MonDO (Monarch Disease Ontology) is a semi-automatically constructed ontology that merges in multiple disease resources to yield a coherent merged ontology.
The procedure is broken into the following steps:
- Pre-processing of ontologies -- see the src/ directory
- Gathering of loose ontology mappings, and translating these into weighted candidate axioms
- Estimation of the most likely Ontology using kBOOM
- Merging equivalence sets and post-processing.
See the Makefile in the src/ontology directory for the execution of these steps.
Each disease resource or ontology is pre-processed. In some cases, only a subset of the ontology is used
- ORDO/Orphanet
- DO
- GARD -- aligned as post-processing step
- OMIM -- note we only use labels from OMIM
- MedGen -- not yet incorporated
- NCIT -- aligned as post-processing step
- OMIA -- Mendelian diseases in non-human animals
- MESH -- We use MEDIC as our initial pre-processed set
- DiseaseClusters -- Additional groupings of OMIM. Includes DECIPHER.
We use mappings as an input, but not as an end-goal. Our end goal is a coherent merged ontology, with strictly defined relationships between them (next step). However, we make use of mappings to generate candidate weighted axioms.
We make use of some ready-made mappings. Additionally, we supplant these with our own using approaches such as entity matching.
See README-editors