def test_complete_assembly_with_snps_and_indels(test_data):
assert os.path.exists(test_data["dirname"])
pre_out = "tmp.complete_assembly_with_snps_and_indels"
subprocess.check_output(f"rm -rf {pre_out}*", shell=True)
fq1 = f"{pre_out}.1.fq"
fq2 = f"{pre_out}.2.fq"
all_amplicon_names = set([x[0] for x in test_data["amplicons"]])
make_catted_paired_reads_for_amplicon_set(test_data, all_amplicon_names, fq1, fq2)
outdir = f"{pre_out}.out"
ref_fasta = f"{pre_out}.ref.fa"
ref_seq = copy.copy(test_data["ref_seq"])
ref_seq[500] = "A" if ref_seq[500] != "A" else "G"
ref_seq.insert(1100, "A")
ref_seq.pop(1200)
nucleotides_list_to_fasta_file(ref_seq, "ref", ref_fasta)
one_sample_pipeline.run_one_sample(
"illumina",
outdir,
ref_fasta,
fq1,
fq2=fq2,
tsv_of_amp_schemes=test_data["schemes_tsv"],
keep_intermediate=True,
)
# TODO: check that we got the expected output
# Should be something like this in the VCF (which doesn't yet exist):
# CHROM POS ID REF ALT QUAL FILTER INFO FORMAT sample
# ref 501 0 A C . PASS . GT 1/1
# ref 1099 1 GA G . PASS . GT 1/1
# ref 1199 2 C CG . PASS . GT 1/1
subprocess.check_output(f"rm -rf {pre_out}*", shell=True)
def _test_not_expected_amplicons(test_data):
assert os.path.exists(test_data["dirname"])
pre_out = "tmp.not_expected_amplicons"
subprocess.check_output(f"rm -rf {pre_out}*", shell=True)
fq1 = f"{pre_out}.1.fq"
fq2 = f"{pre_out}.2.fq"
all_amplicon_names = set([x[0] for x in test_data["amplicons"]])
make_catted_paired_reads_for_amplicon_set(test_data, all_amplicon_names, fq1, fq2)
amplicons = [
("amplicon1", 50, 900),
("amplicon2", 850, 1200),
("amplicon3", 1165, 1700),
]
amplicons_json = f"{pre_out}.amplicons.json"
make_amplicons(amplicons_json, amplicons=amplicons)
outdir = f"{pre_out}.out"
try:
one_sample_pipeline.run_one_sample(
"illumina",
outdir,
test_data["ref_fasta"],
fq1,
fq2=fq2,
amplicon_json=amplicons_json,
keep_intermediate=True,
)
except:
pass
# TODO: check that we got the expected output
subprocess.check_output(f"rm -rf {pre_out}*", shell=True)
def _test_complete_assembly_no_reads_map(test_data):
assert os.path.exists(test_data["dirname"])
pre_out = "tmp.no_reads_map"
subprocess.check_output(f"rm -rf {pre_out}*", shell=True)
fq1 = f"{pre_out}.1.fq"
fq2 = f"{pre_out}.2.fq"
# make some garbage reads so they don't map
with open(fq1, "w") as f1, open(fq2, "w") as f2:
print("@read1/1", "A" * 100, "+", "I" * 100, sep="\n", file=f1)
print("@read1/2", "A" * 100, "+", "I" * 100, sep="\n", file=f2)
outdir = f"{pre_out}.out"
try:
one_sample_pipeline.run_one_sample(
"illumina",
outdir,
test_data["ref_fasta"],
fq1,
fq2=fq2,
amplicon_json=test_data["amplicons_tsv"],
)
# This test should fail on viridian, producing no consensus
# TODO specify that it was the consensus file that's missing
except utils.OutputFileError as error:
if str(error) != str(
os.path.abspath(
os.path.join(outdir, "Processing/viridian/consensus.final_assembly.fa")
)
):
raise error
subprocess.check_output(f"rm -rf {pre_out}*", shell=True)
def run(options):
fq1, fq2 = utils.check_tech_and_reads_opts_and_get_reads(options)
if options.force:
logging.info(f"--force option used, so deleting {options.outdir} if it exists")
subprocess.check_output(f"rm -rf {options.outdir}", shell=True)
one_sample_pipeline.run_one_sample(
options.tech,
options.outdir,
options.ref_fasta,
fq1,
fq2=fq2,
built_in_amp_schemes=options.built_in_amp_schemes,
tsv_of_amp_schemes=options.amp_schemes_tsv,
force_amp_scheme=options.force_amp_scheme,
keep_intermediate=options.debug,
keep_bam=options.keep_bam,
target_sample_depth=options.target_sample_depth,
sample_name=options.sample_name,
min_sample_depth=options.min_sample_depth,
max_percent_amps_fail=options.max_percent_amps_fail,
viridian_cons_max_n_percent=options.max_cons_n_percent,
frs_threshold=options.frs_threshold,
self_qc_depth=options.self_qc_depth,
log_liftover=options.log_liftover,
trim_5prime=options.trim_5prime,
command_line_args=options,
)
def test_complete_assembly_from_all_good_amplicons_unpaired(test_data):
assert os.path.exists(test_data["dirname"])
pre_out = "tmp.complete_assembly_from_all_good_amplicons_unpaired"
subprocess.check_output(f"rm -rf {pre_out}*", shell=True)
fq = f"{pre_out}.fq"
all_amplicon_names = set([x[0] for x in test_data["amplicons"]])
make_catted_unpaired_reads_for_amplicon_set(test_data, all_amplicon_names, fq)
outdir = f"{pre_out}.out"
one_sample_pipeline.run_one_sample(
"ont",
outdir,
test_data["ref_fasta"],
fq,
tsv_of_amp_schemes=test_data["schemes_tsv"],
keep_intermediate=True,
)
# TODO: check that we got the expected output
subprocess.check_output(f"rm -rf {pre_out}*", shell=True)
def test_reads_are_wgs_not_amplicon(test_data):
assert os.path.exists(test_data["dirname"])
pre_out = "tmp.reads_are_wgs_not_amplicon"
subprocess.check_output(f"rm -rf {pre_out}*", shell=True)
fq1 = f"{pre_out}.1.fq"
fq2 = f"{pre_out}.2.fq"
tiling_reads(test_data["ref_seq"], 150, 350, fq1, fq2, step=2)
outdir = f"{pre_out}.out"
one_sample_pipeline.run_one_sample(
"illumina",
outdir,
test_data["ref_fasta"],
fq1,
fq2=fq2,
tsv_of_amp_schemes=test_data["schemes_tsv"],
keep_intermediate=True,
)
# TODO: check that we got the expected output
subprocess.check_output(f"rm -rf {pre_out}*", shell=True)
def test_assembly_amplicon_3_no_reads(test_data):
assert os.path.exists(test_data["dirname"])
pre_out = "tmp.assembly_amplicon_3_no_reads"
subprocess.check_output(f"rm -rf {pre_out}*", shell=True)
fq1 = f"{pre_out}.1.fq"
fq2 = f"{pre_out}.2.fq"
amplicon_names = {"amplicon1", "amplicon2", "amplicon4", "amplicon5"}
make_catted_paired_reads_for_amplicon_set(test_data, amplicon_names, fq1, fq2)
outdir = f"{pre_out}.out"
one_sample_pipeline.run_one_sample(
"illumina",
outdir,
test_data["ref_fasta"],
fq1,
fq2=fq2,
tsv_of_amp_schemes=test_data["schemes_tsv"],
keep_intermediate=True,
)
# TODO: check that we got the expected output
subprocess.check_output(f"rm -rf {pre_out}*", shell=True)