Semantic resources project/Use Cases/Meeting Notes/BradHyman/Meeting 10062009

PPP3CA, aka "Calcineurin"

XIAP, "X-linked inhibitor of apoptosis"

Meeting: October 6th, 9am In attendance: BH, TC, TWD

BH: we might start by imagining that we want to find antibodies to calmodulin, or calcineurin.
 * we can search databases for that name...
 * ... but remember: "Don't trust Santa Cruz."

BH: this is a common activity that we want to perform:
 * For a given protein,
 * find the antibodies that have been used to detect it,
 * and from there, find the papers that used it (_along with the way in which the antibody was used_)
 * there's an evaluation step in there: a grad student, or someone, will take the paper, look at the figures, and figure out "did it work alright in this paper?"
 * and then they email the corresponding author for materials.
 * TWD: Use case PROTEIN -> ANTIBODIES -> PAPERS (-> CORRESPONDING AUTHORS?)

BH: questions for antibodies also include, "what is the epitope of the antibody," or "is this a neo-epitope antibody?"
 * From a Nature Glossary, an "anti-neoepitope antibody" is a "An antibody that specifically recognizes the free amino or carboxyl groups of the amino-acid residues from a cleaved scissile bond that forms the new N and C termini of the cleaved product."

BH: there is what I would call "the XIAP problem"
 * 7000 different nomenclatures
 * spliceforms, variants
 * How do we get it? Can we clone it ourselves? Is it already in a vector sitting in a supplier somewhere?
 * Question: can I find someone who "has it in their fridge?"
 * There's a lot of pressure to share materials and reagents: social pressure, NIH requirements
 * TWD: suggests that a way to specify "what we have in our fridge" is an important component here.

TC: "experience using..." statements
 * Codifying "lore" or "cultural knowledge"
 * It's not just the antibody information, or the reagent information, it's also the question of "did this work for me?," or "did it work for you?"

TWD: Expanding the question to "reagents in general," what information would be useful in your everyday research?
 * BH: anything where alternate nomenclature is an issue
 * there are multiple genes called XIAP, and we want to make sure we find the *correct* mouse model that expresses a particular variant.
 * antibodies have multiple (common) names, and synonyms, and multiple antibodies with the same name.
 * TC: asks whether more detailed mouse model information would be immediately useful?
 * BH: Doesn't like the JAX website, but it has the information that they need.
 * BH: there are two websites where all this information is available, it's already consolidated
 * BH: of course, you run into the same nomenclature issue ("which XIAP") that you do with antibodies or genes.

TWD: Use case, GENE (ALLELE) NAMES -> CORRECT MOUSE MODELS

TC: there's also the PAPERS -> REAGENTS direction
 * actual protocol data is less important
 * BH: basically, if you can find a few papers, you can get a grad student to walk through the protocols and figure out (stepwise) the differences
 * BH: The real problem is, you can search for "calcineurin", and you get 500+ papers.
 * How can we drill down?
 * Could we drill down by reagents? eg. antibodies, or experimental type
 * An example -- search for calcineurin papers, then find those that used a particular antibody, and then find the subset of those that used it in an IP.
 * FILTERING.

BH: must remember, antibodies have lots of degenerate naming schemes.
 * Again: don't trust SC.