Practical application of ontologies to annotate and analyse large scale raw mouse phenotype data

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Large-scale projects generate complex mouse knockout mutant phenotype data.
• Phenotype data can be annotated using single terms or combinatorial approaches (Entity-Quality model).
• Previous databases favored one annotation method, limiting data comparability.

Purpose of the Study:

• To implement and integrate both single-term and combinatorial ontology approaches for mouse phenotype data annotation.
• To capture and manage large-scale phenotype data generated by Standard Operating Procedures (SOPs).
• To facilitate seamless querying and comparison of phenotype data across different methodologies.

Main Methods:

• A four-tier annotation strategy was applied to SOP data.
• The Entity-Quality (EQ) model was used for individual parameter and qualitative data annotation.
• Mappings between the Mammalian Phenotype (MP) ontology and the EQ model were exploited for querying.

Main Results:

• The implemented system allows for four-tier annotation of phenotype data.
• Qualitative phenotypes were assessed using PATO qualities.
• Querying the database using community-accepted MP terms was enabled through ontology mappings.

Conclusions:

• Well-annotated and comparable phenotype databases are achievable using ontologically derived statements.
• The integrated approach allows scientists to work seamlessly with ontologies.
• This study is the first to implement both combinatorial and single-dedicated approaches for a single phenotypic dataset.