Bundle: mahco-hla
Graph name: http://purl.org/science/graph/mahco-hla

This bundle consists of an unmodified copy of the MaHCO (major histocompatibility complex) ontology:

  • DeLuca DS, Beisswanger E, Wermter J, Horn PA, Hahn U and Blasczyk R. MaHCO: an ontology of the major histocompatibility complex for immunoinformatic applications and text mining, Bioinformatics (Oxford, England), 25, 2064-2070, 2009.

We're mainly using this ontology to obtain URIs for HLA alleles, which are referenced from several other bundles (Bundles/hla, Bundles/omim, Bundles/pdb, Bundles/medline/alleles). For example, for an allele B*3554, the URI is http://purl.org/stemnet/HLA#B_3554, i.e. replace * with _ and add a common prefix.

Serotypes are not named using an underscore and are not zero-extended from single to double digit names. HLA-A8 and A*08 both correspond to the URI http://purl.org/stemnet/HLA#A8.

The alleles are arranged in a class hierarchy, which could prove useful. For example, to determine whether a given class is an HLA allele class, we might check whether the class is a subclass (transitively) of Human_MHC_Allele (http://purl.org/stemnet/HLA#Human_MHC_Allele). The classes for serotypes, e.g. B35, are superclasses of chain variant classes, e.g. B_3554.

The ontology has separate classes for DNA and for gene products (amino acid chains), related via hla:encodes restrictions (every HLA-A DNA chain encodes some HLA-A amino acid chain, which seems a bit backwards to me). See its documentation for more information.


About 25 MeSH subject headings seem to correspond to MaHCO-HLA chain classes. These relations are included, e.g.

@prefix mesh:  <http://purl.org/commons/record/mesh/> .
@prefix hla:   <http://purl.org/stemnet/HLA#> .
@prefix sc:    <http://purl.org/science/owl/sciencecommons/> .
mesh:D015797 sc:mesh-to-chain hla:B35 .

The sc:mesh-to-chain relation is provisional - it's not clear what the semantics of this relation is. It may be replaced by something from the OBO Foundry in the future.

See also: ImmPort