# bioseq.schema
# ldap schema for bio-sequences  
# version 0.3, jul02, d.gilbert

# Note: as much as possible, match attribute name to SRS databank field names
# to simplify srs2ldap: need mapping anyway, but LDAP enforces use of
# primary schema-defined names in query filters (substituting alternate for primary) 

## ldap Syntax Object Identifiers
#binary: 1.3.6.1.4.1.1466.115.121.1.5
#integer:1.3.6.1.4.1.1466.115.121.1.27
#directory string: 1.3.6.1.4.1.1466.115.121.1.15 
#printable string: 1.3.6.1.4.1.1466.115.121.1.44
#numeric string: 1.3.6.1.4.1.1466.115.121.1.36
#octet string: 1.3.6.1.4.1.1466.115.121.1.40

## to some extent this schema should be auto-generated from [srs] databanks/backend
## need to add srs fields or equivalents to handle
## common ones for bioseqs: acc, des, gen(geneName), div
## key, (dat/crd/crlu),  cc
## refs: aut, tit, jnl, vol, fp, yr, mid, pmd, rc.
# dbn = DbName == lib??
# dbxref == dr for swissprot

# = srs field
attributetype ( keywords-oid
        NAME ( 'key' 'Keywords' )
        DESC 'Keywords'
        EQUALITY caseExactMatch
        SUBSTR caseIgnoreSubstringsMatch
        SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )

# = srs field
attributetype ( genename-oid
        NAME ( 'gen' 'GeneName' )
        DESC 'Gene name'
        EQUALITY caseExactMatch
        SUBSTR caseIgnoreSubstringsMatch
        SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )

# = srs field
attributetype ( comment-oid
        NAME ( 'cc' 'Comment' )
        DESC 'Comment'
        EQUALITY caseExactMatch
        SUBSTR caseIgnoreSubstringsMatch
        SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )

# = srs library key 
attributetype ( library-oid 
        NAME ( 'lib' 'Library' )
        DESC 'Databank library'
        EQUALITY caseExactMatch
        SUBSTR caseIgnoreSubstringsMatch
        SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )

# ? use this or use generic 'category' attribute
attributetype ( libgroup-oid 
        NAME ( 'libgroup' 'LibraryGroup' )
        DESC 'Databank library category'
        EQUALITY caseExactMatch
        SUBSTR caseIgnoreSubstringsMatch
        SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )

attributetype ( DatabankField-attr-oid  
	NAME ( 'field' 'DatabankField' )
	DESC 'Databank Field'
	EQUALITY caseIgnoreMatch
	SUBSTR caseIgnoreSubstringsMatch
	SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )
	
attributetype ( DatabankLink-attr-oid  
	NAME ( 'link' 'DatabankLink' )
	DESC 'Linked databank'
	EQUALITY caseIgnoreMatch
	SUBSTR caseIgnoreSubstringsMatch
	SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )

# = srs field
attributetype ( division-oid
        NAME ( 'div' 'Divison' )
        DESC 'Databank division'
	EQUALITY caseIgnoreMatch
	SUBSTR caseIgnoreSubstringsMatch
	SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )

attributetype ( Count-oid  
	NAME ( 'count' )
	DESC 'count of items'
	EQUALITY integerMatch
	SYNTAX 1.3.6.1.4.1.1466.115.121.1.27 ) 

attributetype ( Startitem-oid  
	NAME ( 'start' 'Startitem' )
	DESC 'start of items'
	EQUALITY integerMatch
	SYNTAX 1.3.6.1.4.1.1466.115.121.1.27 ) 

attributetype ( Querystring-oid  
	NAME ( 'query' 'filter' )
	DESC 'Query string'
	EQUALITY caseIgnoreMatch
	SUBSTR caseIgnoreSubstringsMatch
	SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )
	
attributetype ( Data-ref-oid
	NAME 'data-ref' 
	DESC 'referral to ldap://site-info containing data'
	SUP ref  )


## add from gnomap2.schema for gnomap ldif data
attributetype ( Chromosome-attr-oid  
        NAME ( 'chr' 'Chromosome' )
        DESC 'Chromosome'
        EQUALITY caseIgnoreMatch
        SUBSTR caseIgnoreSubstringsMatch
        SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )

attributetype ( Feature-attr-oid  
        NAME ( 'ftk' 'Feature' )
        DESC 'Chromosome feature'
        EQUALITY caseIgnoreMatch
        SUBSTR caseIgnoreSubstringsMatch
        SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )

attributetype ( Loc-start-oid  
        NAME ( 'bpb' 'bstart' 'Base-start' )
        DESC 'start base number of feature'
        EQUALITY integerMatch
        SYNTAX 1.3.6.1.4.1.1466.115.121.1.27
        ) 

attributetype ( Loc-stop-oid  
        NAME ( 'bpe' 'bstop' 'Base-stop' )
        DESC 'last base number of feature'
        EQUALITY integerMatch
        SYNTAX 1.3.6.1.4.1.1466.115.121.1.27
        ) 

## in biomirror-catalog-spec.schema
#attributetype ( id-oid 
#  NAME 'id'
#	DESC 'object id'
#	EQUALITY caseExactMatch
#	SUBSTR caseIgnoreSubstringsMatch
#	SYNTAX 1.3.6.1.4.1.1466.115.121.1.15
#	SINGLE-VALUE )

## in biomirror-catalog-spec.schema
#attributetype ( format-oid 
#	NAME ( 'fmt' 'format' )
#	DESC 'data format'
#	EQUALITY caseIgnoreMatch
#	SUBSTR caseIgnoreSubstringsMatch
#	SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )

# = srs field
attributetype ( acc-oid 
  NAME 'acc'
	DESC 'accession'
	EQUALITY caseIgnoreMatch
	SUBSTR caseIgnoreSubstringsMatch
	SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )

# same as gnomap.schema
# org is srs field, make it primary name ??
attributetype ( Species-attr-oid  
	NAME ( 'org' 'spp' 'Species' )
	DESC 'Species'
	EQUALITY caseIgnoreMatch
	SUBSTR caseIgnoreSubstringsMatch
	SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )

## need to fix this to parsable date - number like 20020130
## crlu == swissprot/trembl/embl date updated
## do we need a different attribute for each date subtype ?
attributetype ( Date-oid  
	NAME ( 'dat' 'crlu' 'crd' 'Date' )
	DESC 'Date'
	EQUALITY caseIgnoreMatch
	SUBSTR caseIgnoreSubstringsMatch
	SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 ) 

## in biomirror-catalog-spec.schema
## replace with core labeledURI attr
#attributetype ( url-oid NAME 'url'
#        DESC 'universal resource locator'
#        SUP labeledURI
#        EQUALITY caseIgnoreMatch
#        SUBSTR caseIgnoreSubstringsMatch )


# ? use standard Definition attribute
# = srs field
attributetype ( Description-attr-oid  
	NAME ( 'des' )
	DESC 'Description'
	EQUALITY caseIgnoreMatch
	SUBSTR caseIgnoreSubstringsMatch
	SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )

## in gnomap schema
# srs field names vary (dr, ??)
attributetype ( Dbxref-oid  
	NAME ( 'dr' 'dbxref'  )
	DESC 'Database cross-reference'
	EQUALITY caseIgnoreMatch
	SUBSTR caseIgnoreSubstringsMatch
	SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 ) 

## in gnomap schema
attributetype ( content-oid  
	NAME ( 'con' 'content' )
	DESC 'data content types'
	EQUALITY caseIgnoreMatch
	SUBSTR caseIgnoreSubstringsMatch
	SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )

## should be numeric value
# = srs field
attributetype ( Seqlength-oid  
	NAME ( 'sl' 'Seqlength' )
	DESC 'length of sequence'
	EQUALITY integerMatch
	SYNTAX 1.3.6.1.4.1.1466.115.121.1.27
	) 
	
# this should be an enum: 
# = srs field
attributetype ( molecule-oid NAME 'mol'
	DESC 'Molecule type'
	EQUALITY caseIgnoreMatch
	SUBSTR caseIgnoreSubstringsMatch
	SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )

# na,bna,aa,baa are in gnomap.schema
# ? use mol type and only one seq attr for na/aa ?
attributetype ( NA-sequence-attr-oid  
	NAME ( 'seq' 'na' 'nucleic-acid' )
	DESC 'sequence of nucleic acid data'
	EQUALITY caseIgnoreMatch
	SYNTAX 1.3.6.1.4.1.1466.115.121.1.15
	) 

attributetype ( Binary-NA-sequence-attr-oid  
	NAME ( 'bseq' 'bna' 'binary-nucleic-acid' )
	DESC 'binary coded sequence of nucleic acid data'
	SUP nucleic-acid
	SYNTAX 1.3.6.1.4.1.1466.115.121.1.5
	) 

attributetype ( AA-sequence-attr-oid  
	NAME ( 'aa' 'amino-acid' )
	DESC 'sequence of amino acid data'
	EQUALITY caseIgnoreMatch
	SYNTAX 1.3.6.1.4.1.1466.115.121.1.15
	) 

attributetype ( Binary-AA-sequence-attr-oid  
	NAME ( 'baa' 'binary-amino-acid' )
	DESC 'binary coded sequence of amino acid data'
	SUP amino-acid
	SYNTAX 1.3.6.1.4.1.1466.115.121.1.5
	) 


# objects ------------

objectclass ( BioseqDirectory-oid NAME 'BioseqDirectory'
	DESC 'Biosequence Data directory'
	SUP top 
  MUST ( id ) 
	MAY ( name $ lib $ description $ url $ web $ con )  
	)

# this is object for directory lib attribute (id == directory.lib ?)
objectclass ( BioseqDatabank-oid NAME 'BioseqDatabank'
	DESC 'Biosequence Databank'
	SUP top 
  MUST ( id ) 
	MAY ( name $ lib $ description $ url $ web  $ con $ fmt ) 
	)

# add some set range info: start,length ; also flavors for Entry/Record/BinRecord ?
objectclass ( BioseqSet-oid NAME 'BioseqSet'
	DESC 'Set of Biosequence Entries'
	SUP BioseqDatabank 
	MAY (  data-ref $ ref $ fmt $ cc $ key $ des $ dat $ spp  )
	)

# general data object - nonsequence - SUP for Bioseq?
# should also be extensibleObject
objectclass ( Biodata-oid NAME 'Biodata'
	DESC 'Biodata Entry'
  MUST ( id ) 
	MAY (  data-ref $ ref $ fmt $ name $ cc $ key $ lib $ des $ dat $ spp  )
	)

objectclass ( BioseqEntry-oid NAME 'BioseqEntry'
	DESC 'Biosequence Entry'
  MUST ( id $ acc ) 
	MAY (  fmt $ cc $ div $ gen $ key $ lib $ des $ dat $ mol $ sl $ spp  )
	)

objectclass ( BioseqRecord-oid NAME 'BioseqRecord'
	DESC 'Biosequence Record'
	SUP BioseqEntry 
	MAY ( seq )
	)

objectclass ( BinaryBioseqRecord-oid NAME 'BinaryBioseqRecord'
	DESC 'Binary encoded Biosequence Record'
	SUP BioseqEntry 
	MAY ( bseq )
	)
 
## not same as gnomap.schema --- are these objects as well as attributes? 
## optional ascii or binary attribute ?
## can we do  MUST ( na | bna ) ??
objectclass ( NA-Sequence-oid NAME 'NA-sequence'
	DESC 'Nucleic acid sequence record'
	SUP BioseqEntry 
	MAY ( na $ bna )
	)
	
objectclass ( AA-Sequence-oid NAME 'AA-sequence'
	DESC 'Amino acid sequence record'
	SUP BioseqEntry 
	MAY ( aa $ baa )
	)