Martel.Re("bp|aa") +
blank_space +
Martel.Opt(residue_type +
blank_space) +
data_file_division +
blank_space +
date +
Martel.AnyEol())
# definition line
# DEFINITION Genomic sequence for Arabidopsis thaliana BAC T25K16 from
# chromosome I, complete sequence.
definition_block = Std.description_block(
Martel.Str("DEFINITION ") +
Std.description(Martel.UntilEol()) +
Martel.AnyEol() +
Martel.Rep(Martel.Str(" ") +
Std.description(Martel.UntilEol()) +
Martel.AnyEol())
)
# accession line
# ACCESSION AC007323
accession = Martel.Group("accession",
Martel.Re("[\w]+"))
accession_block = Martel.Group("accession_block",
Martel.Str("ACCESSION") +
Martel.Rep1(blank_space +
Martel.Rep1(accession +
Martel.Opt(Martel.Str(" ")) +
Martel.Group("date_updated", date) + \
Martel.Str(" (Rel. ") + \
Martel.Digits("release_updated") + \
Martel.Str(", Last updated, Version ") + \
Martel.Digits("version_number") + \
Martel.Str(")") + \
Martel.AnyEol()
DT_block = DT_created_line + DT_updated_line
## DE - description (>=1 per entry)
DE_line = Martel.Str("DE ") + \
Std.description(Martel.UntilEol("description")) + \
Martel.AnyEol()
DE_block = Std.description_block(Martel.Group("description_block",
Martel.Rep1(DE_line)))
## KW - keyword (>=1 per entry)
KW_line = Martel.Str("KW ") + \
Martel.ToEol("keyword_data")
KW_block = Martel.Rep1(KW_line)
## OS - organism species (>=1 per entry)
OS_block = sprot38.OS_block
## OC - organism classification (>=1 per entry)
OC_block = sprot38.OC_block
## OG - organelle (0 or 1 per entry)
OG_block = sprot38.OG_block
