Transcript Shane Burgess

AgBase
Shane Burgess, Fiona McCarthy
Mississippi State University
1. Who we are
• Ontology users:
– Experimentalists: model our own functional genomics data sets
– Tool developers: develop ontology-based tools for functional
modeling
– Ontology training: train researchers in the use of ontologybased tools
• Ontology developers:
– Burgess: involved in development of GO immune function
– McCarthy: involved in development of GO viral process terms…
– AgBase biocurators: GO & PO new term requests and changes
during biocuration
• Ontology contributors:
– funded to provide GO annotations for chicken, agricultural
animals, maize, cotton…
– funded to provide PO annotations for maize & cotton…
2. Why we use ontologies?
• Structure: ability to organize data
• Ability to capture data as information sources
grow.
• Tools: ability to model high throughput data
sets
– summarizing function
– functional enrichment analysis
– hypothesis-testing
• Comparative biology: use of ontological data
to understand differences between species
3. Ontology Tools we are using at
AgBase
• None
• Ontology contributors:
– Initially used excel!
– Developed in house biocuration interface for
annotation
• Ontology developers:
– Mostly minor development so we only use online
ontology browser tools (e.g., QuickGO, AmiGO)
4. Roadblocks
• How to annotate increasing number of sequences from
new sequencing technologies?
–
–
–
–
No literature
Short sequences – similarity/motif finding not effective
Noncoding RNAs
Move to tissue/phenotype ontologies to capture data?
• Visualization
– How to view functional data from multiple ontologies
– Expression data, functional data and genome browser?
– How to link gene products to pathways and phenotypes?
5. Vision
• Need to employ multiple ontologies to get
comprehensive functional modeling
– cross-ontology queries
– rapidly populating other functional ontologies (where
is the “IEA pipeline” equivalent?)
– enrichment analysis tools that handle multiple
ontologies
• Manual biocuration
– more effective at targeting manual effort towards
“high importance” genes
– moving annotation data to new species (pathogens,
fungus, plants, marine invertebrates…)
6. Collaborators?
• Type of person or resource who could help?
• Where are the people driving the need?