def test_parse_output_file(self):
fa = FastANI(self.cpus, force_single=True)
out_txt = [['q1', 'r1', '83.1234', '5', '10'],
['q1', 'r2', '99.1111', '3', '10'],
['q2', 'r1', '1.12', '555', '1111']]
expected = {'q1': {'r1': (83.1234, 0.5), 'r2': (99.1111, 0.3)}, 'q2': {'r1': (1.12, 0.5)}}
path_f1 = os.path.join(self.dir_tmp, 'f1.txt')
path_f2 = os.path.join(self.dir_tmp, 'f2.txt')
path_f3 = os.path.join(self.dir_tmp, 'f3.txt')
with open(path_f1, 'w') as fh:
for x in out_txt:
fh.write(' '.join(x) + '\n')
result_f1 = fa.parse_output_file(path_f1)
self.assertEqual(result_f1, expected)
with open(path_f2, 'w') as fh:
for x in out_txt:
fh.write('\t'.join(x) + '\n')
result_f2 = fa.parse_output_file(path_f2)
self.assertEqual(result_f2, expected)
open(path_f3, 'w').close()
result_f3 = fa.parse_output_file(path_f3)
self.assertEqual(result_f3, {})
def run(self, genomes, no_mash, max_d, out_dir, prefix, mash_k, mash_v,
mash_s, min_af):
"""Runs the pipeline."""
self.check_dependencies(no_mash)
ref_genomes = self._get_ref_genomes()
d_compare = defaultdict(set)
d_paths = {**genomes, **ref_genomes}
# Pre-filter using Mash if specified.
if not no_mash:
dir_mash = os.path.join(out_dir, DIR_ANI_REP_INT_MASH)
mash = Mash(self.cpus, dir_mash, prefix)
self.logger.info(f'Using Mash version {mash.version()}')
mash_results = mash.run(genomes, ref_genomes, max_d, mash_k,
mash_v, mash_s)
for qry_gid, ref_hits in mash_results.items():
d_compare[qry_gid] = d_compare[qry_gid].union(
set(ref_hits.keys()))
# Compare against all reference genomes.
else:
for qry_gid in genomes:
d_compare[qry_gid] = set(ref_genomes.keys())
self.logger.info('Calculating ANI with FastANI.')
fastani = FastANI(self.cpus, force_single=True)
fastani_results = fastani.run(d_compare, d_paths)
ANISummaryFile(out_dir, prefix, fastani_results)
ANIClosestFile(out_dir, prefix, fastani_results, genomes, min_af)
def test_run(self):
"""Test that FastANI produces the expected output (version dependent)"""
fa = FastANI(self.cpus, force_single=True)
"""
a = d__Archaea; p__Altiarchaeota; c__Altiarchaeia; o__Altiarchaeales; f__Altiarchaeaceae; g__Altiarchaeum; s__Altiarchaeum sp001873845
b = d__Archaea; p__Altiarchaeota; c__Altiarchaeia; o__Altiarchaeales; f__Altiarchaeaceae; g__Altiarchaeum; s__Altiarchaeum sp002083985
c = d__Bacteria; p__Aquificota; c__Desulfurobacteriia; o__Desulfurobacteriales; f__Desulfurobacteriaceae; g__Thermovibrio; s__Thermovibrio ammonificans
x = d__Archaea; p__Altiarchaeota; c__Altiarchaeia; o__GCA-002841105; f__GCA-002841105; g__GCA-002841105; s__GCA-002841105 sp002841105
y = d__Bacteria; p__Aerophobota; c__Aerophobia; o__Aerophobales; f__Aerophobaceae; g__Aerophobus; s__Aerophobus profundus
z = d__Archaea; p__Halobacterota; c__Archaeoglobi; o__JdFR-21; f__JdFR-21; g__JdFR-21; s__JdFR-21 sp002011165
"""
d_compare = {'a': {'x', 'y'},
'b': {'x'},
'c': {'z'}}
d_paths = {'a': os.path.join(self.genome_root,'GCA/001/873/845', 'GCA_001873845.1_genomic.fna.gz'),
'b': os.path.join(self.genome_root,'GCA/002/083/985', 'GCA_002083985.1_genomic.fna.gz'),
'c': os.path.join(self.genome_root,'GCF/000/185/805', 'GCF_000185805.1_genomic.fna.gz'),
'x': os.path.join(self.genome_root,'GCA/002/841/105', 'GCA_002841105.1_genomic.fna.gz'),
'y': os.path.join(self.genome_root,'GCA/000/402/295', 'GCA_000402295.1_genomic.fna.gz'),
'z': os.path.join(self.genome_root,'GCA/002/011/165', 'GCA_002011165.1_genomic.fna.gz')}
result = fa.run(d_compare, d_paths)
expected = {'a': {'x': {'ani': 82.5201, 'af': 0.57},
'y': {'ani': 74.5154, 'af': 0.0}},
'b': {'x': {'ani': 84.5846, 'af': 0.55}},
'c': {'z': {'ani': 74.4978, 'af': 0.01}}}
self.assertEqual(json.dumps(result, sort_keys=True), json.dumps(expected, sort_keys=True))
def run(self, genomes, no_mash, max_d, out_dir, prefix, mash_k, mash_v,
mash_s, min_af, mash_db):
"""Runs the pipeline.
Parameters
----------
genomes : dict[str, str]
Dict[genome_id] = fasta_path
no_mash : bool
True if Mash should be used for pre-filtering, False otherwise.
max_d : float
maximum distance to keep [0-1]
out_dir : str
The directory to write the output files to.
prefix : str
The prefix to use when writing output files.
mash_k : int
k-mer size [1-32]
mash_v : float
maximum p-value to keep [0-1]
mash_s : int
maximum number of non-redundant hashes
min_af : float
alignment fraction to consider closest genome
mash_db : Optional[str]
The path to read/write the pre-computed Mash reference sketch database.
"""
self.check_dependencies(no_mash)
self.logger.info('Loading reference genomes.')
ref_genomes = self._get_ref_genomes()
d_compare = defaultdict(set)
d_paths = {**genomes, **ref_genomes}
# Pre-filter using Mash if specified.
if not no_mash:
dir_mash = os.path.join(out_dir, DIR_ANI_REP_INT_MASH)
mash = Mash(self.cpus, dir_mash, prefix)
self.logger.info(f'Using Mash version {mash.version()}')
mash_results = mash.run(genomes, ref_genomes, max_d, mash_k,
mash_v, mash_s, mash_db)
for qry_gid, ref_hits in mash_results.items():
d_compare[qry_gid] = d_compare[qry_gid].union(
set(ref_hits.keys()))
# Compare against all reference genomes.
else:
for qry_gid in genomes:
d_compare[qry_gid] = set(ref_genomes.keys())
self.logger.info(
f'Calculating ANI with FastANI v{FastANI._get_version()}.')
fastani = FastANI(self.cpus, force_single=True)
fastani_results = fastani.run(d_compare, d_paths)
taxonomy = Taxonomy().read(TAXONOMY_FILE, canonical_ids=True)
ANISummaryFile(out_dir, prefix, fastani_results, taxonomy)
ANIClosestFile(out_dir, prefix, fastani_results, genomes, min_af,
taxonomy)