# bioseq4.schema
# ldap schema for bio-sequence databanks in Bio-Mirror project
# minimal info common to many, or needed for some
# version 0.4, dec02, d.gilbert

## ?? make this schema self-contained? (no refs to core.schema, other ldap schema)

## grep -- '-oid' bioseq4.schema |\
##   perl -e 'while(<>){if (/([\w\-]+-oid)/) \
##   {$o= $1;unless($h{$o}++){ $n++;print "objectidentifier $o 1953.$n\n";}}}' \
##   > ! bioseq4.oids

# 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

## new schema format for openldap 2.1.x
objectIdentifier BioMirrorRoot 1.129.79.225200

objectIdentifier BioSeq BioMirrorRoot:1
objectIdentifier BioSeqattributeType BioSeq:3
objectIdentifier BioSeqObjectClass   BioSeq:4


# = srs field
attributetype ( BioSeqattributeType:1
        NAME ( 'kw' 'key' 'Keywords' )
        DESC 'Keywords'
        EQUALITY caseIgnoreMatch
        SUBSTR caseIgnoreSubstringsMatch
        SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )

# = srs field
attributetype ( BioSeqattributeType:2
        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 ( BioSeqattributeType:3
        NAME ( 'cc' 'Comment' )
        DESC 'Comment'
        EQUALITY caseIgnoreMatch
        SUBSTR caseIgnoreSubstringsMatch
        SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )

attributetype ( BioSeqattributeType:4
        NAME ( 'cr' 'crc' )
        DESC 'CRC value'
        EQUALITY caseIgnoreMatch
        SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )

attributetype ( BioSeqattributeType:5 
        NAME ( 'fl' 'file' )
        DESC 'Databank file'
        EQUALITY caseExactMatch
        SUBSTR caseIgnoreSubstringsMatch
        SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )

# = srs library key 
attributetype ( BioSeqattributeType:6 
        NAME ( 'lib' 'Library' )
        DESC 'Databank library'
        EQUALITY caseIgnoreMatch
        SUBSTR caseIgnoreSubstringsMatch
        SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )

attributetype ( BioSeqattributeType:6.1 
        NAME ( 'dir' 'Directory' )
        DESC 'Biodata Directory'
        EQUALITY caseIgnoreMatch
        SUBSTR caseIgnoreSubstringsMatch
        SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )

attributetype ( BioSeqattributeType:6.2 
        NAME ( 'srv' 'service' )
        DESC 'Bioinformatics Service'
        EQUALITY caseIgnoreMatch
        SUBSTR caseIgnoreSubstringsMatch
        SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )


# ? use this or use generic 'category' attribute
attributetype ( BioSeqattributeType:7 
        NAME ( 'libgroup' 'LibraryGroup' )
        DESC 'Databank library category'
        EQUALITY caseIgnoreMatch
        SUBSTR caseIgnoreSubstringsMatch
        SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )

attributetype ( BioSeqattributeType:8 
	NAME ( 'field' 'DatabankField' )
	DESC 'Databank Field'
	EQUALITY caseIgnoreMatch
	SUBSTR caseIgnoreSubstringsMatch
	SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )
	
attributetype ( BioSeqattributeType:9 
	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 ( BioSeqattributeType:10
        NAME ( 'div' 'Divison' )
        DESC 'Databank division'
	EQUALITY caseIgnoreMatch
	SUBSTR caseIgnoreSubstringsMatch
	SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )

attributetype ( BioSeqattributeType:11  
	NAME ( 'count' )
	DESC 'count of items'
	EQUALITY integerMatch
	SYNTAX 1.3.6.1.4.1.1466.115.121.1.27 ) 

attributetype ( BioSeqattributeType:12  
	NAME ( 'in' 'index' )
	DESC 'index of items'
	EQUALITY integerMatch
	SYNTAX 1.3.6.1.4.1.1466.115.121.1.27 ) 

attributetype ( BioSeqattributeType:13  
	NAME ( 'start' 'Startitem' )
	DESC 'start of items'
	EQUALITY integerMatch
	SYNTAX 1.3.6.1.4.1.1466.115.121.1.27 ) 

attributetype ( BioSeqattributeType:13.1  
	NAME ( 'query' 'filter' )
	DESC 'Query string'
	EQUALITY caseIgnoreMatch
	SUBSTR caseIgnoreSubstringsMatch
	SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )
	
#attributetype ( BioSeqattributeType:14 
#	NAME 'data-ref' 
#	DESC 'referral to ldap://site-info containing data'
#	EQUALITY caseIgnoreMatch
#	SUBSTR caseIgnoreSubstringsMatch
#	SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )
#was ref, ldap v2.1.9 won't allow !??	
# SUP ref  


## add from gnomap2.schema for gnomap ldif data
attributetype ( BioSeqattributeType:15  
        NAME ( 'chr' 'Chromosome' )
        DESC 'Chromosome'
        EQUALITY caseIgnoreMatch
        SUBSTR caseIgnoreSubstringsMatch
        SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )

attributetype ( BioSeqattributeType:16   
        NAME ( 'ft' 'ftk' 'Feature' )
        DESC 'Sequence feature'
        EQUALITY caseIgnoreMatch
        SUBSTR caseIgnoreSubstringsMatch
        SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )

attributetype ( BioSeqattributeType:17 
        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 ( BioSeqattributeType:18
        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 ( BioSeqattributeType:19 
  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 ( BioSeqattributeType:20 
	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 ( BioSeqattributeType:21 
  NAME  ('ac' '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 ( BioSeqattributeType:22 
	NAME ( 'os' '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 ( BioSeqattributeType:23  
	NAME ( 'dt' 'dat' 'lsu' 'lau' '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 ( BioSeqattributeType:24 NAME 'url'
		DESC 'universal resource locator'
		SUP labeledURI
		EQUALITY caseIgnoreMatch
		SUBSTR caseIgnoreSubstringsMatch )

attributetype ( BioSeqattributeType:25 NAME 'web'
	DESC 'web page' SUP labeledURI )

attributetype ( BioSeqattributeType:25.1 NAME  ('taxon' 'taxonomy' )
		DESC 'taxonomic or phylogenetic classification'
		EQUALITY caseIgnoreMatch
		SUBSTR caseIgnoreSubstringsMatch
		SYNTAX 1.3.6.1.4.1.1466.115.121.1.15 )
        
# ? use standard Definition attribute
# = srs field
attributetype ( BioSeqattributeType:26  
	NAME ( 'de' '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 ( BioSeqattributeType:27  
	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 ( BioSeqattributeType:28  
	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 ( BioSeqattributeType:29  
	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 ( BioSeqattributeType:30 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 ( BioSeqattributeType:31 
	NAME ( 'sq' '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 ( BioSeqattributeType:32  
	NAME ( 'bseq' 'bna' 'binary-nucleic-acid' )
	DESC 'binary coded sequence of nucleic acid data'
	SYNTAX 1.3.6.1.4.1.1466.115.121.1.5
	) 
# SUP nucleic-acid

attributetype ( BioSeqattributeType:33  
	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 ( BioSeqattributeType:33.1  
	NAME ( 'baa' 'binary-amino-acid' )
	DESC 'binary coded sequence of amino acid data'
	SYNTAX 1.3.6.1.4.1.1466.115.121.1.5
	) 
# SUP amino-acid

attributetype ( BioSeqattributeType:34 
	NAME ( 'drec' 'data-record' )
	DESC 'biology data record (native format)'
	EQUALITY caseIgnoreMatch
	SYNTAX 1.3.6.1.4.1.1466.115.121.1.15
	) 

attributetype ( BioSeqattributeType:35  
	NAME ( 'brec' 'compressed-data-record' )
	DESC 'gzip compressed data record'
	SYNTAX 1.3.6.1.4.1.1466.115.121.1.5
	) 
# SUP data-record

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

objectClass ( BioSeqObjectClass:1 
  NAME ( 'biodir' 'BioseqDirectory' )
	DESC 'Biosequence Data directory'
	SUP top 
  MUST ( id ) 
	MAY ( cn $ name $ lib $ des $ url $ web $ con )  
	)

# this is object for directory lib attribute (id == directory.lib ?)
objectClass ( BioSeqObjectClass:2 
  NAME 'BioseqDatabank'
	DESC 'Biosequence Databank'
	SUP top 
  MUST ( id ) 
	MAY ( cn $ name $ lib $ des $ url $ web  $ con $ fmt ) 
	)

# add some set range info: start,length ; also flavors for Entry/Record/BinRecord ?
objectClass ( BioSeqObjectClass:3 
  NAME 'BioseqSet'
	DESC 'Set of Biosequence Entries'
	SUP top 
  MUST ( id ) 
	MAY (  cn $ name $ lib $ des $ url $ web  $ con $ fmt
		 $ fmt $ cc $ key $ des $ dat $ spp  )
	)
##	SUP BioseqDatabank 
# data-ref $ ref

objectClass ( BioSeqObjectClass:4 
  NAME 'BiodataFile'
	DESC 'File of Biodatabank Records'
	SUP top 
  MUST ( id ) 
	MAY (  cn $ name $ lib $ des $ url $ web  $ con $ fmt
		$ cc $ key $ dat $ spp  )
	)
#	SUP BioseqDatabank 
# data-ref $ ref 

# general data object - nonsequence - SUP for Bioseq?
# should also be extensibleObject
#objectclass ( BioSeqObjectClass:5 NAME 'Biodata'
#	DESC 'Biodata Entry'
#  MUST ( id ) 
#	MAY (  data-ref $ ref $ fmt $ name $ cc $ key $ lib $ des $ dat $ spp  )
#	)

objectClass ( BioSeqObjectClass:6 
  NAME 'BiodataRecord'
	DESC 'Biodata Record'
  MUST ( id ) 
	MAY (  cn $ name $ acc $ cc $ key $ lib $ des $ format
	    $ Date $ Species 
			$ file $ crc $ index $ drec $ brec 
		 )
	)
	
#			$ data-ref $ ref  
	
#objectclass ( BioSeqObjectClass:7 NAME 'BioseqEntry'
#	DESC 'Biosequence Entry'
#  MUST ( id $ acc ) 
#	MAY (  fmt $ cc $ div $ gen $ key $ lib $ des $ dat $ mol $ sl $ spp  )
#	)
#
#objectclass ( BioSeqObjectClass:8 NAME 'BioseqRecord'
#	DESC 'Biosequence Record'
#	SUP BioseqEntry 
#	MAY ( seq )
#	)
#
#objectclass ( BioSeqObjectClass:9 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 ( BioSeqObjectClass:10 NAME 'NA-sequence'
#	DESC 'Nucleic acid sequence record'
#	SUP BioseqEntry 
#	MAY ( na $ bna )
#	)
#	
#objectclass ( BioSeqObjectClass:11 NAME 'AA-sequence'
#	DESC 'Amino acid sequence record'
#	SUP BioseqEntry 
#	MAY ( aa $ baa )
#	)