Stony Brook University<br />School of Medicine<br />8/25/2011<br />1<br />VIVO Mini-Grant: Integrating the UMLS Ontology i...
Contributors:<br />Erich Bremer*<br />JizuZhi**<br />Tammy DiPrima*<br />Ann Gardner***<br />Naresh Singh****<br />AniketD...
SUNY REACH<br />3<br />
4<br />
5<br />
6<br />
VIVO has an ontology<br />7<br />
The Semantic Web starts simple with a RDF triple<br />Subject<br />Object<br />Predicate<br />8<br />
Builds into a more complex network of interlinked URIs<br />Source data from NCIt<br />9<br />
10<br />
The CUI: Concept Unique Identifier<br />11<br />
UMLS as linked data<br />Developed tool to publish large databases into RDFS<br />Published 2011AA version of the UMLS and...
UMLS RDF in the wider world<br />13<br />
Faceted browser view<br />14<br />
UMLS CUI Alignment Tool<br />15<br />
16<br />
Algorithm for aligning free text<br />Parse free text into component words<br />Build phrases of different word length<br ...
UMLS Web Service<br />Base address: http://link.informatics.stonybrook.edu<br />Sample call: /MeaningLookup/MlServiceServl...
19<br />Alignment to the UMLS CUIs<br />
PubMed RDF Conversion<br />Started with XSLT published in 2008 by Pierre Lindenbaum<br />A prototype project linked 2010 P...
21<br />
PubMed CUI Web Service<br />Base address: http://link.informatics.stonybrook.edu<br />Sample call: /weaver/pubmed2cuis?pmi...
Linking subject areas to publications<br />23<br />
Data  facts<br />UMLS RDF (2011AA release; English language; SRL=0)<br />Number of triples: 110,415,427<br />Number of dif...
Links<br />SPARQL endpoint:<br />http://link.informatics.stonybrook.edu/sparql/<br />CUI alignment tool:<br />http://link....
Acknowledgements<br />Supported through: <br />VIVO: Enabling National Networking of Scientists NIH U24 RR029822 <br />Ori...
Upcoming SlideShare
Loading in …5
×

VIVO Mini-Grant: Integrating the UMLS Ontology into VIVO for Linking Biomedical Scientists

1,511 views
1,419 views

Published on

VIVO 2011 presentation on mini-grant award that Stony Brook won from the NIH VIVO consortium.

Published in: Sports, Education
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
1,511
On SlideShare
0
From Embeds
0
Number of Embeds
1
Actions
Shares
0
Downloads
9
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

VIVO Mini-Grant: Integrating the UMLS Ontology into VIVO for Linking Biomedical Scientists

  1. 1. Stony Brook University<br />School of Medicine<br />8/25/2011<br />1<br />VIVO Mini-Grant: Integrating the UMLS Ontology into VIVO for Linking Biomedical Scientists<br />Moises Eisenberg* and Janos Hajagos*<br />
  2. 2. Contributors:<br />Erich Bremer*<br />JizuZhi**<br />Tammy DiPrima*<br />Ann Gardner***<br />Naresh Singh****<br />AniketDivecha****<br />2<br />Dept. of Medical Informatics* | OSA**| SOM***|Dept. of Computer Science****<br />
  3. 3. SUNY REACH<br />3<br />
  4. 4. 4<br />
  5. 5. 5<br />
  6. 6. 6<br />
  7. 7. VIVO has an ontology<br />7<br />
  8. 8. The Semantic Web starts simple with a RDF triple<br />Subject<br />Object<br />Predicate<br />8<br />
  9. 9. Builds into a more complex network of interlinked URIs<br />Source data from NCIt<br />9<br />
  10. 10. 10<br />
  11. 11. The CUI: Concept Unique Identifier<br />11<br />
  12. 12. UMLS as linked data<br />Developed tool to publish large databases into RDFS<br />Published 2011AA version of the UMLS and corresponding RxNorm Release<br />Public available sources (SRL=0)<br />RxNorm is linked to DrugBank<br />Includes attributes, relationships, and semantic types from the UMLS<br />12<br />
  13. 13. UMLS RDF in the wider world<br />13<br />
  14. 14. Faceted browser view<br />14<br />
  15. 15. UMLS CUI Alignment Tool<br />15<br />
  16. 16. 16<br />
  17. 17. Algorithm for aligning free text<br />Parse free text into component words<br />Build phrases of different word length<br />Query UMLS if phrase exists<br />Sort in descending order of number of words<br />Tie criteria based on number of occurrences in different source vocabularies<br />Most widely used gets a higher rank<br />17<br />
  18. 18. UMLS Web Service<br />Base address: http://link.informatics.stonybrook.edu<br />Sample call: /MeaningLookup/MlServiceServlet?textToProcess=Pediatric%20HIV&format=json<br />Response format:<br />JSON, N-triples, RDF/XML<br />Response content:<br />Best choices and all choices for matching CUIs<br />18<br />
  19. 19. 19<br />Alignment to the UMLS CUIs<br />
  20. 20. PubMed RDF Conversion<br />Started with XSLT published in 2008 by Pierre Lindenbaum<br />A prototype project linked 2010 PubMed to the internal Health Sciences Library MARC holdings data (>800,000,000 triples)<br />Allowed linked data search joining article data with holdings data<br />PubMed XSLT updated to 2011 schema with MeSH aligned to the UMLS CUIs <br />Current translation generated 1,973,880,813 triples<br />20<br />
  21. 21. 21<br />
  22. 22. PubMed CUI Web Service<br />Base address: http://link.informatics.stonybrook.edu<br />Sample call: /weaver/pubmed2cuis?pmid=17952453<br />Response format: JSON<br />Response content: UMLS CUIs with labels<br />22<br />
  23. 23. Linking subject areas to publications<br />23<br />
  24. 24. Data facts<br />UMLS RDF (2011AA release; English language; SRL=0)<br />Number of triples: 110,415,427<br />Number of different sources: 46<br />Number of CUIs: 2,404,344<br />Number of AUIs: 3,594,372<br /><ul><li>REACH VIVO (Data extracted 8/23/2011)</li></ul>Number of people: 684<br />Number of triples: 448,112<br />UMLS alignment of subject areas<br />Number of subject areas: 425<br />Number of UMLS CUIs generated: 899<br />Number of distinct UMLS CUIs: 604<br />PubMed alignment to REACH<br />Number of UMLS CUIs generated: 192,450<br />Number of distinct UMLS CUIs: 11,039<br />Articles with no MeSH is 1,293 out of 15,975<br />24<br />
  25. 25. Links<br />SPARQL endpoint:<br />http://link.informatics.stonybrook.edu/sparql/<br />CUI alignment tool:<br />http://link.informatics.stonybrook.edu/MeaningLookup/<br />Points to start browsing linked data:<br />http://link.informatics.stonybrook.edu/umls/<br />http://link.informatics.stonybrook.edu/umls/SAB<br />Open source code developed at SBU:<br />http://code.google.com/p/py-triple-simple/<br />Native Python RDF utility<br />http://code.google.com/p/sbu-mi-vivo-tools/<br />Automated dumping of VIVO sites RDF and alignment to UMLS and PubMed<br />http://code.google.com/p/spyder-web/<br />Faceted browser and lightweight web service for parameterized SPARQL queries<br />25<br />
  26. 26. Acknowledgements<br />Supported through: <br />VIVO: Enabling National Networking of Scientists NIH U24 RR029822 <br />Original interactive CUI alignment tool created by JakubPezacki (SBU Class of 2010)<br />26<br />

×