See also: IRC log
ACTION: Alan asks Olivier about MeSH terms [recorded in http://www.w3.org/2007/03/19-BioRDF-minutes.html#action01]
<vipul> Don update on mapping of BAM names
ACTION: Elizabeth to ask June about states of Alz [recorded in http://www.w3.org/2007/03/19-BioRDF-minutes.html#action02]
ACTION: Alan to show example of curation contradiction, biology contradiction for demo, based on Kei's action [CONTINUES] [recorded in http://www.w3.org/2007/03/19-BioRDF-minutes.html#action03]
AlanR: two types of
contradiction
... statement -- two people say X said something
different
... biology -- brain has 3, no 4 (latent discovery),
receptors
... demonstrated as little snippets
... in how OWL reasoning helps us
AlanR: general issue is mapping
(sameAs or equivClass)
... Don looking at this for MESH terms
ACTION: Kei look at http://hissa.nist.gov/jb/biordf-demo/bams_cell_names.txt too. Alan send list to Cell ontology people for maps as well. [CONTINUES] [recorded in http://www.w3.org/2007/03/19-BioRDF-minutes.html#action05]
ACTION: Don will try to map these lists to MESH. Talk to Bill Bug. [CONTINUES] [recorded in http://www.w3.org/2007/03/19-BioRDF-minutes.html#action06]
Olivier: OWL version of MESH is
unoffical and unsupported
... MESH requested it be withdrawn because it has bad
features:
... wrond IDs
... oversimplifies MESH
... leads to inaccurate relations
... (liver is a kind of abdomen)
AlanR: were planning to use it because they had URIs for MESH terms
Olivier: they are the wrong URIs
-- used tree number, which simply indicates the position in the
MESH tree
... should use MESH id (starts with D for main headings), and
not numbers with '.'s in them
... a main heading have several tree numbers
... not unique
AlanR: we need URIs for MESH terms
Olvier: point well taken
... for this purpose, make up a URI: NLM + "mesh" + MESH id
Matthias: did this. will send URL after meeting
Olivier: probably much to use
Matthias's version
... Gus Shriber also translated MESH to SKOS last
summer
...have mapped every UMLs concept to MESH
ACTION: ScottM to get a URL from Gus Shriber for MESH work [recorded in http://www.w3.org/2007/03/19-BioRDF-minutes.html#action07]
ACTION: Olivier to provide guidance to Don Doherty for finding mappings between BAM and MeSH [recorded in http://www.w3.org/2007/03/19-BioRDF-minutes.html#action08]
<samwaldmatthias> Here is the URL for my newer SKOS version of MeSH 2007, based on the unique MeSH IDs: http://neuroscientific.net/ont/bio-zen-MESH-new.zip
Kei: we have example inconsistent data in NeuronDB
ACTION: Kei to identify URL for terms in http://hissa.nist.gov/jb/biordf-demo/bams_part_names.txt that are already in neurondb etc [WITHDRAWN] [recorded in http://www.w3.org/2007/03/19-BioRDF-minutes.html#action09]
ACTION: Kei, Bill to identify URL for terms in http://hissa.nist.gov/jb/biordf-demo/bams_part_names.txt that are already in neurondb etc [recorded in http://www.w3.org/2007/03/19-BioRDF-minutes.html#action10]
<mscottm> The Guus Schreiber student that I think was involved with the MESH SKOS work is Willem van Hage http://www.few.vu.nl/~wrvhage/index_work.html
ACTION: Don see what's in brainmaps that would be of use for our scientific use case and demo [CONTINUES] [recorded in http://www.w3.org/2007/03/19-BioRDF-minutes.html#action12]
Don: just looking to see if we can use http://brainmaps.org/
ACTION: Melli will think about what image processing could be done on brainmaps images [CONTINUES] [recorded in http://www.w3.org/2007/03/19-BioRDF-minutes.html#action13]
AlanR: want MESH to PubMed
mapping
... what do MESH descriptions of pubmed terms mean?
Olivier: '*'s reprepsent main
topic in medline paper
... subheadings give facets of the main heading
... main headings identify single MESH term
Note: main medline headings identify a single MESH term
AlanR: first things is MESH term, and each '/' is a facet, and '*' someone signifies special lines
ACTION: Olivier to send (to list) pointer to MESH curation manual [recorded in http://www.w3.org/2007/03/19-BioRDF-minutes.html#action14]
ACTION: Olivier to send pointer to a pubmedID to MESHid mapping [recorded in http://www.w3.org/2007/03/19-BioRDF-minutes.html#action15]
ericP: am importing JAX data into http://www.w3.org/2007/03/entrez-gene/snapshots/
AlanR: start with accurate set of MESH terms
<mscottm> Mark van Assem.
<mscottm> http://thesauri.cs.vu.nl/
AlanR: our demo keeps us on
track
... but our real compelling demo will come from things that
people like June discover as we assemble this
Kei: in senselab, we export in
RDF and OWL
... which should we focus on?
AlanR: prefer OWL -- just a focused sublanguage of RDF
<samwaldmatthias> http://3ba.se/
<samwaldmatthias> http://3ba.se/
-> http://3ba.se/ OntoWiki
matthias: on left, see KBs
-> http://3ba.se/instance.php?model=http://3ba.se/conferences/
scribe: classes on left
... classes and individuals rendered differently
... clicking on a class gives a list of individuals
http://3ba.se/index.php?instance=AIMSA2006
scribe: can click on Edit, or
in-line in the actions box
... would like to use OntoWiki to show that RDF is not
something the scientists use in the labs and they export data,
but instead something that we can use
... OntoWiki not tailored to a particular ontology, just upload
RDFS/OWL and you can visualize the data
vipul: can you edit classes, or just instances?
matthias: can add classes, but this is not the main use case
vipul: could import all our demo data into wiki
matthias: shouldn't be an afterthought
vipul: also considering faceted
browser
... also AlanR's prefuse
<samwaldmatthias> prefuse
scott: faceted browser of
OntoWiki are handy for browsing data
... the faceted browser incorporates selected facets into a
query
... faceted browser shows thumbnails and "..." for plentiful
data for further refinement
<alanr> times up
<alanr> got to go...
matthias: yes, can SPARQL OntoWiki
<alanr> Welcome Michael!
vipul: would like scott amd matthias to recommend UI
<vipul> Gotta go...
<mscottm> Mark van Assem.
<mscottm> http://thesauri.cs.vu.nl/
UNKNOWN_SPEAKER: have a couple weeks slack and volunteering to Alzgene to OWL
Elizabeth: you have contacts at Alzforum?
AlanR: working me right now
<mscottm> Mark van Assem.
<mscottm> http://thesauri.cs.vu.nl/
-> http://thesauri.cs.vu.nl/ MESH mapping
ACTION: Kei, June, Gwen, coordinate creating list of symptoms, stages, life course of the disease [CONTINUES] [recorded in http://www.w3.org/2007/03/19-BioRDF-minutes.html#action16]
ACTION: Kei: Brainpharm OWL should have enough to do query for known contradictions, e.g. disagreement about whether something is in a particular compartment [CONTINUES] [recorded in http://www.w3.org/2007/03/19-BioRDF-minutes.html#action17]
ACTION: Eric mapping demo using dartgrid [DROPPED] [recorded in http://www.w3.org/2007/03/19-BioRDF-minutes.html#action21]
<heuermh> For the record, +1.612.821.aaaa is me, Michael Heuer