def test_unpaired_unsorted():
ref = os.path.join(data_dir, "ref.fa")
reads = os.path.join(data_dir, "reads.fq")
bam = "tmp.minimap_unpaired_unsorted.bam"
subprocess.check_output(f"rm -f {bam}", shell=True)
minimap.run(bam, ref, reads, sort=False)
assert os.path.exists(bam)
assert not bam_is_sorted_and_indexed(bam)
os.unlink(bam)
def test_paired_sorted():
ref = os.path.join(data_dir, "ref.fa")
reads1 = os.path.join(data_dir, "reads_1.fq")
reads2 = os.path.join(data_dir, "reads_2.fq")
bam = "tmp.minimap_paired_sorted.bam"
subprocess.check_output(f"rm -f {bam}", shell=True)
minimap.run(bam, ref, reads1, fq2=reads2, sort=True)
assert os.path.exists(bam)
assert bam_is_sorted_and_indexed(bam)
os.unlink(bam)
os.unlink(f"{bam}.bai")
def map_reads_to_viridian_consensus(self):
logging.info("Mapping reads to consensus from Viridian")
if self.paired:
fq1 = self.sampler.fq_out1
fq2 = self.sampler.fq_out2
else:
fq1 = self.sampler.fq_out
fq2 = None
minimap.run(
self.bam_reads_v_viridian,
self.viridian_fasta,
fq1,
fq2=fq2,
sample_name=self.sample_name,
sort=True,
)
self.update_json_latest_stage("Map reads to Viridian consensus")
def run(options):
fq1, fq2 = utils.check_tech_and_reads_opts_and_get_reads(options)
logging.info("Gathering amplicons scheme files")
if options.built_in_amp_schemes is None and options.amp_schemes_tsv is None:
logging.info("No primer schemes provided. Using all built in schemes")
options.built_in_amp_schemes = list(
amplicon_schemes.get_built_in_schemes().keys())
(
amplicon_scheme_name_to_tsv,
amplicon_scheme_list,
) = amplicon_schemes.load_list_of_amplicon_sets(
built_in_names_to_use=options.built_in_amp_schemes,
tsv_others_to_use=options.amp_schemes_tsv,
)
temp_sam = f"{options.outprefix}.tmp.sam"
logging.info("Mapping reads to reference")
minimap.run(
temp_sam,
options.ref_fasta,
fq1,
fq2=fq2,
sample_name=options.sample_name,
sort=False,
)
logging.info("Detecting primers for each read")
bam_out = f"{options.outprefix}.bam" if options.make_bam else None
results = detect_primers.gather_stats_from_bam(temp_sam, bam_out,
amplicon_scheme_list)
results[
"amplicon_scheme_set_matches"] = detect_primers.amplicon_set_counts_to_json_friendly(
results["amplicon_scheme_set_matches"])
json_out = f"{options.outprefix}.json"
logging.info(f"Tidying files and writing final JSON {json_out}")
if not options.debug:
os.unlink(temp_sam)
utils.write_json(json_out, results)
def initial_read_map_and_detect_amplicon_scheme(self):
logging.info("Mapping reads to reference")
unsorted_bam = os.path.join(self.processing_dir,
"map_reads.unsorted.bam")
minimap.run(
unsorted_bam,
self.ref_genome,
self.fq1,
fq2=self.fq2,
sample_name=self.sample_name,
sort=False,
)
self.update_json_latest_stage("Initial map reads")
logging.info("Detecting amplicon scheme and gathering read statistics")
primer_stats = detect_primers.gather_stats_from_bam(
unsorted_bam, self.unsorted_read_tagged_bam,
self.amplicon_scheme_list)
self.log_dict["read_and_primer_stats"] = primer_stats
self.log_dict["read_and_primer_stats"][
"amplicon_scheme_set_matches"] = detect_primers.amplicon_set_counts_to_json_friendly(
self.log_dict["read_and_primer_stats"]
["amplicon_scheme_set_matches"])
if self.force_amp_scheme is None:
self.log_dict["amplicon_scheme_name"] = self.log_dict[
"read_and_primer_stats"]["chosen_amplicon_scheme"]
else:
self.log_dict["chosen_amplicon_scheme"] = self.force_amp_scheme
chosen_scheme = primer_stats["chosen_amplicon_scheme"]
self.amplicon_tsv = self.amplicon_scheme_name_to_tsv[chosen_scheme]
self.amplicon_set = primers.AmpliconSet.from_tsv(self.amplicon_tsv,
name=chosen_scheme)
self.update_json_latest_stage(
"Gather read stats and detect primer scheme")
def test_gather_stats_from_bam():
# Make a toy genome, two amplicon schemes, and a few reads. Map to make
# the unsorted BAM, then we can test gather_stats_from_bam().
random.seed(42)
ref_seq = "".join(random.choices(["A", "C", "G", "T"], k=1100))
ref_fasta = "tmp.gather_stats_from_bam.ref.fa"
with open(ref_fasta, "w") as f:
print(">ref", file=f)
print(ref_seq, file=f)
tsv_files = {
"scheme1": os.path.join(data_dir,
"gather_stats_from_bam.amplicons_1.tsv"),
"scheme2": os.path.join(data_dir,
"gather_stats_from_bam.amplicons_2.tsv"),
}
# scheme1:
# 100-300, 290-800, 790-1000
#
# scheme2:
# 100-300, 290-500, 490-700, 790-1001
unpaired_read_coords = [(100, 290), (110, 290), (300, 800), (750, 900)]
unpaired_reads_fa = "tmp.gather_stats_from_bam.reads.fa"
_write_sim_reads(ref_seq, unpaired_read_coords, unpaired_reads_fa)
unpaired_bam = "tmp.gather_stats_from_bam.unpaired.bam"
minimap.run(unpaired_bam, ref_fasta, unpaired_reads_fa, sort=False)
reads1_coords = [(100, 200), (110, 210), (310, 410)]
reads2_coords = [(200, 300), (200, 300), (900, 1000)]
reads1_fa = "tmp.gather_stats_from_bam.reads_1.fa"
reads2_fa = "tmp.gather_stats_from_bam.reads_2.fa"
_write_sim_reads(ref_seq, reads1_coords, reads1_fa, suffix="/1")
_write_sim_reads(ref_seq,
reads2_coords,
reads2_fa,
revcomp=True,
suffix="/2")
paired_bam = "tmp.gather_stats_from_bam.paired.bam"
minimap.run(paired_bam, ref_fasta, reads1_fa, fq2=reads2_fa, sort=False)
amplicon_sets = [
primers.AmpliconSet.from_tsv(v, name=k) for k, v in tsv_files.items()
]
tmp_bam_out = "tmp.bam"
subprocess.check_output(f"rm -f {tmp_bam_out}", shell=True)
got = detect_primers.gather_stats_from_bam(unpaired_bam, tmp_bam_out,
amplicon_sets)
assert got == {
"total_reads": 4,
"reads1": 0,
"reads2": 0,
"unpaired_reads": 4,
"mapped": 4,
"match_any_amplicon": 3,
"read_lengths": {
190: 1,
180: 1,
500: 1,
150: 1
},
"template_lengths": {
190: 1,
180: 1,
500: 1,
150: 1
}, # TODO: check this
"amplicon_scheme_set_matches": {
("scheme1", ): 1,
("scheme1", "scheme2"): 2
},
"amplicon_scheme_simple_counts": {
"scheme1": 3,
"scheme2": 2
},
"chosen_amplicon_scheme": "scheme1",
}
assert os.path.exists(tmp_bam_out)
os.unlink(tmp_bam_out)
got = detect_primers.gather_stats_from_bam(paired_bam, tmp_bam_out,
amplicon_sets)
assert got == {
"total_reads": 6,
"reads1": 3,
"reads2": 3,
"unpaired_reads": 0,
"mapped": 6,
"match_any_amplicon": 2,
"read_lengths": {
100: 6
},
"template_lengths": {
200: 1,
190: 1,
690: 1
}, # TODO: check this
"amplicon_scheme_set_matches": {
("scheme1", "scheme2"): 2
},
"amplicon_scheme_simple_counts": {
"scheme1": 2,
"scheme2": 2
},
"chosen_amplicon_scheme": "scheme2",
}
assert os.path.exists(tmp_bam_out)
os.unlink(tmp_bam_out)
os.unlink(ref_fasta)
os.unlink(reads1_fa)
os.unlink(reads2_fa)
os.unlink(unpaired_reads_fa)
os.unlink(unpaired_bam)
os.unlink(paired_bam)