def prepare_nfasta_for_indexing(input_file: str,
output_dir: str,
preserve_headers: bool = False,
chop: bool = False,
chunk_length: int = int(3.6 * 10**9)):
array = FASTAArray.parse(Utilities.load_string(input_file))
if not preserve_headers:
array._fix_headers()
output_dir = Utilities.ends_with_slash(output_dir)
os.makedirs(output_dir, exist_ok=True)
output_file_mask = (output_dir + Utilities.filename_only(input_file))
annotation_file = "{}_annotation.tsv".format(output_file_mask)
array.dump_annotation(annotation_file)
arrays_dict = {"{}.fasta".format(output_file_mask): array}
if chop and array.get_total_length() >= chunk_length:
print("Too large reference nFASTA file: '{}'. Splitting sequences".
format(input_file))
arrays_dict = array._chop_sequences(chunk_length)
arrays_dict = {
"{a}_{i}.fasta".format(a=output_file_mask, i=i): arrays_dict[i]
for i in arrays_dict
}
refdatas_dict = {}
counter = 0
for chunk_file in arrays_dict:
counter += 1
arrays_dict[chunk_file].dump_fastas(chunk_file)
refdatas_dict["sequence_{}".format(counter)] = {
"reference_nfasta": chunk_file,
"annotation": annotation_file
}
print("FASTA files created: {}".format(counter))
return refdatas_dict
def __init__(self, parsed_dictionary: dict):
self._nfasta = parsed_dictionary["reference_nfasta"]
self.db_name = parsed_dictionary.get("alias")
if not self.db_name:
self.db_name = Utilities.filename_only(self._nfasta)
self._reference_mask = Utilities.ends_with_slash(os.path.dirname(os.path.realpath(self._nfasta))) + self.db_name
self.bowtie_index_mask = parsed_dictionary["ebwt_mask"]
self.bowtie2_index_mask = parsed_dictionary["bt2_mask"]
self.samtools_index_file = parsed_dictionary["fai"]
self.bedtools_genome_file = parsed_dictionary["genome"]
self.annotation_file = parsed_dictionary["annotation"]
def compile(input_file: str,
output_dir: str,
preserve_headers: bool = False,
chop: bool = False,
chunk_length: int = int(3.6 * 10**9)):
import json
from modules.FASTAArray import FASTAArray
from modules.RefDataLine import RefDataLine
output_dir = Utilities.ends_with_slash(output_dir)
os.makedirs(output_dir, exist_ok=True)
refdatas_dict = FASTAArray.prepare_nfasta_for_indexing(
input_file=input_file,
output_dir=output_dir,
preserve_headers=preserve_headers,
chop=chop,
chunk_length=chunk_length)
output_dict = {}
for sequence_id in refdatas_dict:
annotation_dict = refdatas_dict[sequence_id]
nfasta_file = annotation_dict.get("reference_nfasta")
if not nfasta_file:
continue
indexing_dict = {"alias": Utilities.filename_only(nfasta_file)}
indexing_dict.update(RefDataLine.fill_dict(nfasta_file))
indexing_dict.update(annotation_dict)
print("Processing nFASTA: '{}'".format(nfasta_file))
refdata = RefDataLine(indexing_dict)
refdata.index()
output_dict[sequence_id] = indexing_dict
output_file = "{a}{b}_refdata.json".format(
a=Utilities.ends_with_slash(output_dir),
b=Utilities.filename_only(input_file))
Utilities.dump_string(
string=json.dumps(output_dict, sort_keys=False, indent=4) + "\n",
file=output_file)
print("Created reference data linker: '{}'".format(output_file))
return output_file
def fill_dict(nfasta_file: str):
mask = Utilities.ends_with_slash(os.path.dirname(os.path.realpath(nfasta_file))) + Utilities.filename_only(nfasta_file)
d = {"ebwt_mask": "{}_colorspace".format(mask),
"bt2_mask": "{}_bowtie2".format(mask),
"fai": "{}_samtools.fai".format(mask),
"genome": "{}_samtools.genome".format(mask),
"annotation": "{}_annotation.tsv".format(mask)}
return d