PRO

Protein Ontology Kick-Off Meeting

The inaugural meeting of the Protein Ontology project will take place in Georgetown University (Harris Building, Conference Room 4200), Washington DC on December 2-4, 2007. This is an internal meeting.

The goals of the meeting are:

• to present the PRO ontology and to subject it to preliminary critique;
• to clarify the relations between PRO and its neighbor ontologies, between PRO and the representations in protein sequence databases, and between PRO and GO sequence annotations;
• to establish plans for PRO's further development, dissemination and use.

Agenda

Sunday, December 2

• Arrival of Workshop Participants

6:30pm Dinner for Workshop Participants [Location to be Announced]

Monday, December 3

9:00am-10:30am Session 1: Introduction to the Protein Ontology (Cathy Wu)

10:45am-12:15pm Session 2: The Protein Ontology within the OBO Foundry (Barry Smith)

• With presentation by Lindsay Cowell on OBO Foundry Methodologies

12:15pm-1:30pm Lunch

1:30pm-3:00pm Session 3: The Protein Ontology and Its Neighbors (Darren Natale)

• With presentation by James W. Brown on the RNA Ontology and by Richard H. Scheuermann on HLA Proteins

3:30pm-5:00pm Session 4: PRO, GO and GO Annotations (Chris Mungall)

6:30pm Dinner [Location to be Announced]

Tuesday, December 4

8:30am Continental Breakfast

9:00am-10:30am Session 5: Ontologies and Databases (Alan Ruttenberg)

• With presentation by John Westbrook on the Protein DataBase (PDB)

10:45am-12:15pm Session 6: The Protein Ontology and Its Users (Judith Blake)

12:15pm-1:30pm Lunch

1:30pm-4:00pm Session 8: Next Steps (Suzanna Lewis)

Relations between PRO and the Gene Ontology

One goal of the meeting is to clarify the relations between PRO and those OBO Foundry ontologies with which it has the potential to overlap.

Relations between PRO and GO (The Gene Ontology)

The Gene Ontology term 'cellular component' is defined as meaning: the part of a cell or its extracellular environment in which a gene product is located. The following gloss is provided:

The GO-CC ontology accordingly defines parts of a cell or its extracellular environment in which gene products are located. This includes the cell itself, it's parts, both internal and external (endoplasmic reticulum, extracellular matrix), organelles and macromolecular complexes of various sizes and compositions, including protein complexes.

GO will contain terms for attributes of protein entities, including

• 1. terms for the molecular functions proteins perform
• 2. terms for the biological processes in which proteins are involved
• 3. terms for parts of cells (or the extracellular environment) in which proteins can be located, including terms for complexes of protein molecules.

GO will not contain terms for individual protein entities, and will not contain terms for transient multi-protein associations.

Protein complexes are defined by GO as: any protein group composed of two or more subunits, which may or may not be identical [in type]. This definition includes homodimers, but not single protein molecules. As an operational criterion, the GO excludes unstable or random assemblages of parts.

The GO will frequently describe the typical constituents of a protein complex in a free text description embedded in a definition of some other term, but not explicitly using part_of links. Protein complex members can also be extracted from annotations of gene products, but this information may be incomplete and yields only organism-specific protein sequence identifiers.

Relations between PRO and Other Neighbor Ontologies

Relations between PRO and SO (The Sequence Ontology) SO will contain

• 1. terms for features of sequences of protein molecules
• 2. terms for qualities of features of sequences of protein molecules
• 3. terms for variants of sequences of protein molecules

Relations between PRO and RNAO (The RNA Ontology)

RNAO will contain terms for evolutionary relations (like homology) between different RNA molecules and parts of homologous RNA -- homology at the motif or nucleotide level -- with the goal of representing such relations in databases and in multiple sequence alignments.

Relations between RNAO and SO

• SO will contain terms for 1-dimensional features of sequences.
• RNAO will contain terms for 2- and 3-dimensional features of sequences.

Participants

Cecilia Arighi (Georgetown University Medical Center)

Judith Blake (The Jackson Laboratory)

Vivien Bonazzi (NHGRI / NIH)

James W Brown (Department of Microbiology, NC State University)

Lindsay Cowell (Duke Institute for Genome Sciences & Policy)

Harold Drabkin (The Jackson Laboratory)

Karen Eilbeck (Eccles Institute of Human Genetics, University of Utah)

Zhangzhi Hu (Georgetown University Medical Center)

Suzanna Lewis (Lawrence Berkeley National Lab)

Jerry Li (NIGMS / NIH)

Hongfang Liu (Georgetown University Medical Center)

Chris Mungall (Lawrence Berkeley National Lab)

Darren Natale (Georgetown University Medical Center)

Alan Ruttenberg (Science Commons)

Richard Scheuermann (The University of Texas Southwestern Medical Center at Dallas)

Barry Smith (University at Buffalo)

John Westbrook (Rutgers, The State University of New Jersey)

Cathy Wu (Georgetown University Medical Center)

Protein Ontology Specific Aims

The Protein Ontology (PRO) project is funded by NIGMS / NIH Grant 1 R01 GM080646-01, PI: Cathy Wu.

The Aims of the project are:

Aim 1. Develop a Protein Evolution (ProEvo) ontology to describe proteins based on evolutionary relationships. In essence, ProEvo will reflect protein families (using sequence or structure similarities) in an ontology framework.

Aim 2. Develop a Protein Forms (ProForm) ontology to represent the multiple protein end-products from a gene. This will include "normal" and mutant forms, forms derived from different splice variants, and cleaved and post-translationally modified products.

Aim 3. Specify the relationships between the ProEvo, ProMod and other OBO Foundry ontologies. Several ontologies provide qualities that can be attributed to various forms of a protein or to an entire protein family. These qualities, in effect, can annotate the protein forms or families.

Aim 4. Disseminate PRO ontology, and demonstrate its usefulness in health-related research via scientific case studies.

Literature

Darren A. Natale, Cecilia N. Arighi, Winona Barker, Judith Blake, Ti-Cheng Chang, Zhangzhi Hu, Hongfang Liu, Barry Smith, and Cathy H. Wu, "Framework for a Protein Ontology", Proceedings of the First International Workshop on Text Mining in Bioinformatics, 2006, p. 29-36.[1]