See also: IRC log
<michel> boycer: has been lurking from LODD, interested in TM group
<michel> ... Drug Interaction Knowledge Base http://thedatahub.org/dataset/the-drug-interaction-knowledge-base
<michel> ... developing use case with scientific discourse
<michel> ... some work could be of benefit to this group
boycer: Pieces of drug-interaction kb might be of interest here.
<michel> ... link shows abstract for an article that reports a human study on drug interaction
<michel> ... concludes "Both clopidogrel and ticlopidine significantly inhibited the CYP2B6-catalyzed bupropion hydroxylation. "
<michel> ... quantitative data in the results
<michel> ... dailymed ; *the* resource for drug labeling
<michel> ... structured product labeling
<michel> ... search for clopidogrel
<adrien> does not work here
<adrien> got it
<boycer> Therefore, the potential exists for a drug interaction between bupropion and drugs that are substrates of or inhibitors/inducers of the CYP2B6 isoenzyme (e.g., orphenadrine, thiotepa, cyclophosphamide, ticlopidine, and clopidogrel).
<michel> ... not found in the first, but is found in the second
<michel> ... interaction not mentioned in one of these package inserts
<michel> ... is it possible to build data using the semantic web so that this information is automatically propagated
<michel> ... built a prototype to give a mashup view
<michel> ... package insert for escitalopram
<michel> ... can we use simple string matching techniques to match compound in labels; what are the claims, and link to a resource that supports or refutes the claim
scribe: whole sentence can be linked; click goes out to PubMed
<michel> ... 5 slides about the drug interaction knowledge base
<michel> ... how does this differ from drugbank? drugbank is a much larger project
<michel> ... makes links to drug interactions, but no real evidence
<michel> ... dikb models the evidence for each of the claims
scribe: DIKB is an evidential system, with evidence both for and against hypothises
<michel> ... taxonomy of evidence (slide 3)
<michel> ... using D2R to make a SPARQL endpoint
<Lena_> drugs are not linked to drugbank?
<Lena_> no prob here
Can you distinguish between inhibition of MOA or of, say, metabolism, of a given drug?
<adrien> works well here
boycer: Not yet linked to drugbank, but working on
Lena: Can use SILK
... w partial string-matching or concatenation
<michel> boycer: for certain drugs, getting a lot of evidence regarding pharmacokinetics, etc
<michel> ... HCLS use case - let's do this on a bigger scale - more complete view of drug package insert
Lena: use case: proposing W3C use
this at a larger scale to create more complete view of drug
... to curate other drug sources
<michel> ... if info in literature, would sources agree that the information existed
<michel> bobP: Can you distinguish between MOA or metabolism, of a given drug?
<michel> boycer: pharmacokinetics - ADME ; MOA about targeting
Lena: Metabolic Inhibition Pharmaco KB
<BobF> PK = what the body does to the drug; PD = what the drug does to the body; mech of action = how the drug works
<michel> boycer: could DIKB link to the TMO?
Michel: What to do with your ontology-of-evidence?
boycer: Wrote paper on this
... would like to see this become part of larger effort re scientific claims
... what is the study type? useful to users, to limit evidence for claims
<michel> ... in vitro vs in vivo
boycer: clinical care evidence
different fro others
... tried also to automatically iterate thru evidence: Pharmacists were too conservative
<Lena_> can I use this in my projects? ;-)
Michel, boycer: This is more epistemological
Michel: HCLS long term strategy it to have a number of these ontologies to work together
boycer: Use case does not require
sophisticated evidence taxonomy
... just find an approach to mash-up w. labels
... Lena, complete for anti-depressants but not for some others
... complete for the drugs we have studied
... goal is to move resource to wikipedia w expert communities
... toward a clinical decision support tool in the long term.
Michel: AMIA oral talk
<BobF> reading now...
<mscottm> Google doc being discussed?
<Lena_> can I be given access on the doc, please?
<mscottm> oh, share privileges
Michel: BobF has set up a truncated version of dbDNP
BobF: Pulled down latest build,
only for human and only for mapped well to chromosome
... then pulled out fields, attributes. Filtered down to a subset of the model
BobF: can do another extraction,
or maybe pull dynamically
... SNP ids, assoc genes, location, etc
... took info, put into Michel DB.
Michel: Working from *far* south here
Scott: Would like to do w
... rather than D2R
Scott: What are SQL queries w SWObjects?
Michel: Will describe this...
Scott: SWObjects can create map,
... Lena has worked w SWObjects extensively
Lena: Linking to mySQL. queries are ugly,
<BobF> db has 4 tables, fields include rsID, chromosome, SNP class, validation status, alleles, and info for associated genes
Lena: so dream query, then map, can load SWObject w the dream query instead of SQL
<Lena_> can you hear me?
<adrien> yes :)
Michel: Will make a read-only account
<BobF> Information about associated genes include: gene symbol, locus ID, mRNA accession, functional class, and (if it affects the protein sequence) protein accession, frame, residue
Michel: Thanks to Richard for presentation.
(Michel now broken up...)
<adrien> michel, we can not here the end
<adrien> ear the end
<BobF> I need to run - thanks, all
<adrien> bye everyone
<michel> ACTION: item to michel to make read-only account to mysql database [recorded in http://www.w3.org/2011/10/20-hcls2-minutes.html#action01]
<michel> ACTION: item to lena / scott to create an SWObjects endpoint over the data, where the query is on the snpid - and get related data [recorded in http://www.w3.org/2011/10/20-hcls2-minutes.html#action02]
michel - is that a wrap?
This is scribe.perl Revision: 1.136 of Date: 2011/05/12 12:01:43 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) No ScribeNick specified. Guessing ScribeNick: bobP Inferring Scribes: bobP WARNING: No "Topic:" lines found. WARNING: No "Present: ... " found! Possibly Present: BobF Bob_Powers IPcaller Lena Lena_ Michel P2 Scott Scott_Bauer aaaa aabb aacc adrien bobP boycer epichler ericP https mscottm rkiefer You can indicate people for the Present list like this: <dbooth> Present: dbooth jonathan mary <dbooth> Present+ amy WARNING: No meeting title found! You should specify the meeting title like this: <dbooth> Meeting: Weekly Baking Club Meeting WARNING: No meeting chair found! You should specify the meeting chair like this: <dbooth> Chair: dbooth Got date from IRC log name: 20 Oct 2011 Guessing minutes URL: http://www.w3.org/2011/10/20-hcls2-minutes.html People with action items: item WARNING: No "Topic: ..." lines found! Resulting HTML may have an empty (invalid) <ol>...</ol>. Explanation: "Topic: ..." lines are used to indicate the start of new discussion topics or agenda items, such as: <dbooth> Topic: Review of Amy's report[End of scribe.perl diagnostic output]