Search Mailing List Archives


Limit search to: Subject & Body Subject Author
Sort by: Reverse Sort
Limit to: All This Week Last Week This Month Last Month
Select Date Range     through    

[bioontology-support] FW: Feedback

Trish Whetzel whetzel at stanford.edu
Thu Feb 25 14:57:07 PST 2010


Hi William, 

 

The issue with the duplicate results has now been resolved. For the moment,
the mappings for UMLS ontologies are not available. We are working on
getting a timeline for that. 

 

As a result to codify the architecture for the Annotator, the UMLS
ontologies are no longer being accessed from the UMLS database. As a result
the UMLS ontology identifier, e.g. human for NCIt, is now added to the
output rather than the CUI.   In order to get the CUI,  the conceptId can be
used in the search Web service adding the parameter to “include attributes”.
We are working on a new concept service that will accept the CUI as the
conceptId. 

 

Trish 

 

 

From: bioontology-support-bounces at lists.stanford.edu
[mailto:bioontology-support-bounces at lists.stanford.edu] On Behalf Of Lau,
William (NIH/CIT) [E]
Sent: Thursday, February 25, 2010 1:25 PM
To: 'support at bioontology.org'
Subject: [bioontology-support] FW: Feedback

 

I wasn’t sure if my earlier email was sent out correctly.

 

Thanks for your attention.

 

William

 

From: Lau, William (NIH/CIT) [E] 
Sent: Tuesday, February 23, 2010 7:48 PM
To: 'Trish Whetzel'
Subject: RE: [bioontology-support] Feedback

 

Hi Trish,

 

Thanks for your quick response. Here is a sample input:

“Cloning and characterization of two cytoplasmic dynein intermediate chain
genes in mouse and human.”

 

I am using the default parameters except for ontologyToKeepInResult, which I
put in:  “42142,39966,42331,42333,40659,42303,42332,42165,42334”

 

I am getting XML output, but for readability, I am pasting less than 5% of
the text output below. 

 

There are a few issues:

(1)    As you can see,  the first two AnnotationBean objects are exactly the
same. There are many of these in fact, which makes the output very very
large due to the redundancy, and as a result, takes a long time to process.

(2)    In the previous version, I remember the term “human” mapped to MSH,
but now it doesn’t. It seems to be true for other ontologies as well too.
Only some ontologies are working.

(3)    The localConceptId for some terms are now the term name. NCIt, for
example, the localConceptId for human is 42331/Human, instead of
42331/C14225, which is the actual identifier in the ontology. In the old
version, the UMLS CUIs are used as Ids. I can use it to translate to the
naïve Id, but now it’s not as easy.

 

Please let me know if you have solutions for these issues or bugs.

 

Thanks much,

William

 

 

ObaResultBean [ 
ResultBean [ 
        resultID = OBA_RESULT_da43
        dictionary = [id = 3, name = OBS_DICO_7281, dateCreated = 2009-06-17
16:19:47.0]
        statistics = [(MAPPING, 54) , (ISA_CLOSURE, 0) , (MGREP, 60) ]
        parameters = [longestOnly = false, wholeWordOnly = true,
filterNumber = true, withSynonyms = true, scored = true, ontologiesToExpand
= [], ontologiesToKeepInResult = [42142, 39966, 42331, 42333, 40659, 42303,
42332, 42165, 42334], isVirtualOntologyId = false, semanticTypes = [],
levelMax = 0, mappingTypes = [], stopWords = [], withDefaultStopWords =
false, isStopWordsCaseSenstive = false, text to annotate = Cloning and
characterization of two cytoplasmic dynein intermediate chain genes in mouse
and human. ]
]
        ontologies = [[NCI Thesaurus, nbAnnotation: 83, score: 5616, (42331,
09.09, 1032)], [Foundational Model of Anatomy, nbAnnotation: 23, score:
1704, (39966, 3.0, 1053)], [Gene Ontology, nbAnnotation: 8, score: 192,
(42332, 1.794, 1070)]]
        annotations = [AnnotationBean [ 
               score = 132
               concept = [localConceptId: 42331/Human, conceptId: 3183531,
localOntologyId: 42331, isTopLevel: 0, preferredName: Human, synonyms:
[Human, General, Homo sapiens], semanticTypes: [[id: 4962845, semanticType:
T999, description: NCBO BioPortal concept], [id: 4992520, semanticType:
T016, description: Human]]]
               context = [MAPPING(false), mappedConceptID =
40739/p11:birnlex_516, mappingType = Automatic, termName = Human, from = 94,
to = 98]
], AnnotationBean [ 
               score = 132
               concept = [localConceptId: 42331/Human, conceptId: 3183531,
localOntologyId: 42331, isTopLevel: 0, preferredName: Human, synonyms:
[Human, General, Homo sapiens], semanticTypes: [[id: 4962845, semanticType:
T999, description: NCBO BioPortal concept], [id: 4992520, semanticType:
T016, description: Human]]]
               context = [MAPPING(false), mappedConceptID =
40739/p11:birnlex_516, mappingType = Automatic, termName = Human, from = 94,
to = 98]
],

 

 

 

From: Trish Whetzel [mailto:whetzel at stanford.edu] 
Sent: Tuesday, February 23, 2010 6:58 PM
To: Lau, William (NIH/CIT) [E]
Cc: support at bioontology.org
Subject: Re: [bioontology-support] Feedback

 

Dear Dr. Lau, 

 

Can you provide an example of the input text, parameters, and output you are
using? 

 

Thanks,

Trish 

 

 

On Feb 23, 2010, at 3:21 PM, admin at bioontology.org wrote:

 

Name : William Lau

Email : william.lau at nih.gov

Comment:
I'm using the Annotator REST service, but there are several bugs. First, the
same terms are returned many many times. The xml output is very large even
for one sentence. From what I can tell the the data are the same, just get
repeated a lot. That slows down the performance dramatically. Also, it
doesn't seem to be able to identify MeSH concepts.

I am not sure if the observations that I see are effects of the upgrade to
version 2.3.
_______________________________________________
bioontology-support mailing list
bioontology-support at lists.stanford.edu
https://mailman.stanford.edu/mailman/listinfo/bioontology-support

 

Trish Whetzel, PhD

Outreach Coordinator

The National Center for Biomedical Ontology

Ph: 650-721-2378

whetzel at stanford.edu

http://www.bioontology.org

 

 

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.stanford.edu/pipermail/bioontology-support/attachments/20100225/6a9366e3/attachment.html>


More information about the bioontology-support mailing list