Appendix A, File Formats

The currently supported file formats include GDE data files, Genbank formatted files (with type extensions), a generic flat file format, and a color mask file.

• GDE format. GDE format is a tagged field format used for storing all available information about a sequence. The format matches very closely the GDE internal structures for sequence data. The format consists of text records starting and ending with braces ('{}'). Between the open and close braces are several tagged field lines specifying different pieces of information about a given sequence. The tag values can be wrapped with double quote characters ('""') as needed. If quotes are not used, the first whitespace delimited string is taken as the value. The allowable fields are:
  

  {

  name "Short name for sequence"

  longname "Long (more descriptive) name for sequence"

  sequence-ID "Unique ID number"

  creation-date "mm/dd/yy hh:mm:ss"

  direction [-1|1]

  strandedness [1|2]

  type [DNA|RNA||PROTEIN|TEXT|MASK]

  offset (-999999,999999)

  group-ID (0,999)

  creator "Author's name"

  descrip "Verbose description"

  comments "Lines of comments that can be fairly arbitrary text about a sequence. Return characters are allowed, but no internal double quotes or brace characters. Remember to close with a double quote"

  sequence "gctagctagctagctagctcttagctgtagtcgtagctgatgctagct gatgctagctagctagctagctgatcgatgctagctgatcgtagctgacg gactgatgctagctagctagctagctgtctagtgtcgtagtgcttattgc"

  }
  
  Any fields that are not specified are assumed to be the default values. Offsets can be negative as well as positive. Genbank entries written out in this format will have all (") converted to ('), and all ({}) converted to ([]) to avoid confusion in the parser. Leading and trailing gaps are removed prior to writing each sequence. This format is deliberately verbose in order to be simple to duplicate.
  
  

• Genbank format: GDE can read a concatenated list of Genbank entries, and extract certain fields from such files. The default method for storing nucleic acid, amino acid, masking sequences or text is in Genbank format. The following fields are recognized:
  
  
  • LOCUS: Short name for this sequence (Maximum of 32 characters)Ý
  • DEFINITION: Definition of sequence (Maximum of 80 characters)
  • ORGANISM: Full name of organism (Maximum of 80 characters)
  • AUTHORS: Authors of this sequence (Maximum of 80 characters)
  • ACCESSION: ID Number for this sequence (Maximum of 80 characters)
  • ORIGIN: Beginning of sequence dataÝ
  • // End of sequence dataÝ

  
  
  All other lines are retained as comments. The LOCUS line also specifies what type of sequence follows. The form of this line is:
  
  

  • LOCUS name size bp type date

  
  
  where name is the Genbank Locus name, size is total base count, type is one of DNA, RNA, PROTEIN, MASK, or TEXT and date is of the form dd-MON-yyyy. In this way, the standard Genbank format is extended to store all text, mask and protein data. The Genbank character set has also been extended in order to support these other data types. Valid characters are:
  
  

  • DNA/RNA: Full IUPAC coding as well as '-' and '~' characters for alignment gaps
  • Protein: All valid single letter codes plus '-' and '~'. Other ASCII characters may be inserted, however external functions may be confused by such characters.
  • Mask: All legal printable ASCII characters. If used as a selection mask, all columns containing a '0' will be removed from any analysis.
  • Text: All valid ASCII characters.

Here is a valid Genbank entry for two E.coli tRNA's:

LOCUS ECOTRNT4 76 bp RNA 28-JAN-1991

DEFINITION E. coli (T4 infected) vulnerable tRNA (A).

ORGANISM Escherichia coli AUTHORS Amitsur,M., Levitz,R. and Kaufmann,G.

FEATURES From To/Span Description tRNA 1 76 vulnerable tRNA(A)

BASE COUNT ?

ORIGIN

1 GGGUCGUUAG CUCAGUUGGU AGAGCAGUUG ACUUUUAAUC AAUUGGNCGC AGGUUCGAAU 61 CCUGCACGAC CCACCA

//

 

LOCUS

ECOTRQ1 75 bp RNA 28-JAN-1991

DEFINITION E.coli Gln-tRNA-1.

ORGANISM Escherichia coli

AUTHORS Yaniv,M. and Folk,W.R.

SOURCE -REFERENCE [1] JOURNAL J. Biol. Chem. 250, 3243-3253 (1975)

FEATURES From To/Span Description tRNA 1 75 Gln-tRNA-1 (NAR: 0510) refnumbr 1 1 sequence not numbered in [1]

BASE COUNT ?

ORIGIN

1 UGGGGUAUCG CCAAGCGGUA AGGCACCGGU UUUUGAUACC GGCAUUCCCU GGUUCGAAUC 61 CAGGUACCCC AGCCA

//

 

• Flat file format: This is a simplified format for importing sequence data, and passing it out to analysis functions. Very little information is actually retained in this format, and should be used carefully so as not to lose attribute information. It is defined as follow:
  
  
  • type_character short_name
  • sequence_data
  • sequence_data
  • sequence_data
  • ...

  
  
  The type character is # for DNA/RNA, % for protein sequence, @ for mask sequence, and " for text. The short name is the same as the LOCUS line in Genbank. This is followed by lines of sequence, each ending with a return character.These lines are read until the next type character is encountered, or until the end of the file is reached. Care should be taken in using this format with text as space characters are stripped automatically. As of release 2.0, flat file format allows for an optional offset to be specified in parentheses after the sequence name. An offset represents how many leading gap characters should be placed before the start of a sequence. If this offset does not exist, then it is defined to be 0.
  
  Here is a sample flat file for two Ecoli tRNA's:
  
  #ECOTRNT4 GGGUCGUUAGCUCAGUUGGUAGAGCAGUUGACUUUUAAUCAAUUGGNCGCAGGUUCGAAU CCUGCACGACCCACCA

  #ECOTRQ1 UGGGGUAUCGCCAAGCGGUAAGGCACCGGUUUUUGAUACCGGCAUUCCCUGGUUCGAAUC CAGGUACCCCAGCCA
  
  

• Color mask: The format for a color mask has been kept simple to make implementation of color functions easy. The format optionally defines which sequence to color, whether or not to color alignment gaps in the existing sequence, and how long the following mask will be. It is then followed by a list of decimal color codes (range 0 to 15) for each position in the sequence. There are four keywords used in the color mask file. Those keywords are:
  
  
  • name:short name. If short name matches a currently loaded sequence, then impose this color mask on that sequence. If this line is omitted, then color all sequences this color, and the color mask is expected to start at the leftmost column on the screen.
  • length:length. The following list in length long
  • nodash: Skip over dash characters when imposing this color mask on the named sequence. This allows an unaligned color mask to be placed over aligned sequence.
  • start: Begin reading the color mask on the next line.

  
  
  Here is a sample color mask file:

  
  
  name:test_sequence

  length:10

  nodash:

  start: 3 3 3 6 5 3 3 3 2 7

  The colors in the default color lookup table are: 0 White 8 Black 1 Yellow 9 Grey 1 2 Violet 10 Grey 2 3 Red 11 Grey 3 4 Aqua 12 Grey 4 5 Lime Green 13 Grey 5 6 Blue 14 Grey 6 7 Purple 15 White