tmp.prepare_nextflow_input_files.in.vcf'
reads_file = 'tmp.prepare_nextflow_input_files.in.reads'
with open(ref_fasta, 'w') as f:
pass
with open(data_tsv, 'w') as f:
print(vcf_file, reads_file, sep='\t', file=f)
with open(vcf_file, 'w'), open(reads_file, 'w'):
pass
if os.path.exists(outdir):
shutil.rmtree(outdir)
pipeline = multi_sample_pipeline.MultiSamplePipeline(
ref_fasta, data_tsv, outdir)
pipeline._make_output_dir()
pipeline._prepare_nextflow_input_files()
self.assertTrue(os.path.exists(outdir))
self.assertTrue(os.path.exists(pipeline.nextflow_input_tsv))
shutil.rmtree(outdir)
os.unlink(ref_fasta)
os.unlink(data_tsv)
os.unlink(vcf_file)
os.unlink(reads_file)
def test_run_with_small_var_vcf_chunking_total_splits(self):
'''test run with chunking small variatn VCF file using total_splits option'''
input_tsv = 'tmp.multi_sample_pipeline.run.in.tsv'
ref_fasta = os.path.join(data_dir, 'run.ref.0.fa')
with open(input_tsv, 'w') as f:
for i in '1', '2':
reads = os.path.join(data_dir, 'run.reads.' + i + '.sorted.bam')
vcf = os.path.join(data_dir, 'run.calls.' + i + '.vcf')
print(vcf, reads, sep='\t', file=f)
outdir = 'tmp.multi_sample_pipeline.run.out'
if os.path.exists(outdir):
shutil.rmtree(outdir)
pipeline = multi_sample_pipeline.MultiSamplePipeline(ref_fasta, input_tsv, outdir, total_splits=3, min_large_ref_length=10, testing=True, clean=False)
pipeline.run()
expected_vcf = os.path.join(data_dir, 'run.out.vcf')
expected_header, expected_lines = vcf_file_read.vcf_file_to_list(expected_vcf)
got_vcf = os.path.join(outdir, 'combined_calls.vcf')
self.assertTrue(os.path.exists(got_vcf))
got_header, got_lines = vcf_file_read.vcf_file_to_list(got_vcf)
# the datei, minos version, and bcftools verisons might not match
expected_header = [x for x in expected_header if not (x.startswith('##fileDate') or x.startswith('##source=minos') or x.startswith('##bcftools_mergeVersion'))]
got_header = [x for x in got_header if not (x.startswith('##fileDate') or x.startswith('##source=minos') or x.startswith('##bcftools_mergeVersion'))]
self.assertEqual(expected_header, got_header)
self.assertEqual(expected_lines, got_lines)
shutil.rmtree(outdir)
os.unlink(input_tsv)
def _test_run_with_small_var_vcf_chunking_vars_per_split(self):
"""test run with chunking small variant VCF file using variants_per_split option"""
input_tsv = "tmp.multi_sample_pipeline.run.in.tsv"
ref_fasta = os.path.join(data_dir, "run.ref.0.fa")
with open(input_tsv, "w") as f:
for i in "1", "2":
reads1 = os.path.join(data_dir,
"run.reads." + i + ".sorted.bam")
reads2 = os.path.join(data_dir,
"run.reads." + i + ".sorted.bam")
vcf = os.path.join(data_dir, "run.calls." + i + ".vcf")
print(vcf, reads1, reads2, sep="\t", file=f)
outdir = "tmp.multi_sample_pipeline.run.out"
if os.path.exists(outdir):
shutil.rmtree(outdir)
pipeline = multi_sample_pipeline.MultiSamplePipeline(
ref_fasta,
input_tsv,
outdir,
variants_per_split=3,
min_large_ref_length=10,
testing=True,
clean=False,
)
pipeline.run()
expected_vcf = os.path.join(data_dir, "run.out.vcf")
expected_header, expected_lines = vcf_file_read.vcf_file_to_list(
expected_vcf)
got_vcf = os.path.join(outdir, "combined_calls.vcf")
self.assertTrue(os.path.exists(got_vcf))
got_header, got_lines = vcf_file_read.vcf_file_to_list(got_vcf)
# the datei, minos version, and bcftools verisons might not match
expected_header = [
x for x in expected_header
if not (x.startswith("##fileDate") or x.startswith(
"##source=minos") or x.startswith("##bcftools_mergeVersion"))
]
got_header = [
x for x in got_header
if not (x.startswith("##fileDate") or x.startswith(
"##source=minos") or x.startswith("##bcftools_mergeVersion"))
]
self.assertEqual(expected_header, got_header)
self.assertEqual(expected_lines, got_lines)
shutil.rmtree(outdir)
os.unlink(input_tsv)
