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 mp_parse_pfasta_header(header: str):
out = regex_based_tokenization(
{
"tadb_id": ("^TADB\|([^ ]+) *", "(^TADB\|[^ ]+ *)"),
"protein_symbol":
(" *\[([^\[\]]+)\] *$", "( *\[[^\[\]]+\] *$)"),
"protein_host": (" *\[([^\[\]]+)\] *$", "( *\[[^\[\]]+\] *$)"),
"protein_geninfo_id": ("^gi\|([0-9]+)\|*", "(^gi\|[0-9]+\|*)"),
"protein_refseq_id":
("^[\| ]*ref\|([^\|]+)[\| ]*", "(^[\| ]*ref\|[^\|]+[\| ]*)"),
"protein_description": ("(.*)", "(.*)"),
}, header)
out["protein_header"] = out.pop("source_string")
return out
def tokenize_reads_file_name(s: str):
d = regex_based_tokenization(
{
"extension": ["\.(.{2,8})$", "(\..{2,8})$"], # E.g. '.fastq.gz'
"last_segment": [
"[^A-Za-z0-9]([A-Za-z0-9]+)$", "([^A-Za-z0-9][A-Za-z0-9]+)$"
], # The last segment is always 001,
"read_index": ["[^A-Za-z0-9](R[0-9]+)$", "([^A-Za-z0-9]R[0-9]+)$"],
"lane_number":
["[^A-Za-z0-9](L[0-9]+)$", "([^A-Za-z0-9]L[0-9]+)$"],
"sample_sheet_number":
["[^A-Za-z0-9](S[0-9]+)$", "([^A-Za-z0-9]S[0-9]+)$"],
"sample_name": ["(.+)", "(.+)"],
},
os.path.basename(s))
d["reads_file"] = s
return d
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