Time: 11.00 am
Day:  Tuesday, Feb 10
Speaker: Ben Shoemaker

Discovery of protein interactions and of new PubChem data

Proteins function by forming interactions with other proteins and with other types of molecules.  Comparisons of how related 
proteins interact with their environment reveal diverse mechanisms for adapting those functions.  We have used the wealth of
 atomic detail of interaction data from protein structures to create the Conserved Binding Mode Database and more recently a
 Protein Interaction Database.  A short review will be given of how protein-interaction data has been used to pursue a variety 
of research questions.  The role of the Protein Interaction Database is to extract interactions between proteins and all other 
biomolecular entities including ligands, DNA and RNA molecules, peptides and other proteins.  Beyond collecting observed interactions,
 the database infers interactions from similar protein structures and groups of similar binding interfaces.  All clusters of binding 
interfaces are ranked against each other for biological relevance.  By so doing, large numbers of interactions can be collapsed 
to display a concise summary of relevant interactions.  Once public, interaction annotations would be most visible from Protein 
Entrez and along with CDD curator annotations would provide increased discoverability within the Structure databases.

For the case of protein-ligand interactions, the more comprehensive way to analyze and store detailed functional information is 
to look at the bioassay activity data available through PubChem. One example of making data in PubChem more accessible would be 
to include interactions from protein structures as activity data in PubChem.  A few more PubChem examples will be highlighted 
to show the importance of handling new bioassay and chemical information (via the PubChem Deposition Gateway).