def mp_parse_nfasta_header(header: str):
# E.g. "10001|vfid|47953|vsiid|68790|ssid|SubName: Full=Leader peptidase PilD; SubName: Full=Type 4 prepilin peptidase VcpD; SubName: Full=Type IV-A prepilin peptidase PilD;"
_REGEXES = {
"vfid": ("^([^|]+)[ ]*\|vfid[\|]*", "^([^|]+[ ]*\|vfid[\|]*)"),
"vsiid": ("^([^|]+)[ ]*\|vsiid[\|]*", "^([^|]+[ ]*\|vsiid[\|]*)"),
"ssid": ("^([^|]+)[ ]*\|ssid[\|]*", "^([^|]+[ ]*\|ssid[\|]*)"),
"feature_names": ("^[ ]*([^|]+)$", "^([ ]*[^|]+)$")
}
out = regex_based_tokenization(_REGEXES, header)
for key, value in out.items():
value = value.strip()
if value.isnumeric():
out[key] = int(value)
out["former_id"] = out.pop("source_string")
out.update({
k: safe_findall(v, out["feature_names"])
for k, v in {
"gene_host": "\[([^\]]+)\] *$",
"recname_full": "[^_]*RecName:[_ ]Full=([^;]+);",
"subname_full": "[^_]*SubName:[_ ]Full=([^;]+);",
}.items()
})
return out
def mp_parse_nfasta_header(header: str):
_VFDB_REGEXES = {
"VFID": ("^([^\(\)]+)", "^([^\(\)]+)"),
"gene_host": ("\[([^\]]+)\] *$", "(\[[^\]]+\] *$)"),
"gene_name": ("\[([^\]]+)\] *$", "(\[[^\]]+\] *$)"),
"gene_description": ("([^\(\)]+)$", "([^\(\)]+)$"),
"gene_symbol": ("\(([^\(\)]+)\) *$", "^\([^\(\)]+\) *$"),
"gene_accession_id": ("^\(([^\(\)]+)\)", "^\([^\(\)]+\) *"),
}
out = regex_based_tokenization(_VFDB_REGEXES, header)
out["former_id"] = out.pop("source_string")
out["vfdb_number"] = int(safe_findall("[0-9]+", out["VFID"]))
return out
def regex_based_tokenization(regex_dict: dict,
string: str,
include_source: bool = True,
verbose: bool = False):
# Column name, regex to extract, regex to excise
out = dict()
if include_source:
out["source_string"] = string
_string = str(string)
for key, regexes in regex_dict.items():
_string = _string.strip()
if len(regexes) == 1:
extract_regex = excise_regex = regexes[0]
else:
extract_regex = regexes[0]
excise_regex = regexes[1]
token = safe_findall(extract_regex, _string, verbose=verbose)
out[key] = safe_findall(extract_regex, _string,
verbose=verbose).strip()
_string = re.sub(excise_regex, "", _string)
return out
def mp_parse_nfasta_header(header: str):
out = regex_based_tokenization(
{
"tadb_id": ("^TADB\|([^ ]+) *", "(^TADB\|[^ ]+ *)"),
"gene_symbol": (" *\[([^\[\]]+)\] *$", "( *\[[^\[\]]+\] *$)"),
"gene_geninfo_id":
("^gi\|([0-9]+)[\| ]*", "(^gi\|[0-9]+[\|]*)"),
"gene_refseq_id":
("^[\| ]*ref\|([^\|]+)[\| ]*", "(^[\| ]*ref\|[^\|]+[\| ]*)"),
"dna_strand": ("^[\| ]*:([c]*)", "(^[\| ]*:[c]*)"),
"start_locus": ("^([0-9]+)[ -]*", "(^[0-9]+[ -]*)"),
"end_locus": ("^[ -]*([0-9]+)[ -]*", "(^[ -]*[0-9]+[ -]*)"),
"gene_description": ("(.*)", "(.*)"),
}, header)
out["former_id"] = out.pop("source_string")
out["is_antisense_dna_strand"] = out["dna_strand"] == "c"
out["tadb_number"] = safe_findall("^[A-Z]*([0-9]+)",
str(out["tadb_id"]).upper(),
verbose=False)
return out
def parse_blast_record(blast_record: Record):
# Based on: https://biopython.org/DIST/docs/tutorial/Tutorial.html#htoc95
high_scoring_pairs = OrderedDict()
for alignment in blast_record.alignments:
for hsp in alignment.hsps:
if hsp.expect < E_VALUE_THRESH:
t = alignment.title
d = dict(title=t,
length=alignment.length,
expect=hsp.expect,
score=hsp.score,
bits=hsp.bits,
identities=hsp.identities,
positives=hsp.positives,
query=hsp.query,
match=hsp.match,
sbjct=hsp.sbjct,
geninfo_id=safe_findall("gi\|([^|]+)\|", t).strip())
high_scoring_pairs[t] = {
k: clear_non_printing_chars(v)
for k, v in d.items()
}
return high_scoring_pairs