def test_write_fasta_1(self):
"""Verify write_fasta() creates readable fasta formatted file"""
fileio.write_fasta(self.fasta_dict_1, self.fasta_file)
record = SeqIO.read(self.fasta_file, "fasta")
id = list(self.fasta_dict_1.keys())[0]
seq = self.fasta_dict_1[id]
self.assertEqual(record.id, id)
self.assertEqual(str(record.seq), seq)
def create_genes_fasta(alchemist, values, fasta_dir, db_name,
mol_type="prot", verbose=False, data_cache=None):
if data_cache is None:
data_cache = {}
fasta_path = fasta_dir.joinpath(".".join([db_name, "fasta"]))
gs_to_ts = map_translations(alchemist, values)
pdm_fileio.write_fasta(gs_to_ts, fasta_path)
return fasta_path
def test_reintroduce_fasta_duplicates_2(self):
"""Verify reintroduce_duplicates() recognizes duplicate sequences"""
fileio.write_fasta(self.test_fa_1_gs_to_ts, self.fasta_file)
fileio.reintroduce_fasta_duplicates(self.test_fa_2_ts_to_gs,
self.fasta_file)
with self.fasta_file.open(mode="r") as filehandle:
for record in SeqIO.parse(filehandle, "fasta"):
with self.subTest(seq_id=record.id):
translation = self.test_fa_2_gs_to_ts[record.id]
self.assertTrue(translation is not None)
self.assertEqual(str(record.seq), translation)
def test_write_fasta_2(self):
"""Verify write_fasta() can properly concatenate fasta files"""
fileio.write_fasta(self.fasta_dict_2, self.fasta_file)
records = SeqIO.parse(self.fasta_file, "fasta")
keys = list(self.fasta_dict_2.keys())
for record in records:
self.assertTrue(record.id in keys)
seq = self.fasta_dict_2[record.id]
self.assertEqual(str(record.seq), seq)
def write_phams_process(fasta_dir, aln_dir, pham, pham_translations):
fasta_path = fasta_dir.joinpath("".join([str(pham), "_genes.fasta"]))
aln_path = aln_dir.joinpath("".join([str(pham), "_genes.aln"]))
gs_to_ts = {}
for translation, gene_ids in pham_translations.items():
gs_to_ts[gene_ids[0]] = translation
fileio.write_fasta(gs_to_ts, fasta_path)
if len(pham_translations) > 1:
run_clustalo(fasta_path, aln_path)
fileio.reintroduce_fasta_duplicates(pham_translations, aln_path)
fileio.reintroduce_fasta_duplicates(pham_translations, fasta_path)
def create_pham_hmms_process(working_dir, pham, ts_to_gs, name, M, seq_id,
add_cons, seq_lim):
aln_path = working_dir.joinpath(".".join([str(pham), "aln"]))
hmm_path = working_dir.joinpath(".".join([str(pham), "hhm"]))
gs_to_ts = dict()
for ts, ids in ts_to_gs.items():
gs_to_ts[ts] = ids[0]
pdm_fileio.write_fasta(gs_to_ts, aln_path)
clustalo(aln_path, aln_path, None, None, "fasta", "fasta")
hhmake(aln_path, hmm_path, name, add_cons, seq_lim, M, seq_id)
return (pham, hmm_path)
def estimate_linker_sequence(source_cluster, target_centroid_seq, temp_dir):
guidetree = source_cluster.GT
curr_node = guidetree.clade
while curr_node.is_bifurcating() and len(curr_node.clades) > 1:
left_child = curr_node.clades[0]
right_child = curr_node.clades[1]
left_leaf = get_furthest_sequence(source_cluster, left_child)
right_leaf = get_furthest_sequence(source_cluster, right_child)
left_pid = left_leaf.comment.get("TargetIdentity")
if left_pid is None:
left_pid = alignment.calculate_levenshtein(
left_leaf.comment["Sequence"],
target_centroid_seq,
identity=True)
left_leaf.comment["TargetIdentity"] = left_pid
pass
right_pid = right_leaf.comment.get("TargetIdentity")
if right_pid is None:
right_pid = alignment.calculate_levenshtein(
right_leaf.comment["Sequence"],
target_centroid_seq,
identity=True)
right_leaf.comment["TargetIdentity"] = right_pid
if left_pid > right_pid:
curr_node = left_child
else:
curr_node = right_child
linker = get_furthest_sequence(source_cluster, curr_node)
if linker.comment.get("FilePath") is None:
linker_path = temp_dir.joinpath(".".join([linker.name, "fasta"]))
if not linker_path.is_file():
pdm_fileio.write_fasta({linker.name: linker.comment["Sequence"]},
linker_path)
linker.comment["FilePath"] = linker_path
return linker
def create_pham_alns_process(working_dir, pham, ts_to_gs, mat_out, tree_out,
infile_type, outfile_type):
aln_path = working_dir.joinpath(".".join([str(pham), "aln"]))
mat_path = None
if mat_out:
mat_path = working_dir.joinpath(".".join([str(pham), "mat"]))
tree_path = None
if tree_out:
tree_path = working_dir.joinpath(".".join([str(pham), "tree"]))
gs_to_ts = dict()
for ts, ids in ts_to_gs.items():
gs_to_ts[ids[0]] = ts
pdm_fileio.write_fasta(gs_to_ts, aln_path)
clustalo(aln_path, aln_path, mat_path, tree_path, infile_type,
outfile_type)
pdm_fileio.reintroduce_fasta_duplicates(ts_to_gs, aln_path)
return (pham, aln_path)
def estimate_min_distance(source_cluster,
target_cluster,
temp_dir,
data_dir,
align=False):
if source_cluster.GT is not None:
source_leaf = estimate_linker_sequence(source_cluster,
target_cluster.centroid_seq_str,
temp_dir)
source_linker = source_leaf.name
source_linker_seq = source_leaf.comment["Sequence"]
source_linker_path = source_leaf.comment["FilePath"]
else:
source_linker = source_cluster.CentroidID
source_linker_seq = source_cluster.centroid_seq_str
source_linker_path = temp_dir.joinpath(".".join(
[source_linker, "fasta"]))
if target_cluster.GT is not None:
target_leaf = estimate_linker_sequence(target_cluster,
source_cluster.centroid_seq_str,
temp_dir)
target_linker = target_leaf.name
target_linker_seq = target_leaf.comment["Sequence"]
target_linker_path = target_leaf.comment["FilePath"]
else:
target_linker = target_cluster.CentroidID
target_linker_seq = target_cluster.centroid_seq_str
target_linker_path = temp_dir.joinpath(".".join(
[target_linker, "fasta"]))
if align:
source_len = len(source_linker_seq)
target_len = len(target_linker_seq)
if not source_linker_path.is_file():
pdm_fileio.write_fasta({source_linker: source_linker_seq},
source_linker_path)
if not target_linker_path.is_file():
pdm_fileio.write_fasta({target_linker: target_linker_seq},
target_linker_path)
pairwise_path = data_dir.joinpath("".join(
[source_linker, "__", target_linker, ".fasta"]))
error = True
while error:
try:
linker_alignment = alignment.pairwise_align(source_linker_path,
target_linker_path,
pairwise_path,
tool="needle")
error = False
except ApplicationError:
time.sleep(0.2)
if source_len > target_len:
linker_pid = linker_alignment.annotations["identity"] / source_len
else:
linker_pid = linker_alignment.annotations["identity"] / target_len
linker_pid = linker_pid * 100
else:
linker_pid = alignment.calculate_levenshtein(source_linker_seq,
target_linker_seq)
return linker_pid
def write_centroids_threadtask(temp_dir, source_path, target_path,
source_gs_to_ts, target_gs_to_ts):
temp_dir.mkdir()
pdm_fileio.write_fasta(source_gs_to_ts, source_path)
pdm_fileio.write_fasta(target_gs_to_ts, target_path)