Tim Clark Harvard Medical School &

Massachusetts General Hospital

Pistoia Alliance Conference April 12, 2011

Copyright 2011 Massachusetts General Hospital. All rights reserved.

  Information  sharing  and  integration  requirements  for  curing  complex  disorders.  

 Web  3.0  and  semantic  metadata.    Integrating  ontologies,  documents,  data.  

  Annotation  Ontology  &  Annotation  Framework.    Applications  

  SESL,  Hypothesis  Mgmt,  Nanopublications    Open  Enterprise  Semantic  Model  

  Conclusion  

  Yearly  mortality  (U.S.)              =  642,00  people  

  Yearly  costs  (U.S.)                          =  $676  B  /  4.7%  GDP  

  Prevalence  =  5.3  M  +  76  M  +  14.4  M      

   =    95.7  M  people    

create hypothesis�

design experiment �

run experiment � collect data�

interpret data�

share interpretations�

synthesize knowledge�

   

MCI progressors non progressors

PET imaging of PIB (radiolabelled compound binds amyloid beta A4 protein)

MRI imaging of brain structure showing loss of hippocampal volume

Brain. 2010 Nov;133(Pt 11):3336-3348.

= 218 subjects +

dopaminergic pathway�

α-synuclein, β-amlyoid�

α-synuclein, Tau �

chr 16p11.2 CNV �

chr 16p11.2 CNV �

CRF, glutaminergic system, dopamine, amygdala …�

Alzheimer� Disease�

Parkinson’s �Disease�

Schizophrenia�

Autism�

Bipolar Disorder� Drug �Addiction �

Huntington’s �Disease�

ALS�

Depression �

SIRT2 �

1. We want to organize all the known facts in neurobiology so we can mash them up.

2. There are no “facts” in neurobiology, except uninteresting ones.

3. All we have, are

assertions supported by evidence, of varying quality.

1667 2010

Printing Press Web

We scientists do not attend professional meetings to present our findings ex cathedra, but in order to argue.

John Polanyi, FRS, Nobel Laureate University of Manchester

 Social  Web  (Web  2.0,  read/write)  

 Shared  annotation  with  controlled  terminology  systems  (Sem  Web)  

 

+

  Information  sharing  within  communities  or  tasks  via  Social  Web  (Web  2.0),  wikis  and  forums  

  Information  “permeability”  across  pharma  R&D  projects  /  domains  /  pipeline  stages  via  shared  metadata  (semantic  annotation)  

 Web  3.0  improves    cross-­‐domain  Signal  to  Noise,  institutional  memory  &  data  “findability”  

Genes

Proteins

Biological Processes

Chemical Compounds

Antibodies

Cells

Brain anatomy

…

  Annotation  Ontology  (AO)  is  a  domain-­‐independent  Web  ontology.    Links  document  fragments  to  ontology    terms.  

 Metadata  separate  from  annotated  documents.      SWAN  AF  manages  document  annotation.  

  Interfaces  to  textmining  svcs  &  supports  curation.    Collaborating  with  

  NCBO,  UCSD,  Elsevier,  USC,  Manchester,  EMBL,  Colorado,  EBI,  etc…  

Text

Shared metadata

2) Automatic annotation

Dr.

Pao

lo C

icca

rese

– O

ct 8

, 201

0

Dr.

Pao

lo C

icca

rese

– O

ct 8

, 201

0

  Semantics  on  documents  (SESL)        Vocabulary  standards  &  terminology  development    

  Document  &  data  management    Collaboratories  &  web  communities    Hypothesis  management  (SWAN)    Nanopublications  (OpenPHACTS)  

  Model the thinking behind your research   Database it, web-ify it, RDF-ize it, share it   Link the Models / Hypotheses to

  Claims / Interpretations   Evidence (publications, experiments, data)   Supporting and contradictory claims from others   Evidence for these other claims

  Web 3.0: share, compare and discuss   Manage knowledge while creating it

  Can be public, private, or semi-private

SWAN Ontology: Model of Research Statements

SWAN Ontology: Provenance of Research Statements

Dr.

Pao

lo C

icca

rese

– O

ct 8

, 201

0

Dr.

Pao

lo C

icca

rese

– O

ct 8

, 201

0

Dr.

Pao

lo C

icca

rese

– O

ct 8

, 201

0

Cognitive    Deficits  

(S)  

BACE1  (O)  

Relate  to  (p)  

provenance  context  

With thanks to Barend Mons and Paul Groth…

Mons / Groth model of a nanopublication

swande:Claim  <http://tinyurl.com/4h2am3a>  

Intramembranous  Aβ  behaves  as  chaperones  of  other  membrane  proteins  

rdf:type

dct:title

G1

<http://example.info/person/1>  pav:authoredBy

Vincent  Marchesi  

foaf:name

foaf:Person  

rdf:type

pav: http://purl.org/pav/provenance/2.0/ foaf: http://xmlns.com/foaf/0.1/

G2

With thanks to Paolo Ciccarese

swande:Claim  <http://tinyurl.com/4h2am3a>  

Intramembranous  Aβ  behaves  as  chaperones  of  other  membrane  proteins  

rdf:type

dct:title

G1 <http://example.info/person/1>  pav:authoredBy

G2 <http://example.info/person/0>  pav:curatedBy

G4

Gwen  Wong  

foaf:name

foaf:Person  

rdf:type

With thanks to Paolo Ciccarese

swande:Claim  <http://tinyurl.com/4h2am3a>  

Intramembranous  Aβ  behaves  as  chaperones  of  other  membrane  proteins  

rdf:type

dct:title

G1

<http://example.info/person/1>  pav:contributedBy

<http://example.info/citation/1>  

swanrel:referencesAsSupportiveEvidence

G5

G6 With thanks to Paolo Ciccarese

G8

<http://example.info/alzswan:statement_f3556dcfc331d9b9af9d5c0cfc570ba6_event_1>  

<http://bio2rdf.org/go:0051087>  

rdf:type

Event  of  type  GO  "chaperone  binding"  

rdfs:label

<prefix:actor_1>  

<prefix:target_1>  

<prefix:location_1>  

<http://bio2rdf.org/chebi:53002>  

<http://bio2rdf.org/mesh:D008565>  

<http://bio2rdf.org/go:0005886>  

rdf:type

rdf:type

rdf:type

rdfs:label “Beta amyloid”

rdfs:label “Membrane protein”

rdfs:label “Plasma membrane”

With thanks to Nigam Shah & Paolo Ciccarese

Hyque  triples  

G8

<http://example.info/person/2>  pav:contributedBy

Nigam  Shah  

foaf:name

foaf:Person  

rdf:type G9

swande:Claim  <http://tinyurl.com/4h2am3a>  

Intramembranous  Aβ  behaves  as  chaperones  of  other  membrane  proteins  

rdf:type

dct:title

G1

Hyque  triples  G8

swanrel:derivedFrom

  The  target  hypothesis  will  be  linked  to:    Pathway  &  target  relation  to  disease,    Target  selection  criteria,      Validation  assays  and  criteria,    Experiment  (assay)  provenance,    Experimental  data  and  computations,    Scientist  remarks,  findings  and  discussion.    

  Start  as  a  relatively  simple  model  and  extend  

  Hypotheses  of  therapeutic  action  for  compounds  and  scaffolds,  linked  to    Hypothesis  /  results  for  individual  assays,    Experiment  (assay)  provenance,    Experimental  data,    Group  annotation,      Internal  databases  etc.  

  Start  as  a  relatively  simple  model  and  extend  

Information ecosystem

  Curing  complex  medical  disorders  goes  hand  in  hand  with  next-­‐gen  biomedical  communications  

 Web  3.0  provides  the  technology  framework    Semantic  annotation,  hypothesis  management,  nanopubs:  tools  for  next-­‐gen  biomed  comms  .    

  Requires  /  enables  international  collaborations  of  biomedical  researchers  and  informaticians.  

  Open  enterprise  model  with  semantic  metadata.  

  People    Paolo  Ciccarese  (Harvard)   Maryann  Martone  (UCSD)    Anita  DeWaard  &  Tony  Scerri  (Elsevier)    Adam  West  &  Ernst  Dow  (Eli  Lilly  &  Co)    Carole  Goble  &  Sean  Bechhofer  (U  Manchester)    Karen  Verspoor  &  Larry  Hunter  (U  Colorado)    Gully  Burns  &  Cartik  Ramakrishnan  (USC)    David  Newman  (U  Southampton)    Nigam  Shah  (Stanford  /  NCBO)    Paul  Groth  &  Barend  Mons  (VU  Amsterdam)  

  Funding:  Elsevier,  NIH,  Eli  Lilly,  &  EMD  Serono