--$Revision: 1.9 $
--************************************************************************
--
-- ASN.1 and XML for the components of a GenBank/EMBL/DDBJ sequence record
-- The International Nucleotide Sequence Database (INSD) collaboration
-- Version 1.6, 25 May 2010
--
--************************************************************************

INSD-INSDSeq DEFINITIONS ::=
BEGIN

--  INSDSeq provides the elements of a sequence as presented in the
--    GenBank/EMBL/DDBJ-style flatfile formats, with a small amount of
--    additional structure.
--    Although this single perspective of the three flatfile formats
--    provides a useful simplification, it hides to some extent the
--    details of the actual data underlying those formats. Nevertheless,
--    the XML version of INSD-Seq is being provided with
--    the hopes that it will prove useful to those who bulk-process
--    sequence data at the flatfile-format level of detail. Further 
--    documentation regarding the content and conventions of those formats 
--    can be found at:
--
--    URLs for the DDBJ, EMBL, and GenBank Feature Table Document:
--    http://www.ddbj.nig.ac.jp/FT/full_index.html
--    http://www.ebi.ac.uk/embl/Documentation/FT_definitions/feature_table.html
--    http://www.ncbi.nlm.nih.gov/projects/collab/FT/index.html
--
--    URLs for DDBJ, EMBL, and GenBank Release Notes :
--    ftp://ftp.ddbj.nig.ac.jp/database/ddbj/ddbjrel.txt
--    http://www.ebi.ac.uk/embl/Documentation/Release_notes/current/relnotes.html
--    ftp://ftp.ncbi.nih.gov/genbank/gbrel.txt
--
--    Because INSDSeq is a compromise, a number of pragmatic decisions have
--    been made:
--
--  In pursuit of simplicity and familiarity a number of fields do not
--    have full substructure defined here where there is already a
--    standard flatfile format string. For example:
--
--   Dates:      DD-MON-YYYY (eg 10-JUN-2003)
--
--   Author:     LastName, Initials  (eg Smith, J.N.)
--            or Lastname Initials   (eg Smith J.N.)
--
--   Journal:    JournalName Volume (issue), page-range (year)
--            or JournalName Volume(issue):page-range(year)
--            eg Appl. Environ. Microbiol. 61 (4), 1646-1648 (1995)
--               Appl. Environ. Microbiol. 61(4):1646-1648(1995).
--
--  FeatureLocations are representated as in the flatfile feature table,
--    but FeatureIntervals may also be provided as a convenience
--
--  FeatureQualifiers are represented as in the flatfile feature table.
--
--  Primary has a string that represents a table to construct
--    a third party (TPA) sequence.
--
--  other-seqids can have strings with the "vertical bar format" sequence
--    identifiers used in BLAST for example, when they are non-INSD types.
--
--  Currently in flatfile format you only see Accession numbers, but there 
--    are others, like patents, submitter clone names, etc which will 
--    appear here
--
--  There are also a number of elements that could have been more exactly
--    specified, but in the interest of simplicity have been simply left as
--    optional. For example:
--
--  All publicly accessible sequence records in INSDSeq format will
--    include accession and accession.version. However, these elements are 
--    optional in optional in INSDSeq so that this format can also be used   
--    for non-public sequence data, prior to the assignment of accessions and 
--    version numbers. In such cases, records will have only "other-seqids".
--
--  sequences will normally all have "sequence" filled in. But contig records
--    will have a "join" statement in the "contig" slot, and no "sequence".
--    We also may consider a retrieval option with no sequence of any kind
--    and no feature table to quickly check minimal values.
--
--  Four (optional) elements are specific to records represented via the EMBL
--    sequence database: INSDSeq_update-release, INSDSeq_create-release,
--    INSDSeq_entry-version, and INSDSeq_database-reference.
--
--  One (optional) element is specific to records originating at the GenBank
--    and DDBJ sequence databases: INSDSeq_segment.
--
--********

INSDSet ::= SEQUENCE OF INSDSeq

INSDSeq ::= SEQUENCE {
    locus VisibleString OPTIONAL ,
    length INTEGER ,
    strandedness VisibleString OPTIONAL ,
    moltype VisibleString ,
    topology VisibleString OPTIONAL ,
    division VisibleString OPTIONAL ,
    update-date VisibleString OPTIONAL ,
    create-date VisibleString OPTIONAL ,
    update-release VisibleString OPTIONAL ,
    create-release VisibleString OPTIONAL ,
    definition VisibleString OPTIONAL ,
    primary-accession VisibleString OPTIONAL ,
    entry-version VisibleString OPTIONAL ,
    accession-version VisibleString OPTIONAL ,
    other-seqids SEQUENCE OF INSDSeqid OPTIONAL ,
    secondary-accessions SEQUENCE OF INSDSecondary-accn OPTIONAL,

--  INSDSeq_project has been deprecated in favor of INSDSeq_xrefs .
--  This element may be be removed from a future version of this DTD.

    project VisibleString OPTIONAL ,

    keywords SEQUENCE OF INSDKeyword OPTIONAL ,
    segment VisibleString OPTIONAL ,
    source VisibleString OPTIONAL ,
    organism VisibleString OPTIONAL ,
    taxonomy VisibleString OPTIONAL ,
    references SEQUENCE OF INSDReference OPTIONAL ,
    comment VisibleString OPTIONAL ,
    comment-set SEQUENCE OF INSDComment OPTIONAL ,
    struc-comments SEQUENCE OF INSDStrucComment OPTIONAL ,
    primary VisibleString OPTIONAL ,
    source-db VisibleString OPTIONAL ,
    database-reference VisibleString OPTIONAL ,
    feature-table SEQUENCE OF INSDFeature OPTIONAL ,
    feature-set SEQUENCE OF INSDFeatureSet OPTIONAL ,
    sequence VisibleString OPTIONAL ,  -- Optional for contig, wgs, etc.
    contig VisibleString OPTIONAL ,
    alt-seq SEQUENCE OF INSDAltSeqData OPTIONAL ,

--  INSDSeq_xrefs provides cross-references from a sequence record
--  to other database resources. These cross-references are at the
--  level of the entire record, rather than at the level of a specific
--  feature. These cross-references can include: BioProject, BioSample,
--  Sequence Read Archive, etc.

    xrefs SEQUENCE OF INSDXref OPTIONAL
}

INSDSeqid ::= VisibleString

INSDSecondary-accn ::= VisibleString

INSDKeyword ::= VisibleString

-- INSDReference_position contains a string value indicating the
-- basepair span(s) to which a reference applies. The allowable
-- formats are:
--
--   X..Y  : Where X and Y are integers separated by two periods,
--           X >= 1 , Y <= sequence length, and X <= Y 
--
--           Multiple basepair spans can exist, separated by a
--           semi-colon and a space. For example : 10..20; 100..500
--             
--   sites : The string literal 'sites', indicating that a reference
--           provides sequence annotation information, but the specific
--           basepair spans are either not captured, or were too numerous
--           to record.
--
--           The 'sites' literal string is singly occuring, and
--            cannot be used in conjunction with any X..Y basepair spans.
--
--           'sites' is a convention utilized by GenBank, and might
--           not be presented in XML provided by EMBL and DDBJ.
--
--   References that lack an INSDReference_position element are not
--   attributed to any particular region of the sequence.

INSDReference ::= SEQUENCE {
    reference VisibleString ,
    position VisibleString OPTIONAL ,
    authors SEQUENCE OF INSDAuthor OPTIONAL ,
    consortium VisibleString OPTIONAL ,
    title VisibleString OPTIONAL ,
    journal VisibleString ,
    xref SEQUENCE OF INSDXref OPTIONAL ,
    pubmed INTEGER OPTIONAL ,
    remark VisibleString OPTIONAL
}

INSDAuthor ::= VisibleString

-- INSDXref provides a method for referring to records in
-- other databases. INSDXref_dbname is a string value that
-- provides the name of the database, and INSDXref_dbname
-- is a string value that provides the record's identifier
-- in that database.

INSDXref ::= SEQUENCE {
    dbname VisibleString ,
    id VisibleString
}

INSDComment ::= SEQUENCE {
    type VisibleString OPTIONAL ,
    paragraphs SEQUENCE OF INSDCommentParagraph
}

INSDCommentParagraph ::= VisibleString

INSDStrucComment ::= SEQUENCE {
    name VisibleString OPTIONAL ,
    items SEQUENCE OF INSDStrucCommentItem
}

INSDStrucCommentItem ::= SEQUENCE {
    tag VisibleString OPTIONAL ,
    value VisibleString OPTIONAL ,
    url VisibleString OPTIONAL
}

-- INSDFeature_operator contains a string value describing
-- the relationship among a set of INSDInterval within
-- INSDFeature_intervals. The allowable formats are:
--
--   join :  The string literal 'join' indicates that the
--           INSDInterval intervals are biologically joined
--           together into a contiguous molecule.
--
--   order : The string literal 'order' indicates that the
--           INSDInterval intervals are in the presented
--           order, but they are not necessarily contiguous.
--
--   Either 'join' or 'order' is required if INSDFeature_intervals
--   is comprised of more than one INSDInterval .

INSDFeatureSet ::= SEQUENCE {
    annot-source VisibleString OPTIONAL ,
    features SEQUENCE OF INSDFeature
}

INSDFeature ::= SEQUENCE {
    key VisibleString ,
    location VisibleString ,
    intervals SEQUENCE OF INSDInterval OPTIONAL ,
    operator VisibleString OPTIONAL ,
    partial5 BOOLEAN OPTIONAL ,
    partial3 BOOLEAN OPTIONAL ,
    quals SEQUENCE OF INSDQualifier OPTIONAL ,
    xrefs SEQUENCE OF INSDXref OPTIONAL
}

-- INSDInterval_iscomp is a boolean indicating whether
-- an INSDInterval_from / INSDInterval_to location
-- represents a location on the complement strand.
-- When INSDInterval_iscomp is TRUE, it essentially
-- confirms that a 'from' value which is greater than
-- a 'to' value is intentional, because the location
-- is on the opposite strand of the presented sequence.

-- INSDInterval_interbp is a boolean indicating whether
-- a feature (such as a restriction site) is located
-- between two adjacent basepairs. When INSDInterval_interbp
-- is TRUE, the 'from' and 'to' values will differ by
-- exactly one base for linear molecules. For circular 
-- molecules, if the inter-basepair position falls between
-- the last and the first base, then 'from' will be the
-- final base (equal to the length of the sequence), and
-- 'to' will have a value of 1.

INSDInterval ::= SEQUENCE {
    from INTEGER OPTIONAL ,
    to INTEGER OPTIONAL ,
    point INTEGER OPTIONAL ,
    iscomp BOOLEAN OPTIONAL ,
    interbp BOOLEAN OPTIONAL ,
    accession VisibleString
}

INSDQualifier ::= SEQUENCE {
    name VisibleString ,
    value VisibleString OPTIONAL
}

-- INSDAltSeqData provides for sequence representations other than
-- literal basepair abbreviations (INSDSeq_sequence), such as the
-- CONTIG/CO linetype of the GenBank and EMBL flatfile formats.
-- It also accomodates the specification of accession-number ranges,
-- which are presented on a WGS master record (for the contigs and
-- and scaffolds of a WGS project).

INSDAltSeqData ::= SEQUENCE {
    name VisibleString ,  -- e.g., contig, wgs, scaffold, cage, genome
    items SEQUENCE OF INSDAltSeqItem OPTIONAL
}

INSDAltSeqItem ::= SEQUENCE {
    interval INSDInterval OPTIONAL ,
    isgap BOOLEAN OPTIONAL ,
    gap-length INTEGER OPTIONAL ,
    gap-type VisibleString OPTIONAL ,
    gap-linkage VisibleString OPTIONAL ,
    gap-comment VisibleString OPTIONAL ,
    first-accn VisibleString OPTIONAL ,
    last-accn VisibleString OPTIONAL ,
    value VisibleString OPTIONAL
}

END