NITRC Resource Ontology Discussion Group Forum: open-discussion

ISBI terms

David Kennedy — Tue, 20 Nov 2012 21:20:44 GMT

The table below shows the keywords for ISBI 2013

These keywords are used in assigning papers

[b]Category[/b]
Keyword
[b]Methods[/b]
Active contours and deformable models
Atlas-based methods
Classification methods
Color and multispectral processing
Compressive sampling and sensing
Deconvolution methods
Differential geometry
Fuzzy methods
Image coding and compression
Image fusion
Image quality assessment
Image registration
Image segmentation branch-cut
Image segmentation level-sets
Image segmentation other
Image sequence processing
Image storage and retrieval
Inverse problem solving
Knowledge modeling
Learning-based methods
Mathematical morphology
Mesh construction and analysis
Modeling of image formation
Motion detection and tracking
Multiresolution methods
Optimization methods
Partial differential equations and variational methods
Probabilistic statistical and Monte-Carlo methods
Restoration and enhancement
Sampling and interpolation
Sparse representations
Shape modeling and analysis
Super-resolution methods
Tensor and vector field analysis
Texture segmentation and analysis
Tomographic reconstruction
Visualization and display
[b]Modalities[/b]
Confocal microscopy
Diffuse optical imaging
Diffusion magnetic resonance imaging
Elastography imaging
Electrical impedance tomography
Electrocardiography
Electroencephalography and magnetoencephalography
Electron microscopy
Fluorescence microscopy
Functional magnetic resonance imaging
Infrared imaging
Integrative Multiscale Imaging
Intravital microscopy
Light microscopy
Magnetic resonance imaging
Magnetic resonance spectroscopy
Microarrays
Multi-photon microscopy
Multispectral and hyperspectral imaging
Optical coherence tomography
Positron emission tomography
Radiography and fluoroscopy
Single photon emission computed tomography
Small animal imaging
Super-resolution microscopy
Ultrasound imaging
X-ray computed tomography
[b]Applications[/b]
Anatomical imaging
Biometry and atlases
Brain imaging
Cancer imaging
Cardiac and vascular imaging
Computational anatomy
Computer aided detection and diagnosis
Content-based retrieval
Dynamic imaging
Epidemiology
Functional imaging
Gene expression mapping
Image guided surgery
Image guided therapy
In vivo cellular and molecular imaging
Interventional imaging
Lung imaging
Molecular and cellular screening
Musculoskeletal imaging
Neural engineering
Physiological modeling
Radiation therapy planning and treatment
Single molecule detection
Strain imaging
Therapy efficacy assessment
Tractography

terms to support Imaging Genomics

David Kennedy — Tue, 13 Nov 2012 11:15:14 GMT

Following Li Shen's description, I added the terms as highlighted in the attached picture. From his list, Visualization, data and database, and multivariate analysis all had existing entries, so were not added.

The domain keyword 'Imaging Genomics' was added, and activated for front-page searching, and already has 7 tagged entries as of this writing.

Resources

David Kennedy — Wed, 28 Apr 2010 9:38:50 GMT

A Word 2007 add-in that enables the annotation of Word documents based on terms that appear in Ontologies
http://ucsdbiolit.codeplex.com/

Calais is a rapidly growing toolkit of capabilities that allow you to readily incorporate state-of-the-art semantic functionality within your blog, content management system, website or application -
http://www.opencalais.com/

BIRNLex mailing list

David Kennedy — Mon, 03 Mar 2008 22:34:57 GMT

[Derived from Bill Bug]

Hi All,

I'm writing to encourage you all to subscribe to the BIRNLex mailing list:

http://portal.nbirn.net/mailman/listinfo/birnlex-dev

I know what you're thinking - "Oh yeah - I've got more cycles I can harvest to invest in my 1394th mailing list subscription." Well, I promise to do all I can to make this worth your while. We'll try to keep the posts short and the threads brisk & utilitarian. Those who know my typical emails may doubt that is possible, but believe me, I've burned enough time on various mailing lists to know, we - the BIRN Ontology Task Force - can't afford not to have a list and can't afford to let it drift off task.

There is considerable content added to BIRNLex in the last 6 months. There is also considerable additional content soon to be added - ontology content related to nerve cell types and molecules that will soon be incorporated into BIRNLex from the Neuroscience Information Framework (NIF) project - in addition to many quite significant application driven additions coming throughout BIRN. We are also beginning to incorporate more expressive relations which will assist in providing mereological reasoning capabilities for neuroanatomy and providing species-specific constraints for a variety of entities all within OWL/SWRL/RDF and represented using OBO-Relations, PATO, OBI, BFO and as much as is practical according to OBO Foundry principles. Finally, in the coming year there will be a variety of nervous system disease related phenotype representation projects that will either be using BIRNLex or will be directly adding to BIRNLex.

In other words, now's the time to make certain there is a consistent and clear audit trail for what is being done or is pending in relation to BIRNLex, so we'd like to start leaning more heavily on this mailing list. As needed, content will be culled from here and added to the BIRNLex Wiki. There is already a great deal of BIRNLex-related Wiki content, and this will also be undergoing an update and re-organization, so the list will make it possible for us to disseminate to all interested parties, when useful additions are made to the Wiki.

Finally, I'd like to encourage all users - or potential users - to post requests, critiques, or comments, so that I can make certain your feedback gets channeled into concrete progress on the ontology. I would suggest we set up a term tracker like activity, but I believe for now, it will suffice to start conversing more regularly. We can add such a capability later, should that be necessary. I would stress, however, that BIRNLex, as an application ontology, is focussed as much as possible on re-using more foundational, generic, or domain ontologies - or casting domains in OWL, when they are not otherwise covered elsewhere in a form that promotes re-use within OWL according to the OBO Foundry practices that undergird our biomedical ontology development. Even in the latter case, when BIRNLex constructs significant domain ontologies, it's typically doing this by converting existing terminologies into the required OWL form and extending or enhancing the expressivity, as applications require. None of this activity would be easily followed right now in a term tracker, though, again, that may change later.

So - thank you for listening, and please start - and keep - those requests and comments coming.

Cheers,
Bill

William Bug, M.S., M.Phil. email: wbug@ncmir.ucsd.edu
Ontological Engineer work: (610) 457-0443
Biomedical Informatics Research Network (BIRN)
and
National Center for Microscopy &
Imaging Research (NCMIR)
Dept. of Neuroscience, School of Medicine
University of California, San Diego
9500 Gilman Drive
La Jolla, CA 92093

RE: Tools for working with ontologies

Bill Bug — Sat, 22 Sep 2007 15:38:52 GMT

I forgot to mention there is one more ontology community-based Wiki/Forum that has been around for quite some time and does reference a lot of info. That is the Ontolog Wiki started quite a while ago by Peter Yim, now maintained by Peter, Kurt Conrad and Leo Oberst (the latter of MITRE). I find this Wiki quite dense, making it both hard to see the forest for the trees and difficult to find a particular set of trees.

Still - there is a wealth of information here, and an ongoing set of conference calls associate with this forum.

There is also the OntoWorld Wiki (http://ontoworld.org/wiki/Main_Page) associated with the very important annual International Semantic Web Conference (ISWC).

In both cases, these sites both include a breath of coverage that is in no way limited to the biomedical domain, while also tending to be focussed specifically on the project undertaken by the participants of those forums, as opposed to also casting an eye outward to keep track of all the other relevant projects and tools.

Cheers,
Bill

RE: Tools for working with ontologies

Bill Bug — Sat, 22 Sep 2007 15:16:09 GMT

Hi David,

I would say pieces of what's here you'll find on various sites, but you aren't likely to find it all gathered together like this - at least, I haven't.

It would be wonderful to have this on a Wiki, where others could contribute. In particular, I know there are other journals and meetings that have a specific biomedical ontology focus. AMIA, for instance, has had a very significant ontology contingent over the last decade. Mark Musen SMI group and the UK groups that have worked on clinical ontologies (e.g., GALEN) frequently participate in the AMIA meetings.

There are also a select few sites around the web where you'll find some of these resources aggregated together - along with others.

So - yes - I think that's an excellent suggestion. If you feel the NITRC Wiki - despite NITRC's neuroimaging focus - would be an appropriate place for this, that works for me. Possibly a more appropriate place for such a Wiki to reside would be either the NCBO Resources page (http://www.bioontology.org/resources.html) or the NCBC Scientific Ontology WG Wiki (http://na-mic.org/Wiki/index.php/SDIWG:_NCBC_Scientific_Ontologies).

The NCBC SOWG page is currently focussed on provided a listing of "gold standard" NCBC-approved ontologies to promote data sharing and systems interoperability across NCBCs.

Right now, NCBO almost exclusively focuses on tools they are being funded to create, distribute, and support (http://www.bioontology.org/tools.html), but I believe it would be helpful for them (the experts) to host a list with a wider scope than that, since there is so much more out there and in fairly widespread use.

Having said that, you were right to pick out Protege. It is THE most widespread ontology curation tool in use today. It's quite a remarkable feat they (in collaboration with the U. Manchester OWL group) were able to take a tool based on the highly complex Protege-Frames (OKBC) formalism and adapt it to OWL. Unfortunately, that did not come without a price, and as we've all learned the hard way, there are some "gotchas" one needs to be aware of when working with Protege-OWL. Protege v4 (built on OWL v1.1 from the ground up) is a completely re-engineered tool that eliminates many of these issues, but its still relatively new (about 1 year old) and doesn't have all the functionality of its bigger brother (Protege v3 - currently at v3.3.1).

One of the reasons I assembled this list is because I tend to use > 1 tool at a time when curating/developing ontologies. Frequently, I'll be running ontology generation/processing code in Netbeans coding against the Jena libraries, while also checking the results both in Protege and SWOOP, using an XML editor such as oXygen to check for certain anomologies in the RDF/XML representation, and running QA tests on the structure using the Pellet DIG reasoner.

Cheers,
Bill

RE: Tools for working with ontologies

David Kennedy — Sat, 22 Sep 2007 14:32:53 GMT

Excellent listing, Bill! Should this content also migrate to a publicly editable wiki (so that the content can be dynamically updated and commented upon as it evolves)? Is it already listed anywhere else in such a fashion? If not, I would recommend a NITRC wiki. If we choose to cross-list this in a NITRC wiki, tho, I'd recommend waiting a little as I expect (hope!) that the phpwiki-based system will be upgraded shortly to a mediawiki-based product soon.

RE: Tools for working with ontologies

Bill Bug — Fri, 21 Sep 2007 23:50:39 GMT

Here is a list of biomedical ontology-related repositories and tools I assembled recently for a BIRN information page. It's far from exhaustive - especially given how much ontology-related work & tools takes place outside the biomedical domain - but it's pretty thorough regarding those ontologies and tools being used by bioinformaticists.

Biomedical ontology aggregation sites:
OBO Foundry:
http://www.obofoundry.org/
BioPortal
http://www.bioontology.org/ncbo/faces/index.xhtml
EBI Ontology Lookup Service:
http://www.ebi.ac.uk/ontology-lookup/
Berkeley Gene Ontology Download Matrix
http://www.berkeleybop.org/ontologies/
NCBC Scientific Ontology Working Group
http://www.berkeleybop.org/sowg/table.cgi

Though there is considerable overlap across these sites, each provides what I've found to be invaluable non-overlapping content and/or functionality.

Specific Ontologies:
Basic Formal Ontology (BFO)
http://www.ifomis.uni-saarland.de/bfo/home

OBO Relations Ontology (OBO-RO)
main:
http://obofoundry.org/ro/
OBO-RO-to-BFO bridge:
http://www.obofoundry.org/cgi-bin/detail.cgi?id=ro_bfo_bridge

OBO Phenotypic Quality Ontology (PATO)
main:
http://www.obofoundry.org/cgi-bin/detail.cgi?id=quality
PATO-to-BFO bridge:
http://obofoundry.org/cgi-bin/detail.cgi?id=quality_bfo_bridge

Ontology of Biomedical Investigation
http://www.obofoundry.org/cgi-bin/detail.cgi?id=obi

Ontology Tools:
Editing/Browsing/Searching:
Protege-OWL
http://protege.stanford.edu/overview/protege-owl.html
Protege-OWL v3.3 (OWL v1.0)
http://protege.stanford.edu/download/registered.html#p33
Protege-OWL v4+ (OWL v1.1+)
http://protege.stanford.edu/download/registered.html#p4
OBO-Edit
http://oboedit.org/
SWOOP
http://code.google.com/p/swoop/
GrOWL
http://ecoinformatics.uvm.edu/technologies/growl-knowledge-modeler.html

Data Annotation with ontologies
Phenote
http://www.phenote.org/

Semantic Search Engines
Swoogle
http://swoogle.umbc.edu/

Community Semantic Web and Ontology Specifications
Resource Description Format (RDF)
http://www.w3.org/TR/rdf-primer/
Web Ontology Language (OWL)
v 1.0:
http://www.w3.org/TR/2004/REC-owl-features-20040210/
v 1.1:
http://www.webont.org/owl/1.1/

Descriptive Logic DIG Reasoners ( reason against OWL ontologies)
Pellet
http://pellet.owldl.com/
FaCT++
http://owl.man.ac.uk/factplusplus/
Racer Pro
http://www.racer-systems.com/

RDF and/or OWL Query Frameworks
SPARQL
http://www.w3.org/TR/rdf-sparql-query/
NOTE: there are a variety of systems supporting queries against RDF or OWL data, in addition to building your own means to do so with the libraries listed below. A few such systems are:

Sesame:
http://sourceforge.net/projects/sesame/
(and the related Simile Project)
http://simile.mit.edu/
KAON:
http://kaon.semanticweb.org/
BOCA:
http://ibm-slrp.sourceforge.net/2006/11/20/boca-the-rdf-repository-component-of-the-ibm-semantic-layered-research-platform/
Joseki:
http://www.joseki.org/
Kawari:
http://kowari.org/
Oracle+RDF
http://www.oracle.com/technology/tech/semantic_technologies/index.html

Programming Environments/Libraries
RDF
Jena
http://jena.sourceforge.net/
Redland
http://librdf.org/
RDFLib
http://rdflib.net/store/
IBM Integrated Ontology Development Toolkit (IODT)
http://www.alphaworks.ibm.com/tech/semanticstk
OWL
OWL API
http://owl.man.ac.uk/api.shtml
Pellet
http://pellet.owldl.com/

Academic Journals:
Journal of Web Semantics:
http://www.elsevier.com/wps/find/journaldescription.cws_home/671322/description
Applied Ontology
http://www.iospress.nl/loadtop/load.php?isbn=15705838

Meetings:
ISMB Bioontology Satellite Meeting
http://bio-ontologies.man.ac.uk/
OWLED
http://owl-workshop.man.ac.uk/
Biocurator.org
http://www.biocurator.org/

Tools for working with ontologies

David Kennedy — Fri, 21 Sep 2007 22:19:22 GMT

Protege seems to be the most ubiquitous one. It can be found at: http://protege.stanford.edu/

Tool/Resource Function Keywords

David Kennedy — Fri, 14 Sep 2007 17:16:07 GMT

software tools function keywords. These keyword selections are specific for the software class of resource, and are not necessarily applicable to the Data or Dissemination classes of resource.

The NCBC ontology and NITRC/IATR have a long list of software function keywords.

One issue for me is how these are represented in the NCBC Ontology (public protege) file. The mapping between this 'ontology' and the minimal tool/resource description is not clear to me (or perhaps I not sure how to read the meaning of the protege file...).

According to the 'Collaborative+SoftwareOntology', an Atomic (software!) class, we need to describe: Purpose, DataTypology, Input, Output, AssociatedURL, BiologicalConcept, AreaOfApplication, Algorithm, Document, SoftwareFunction, SoftwareType and Integration and Interoperability Tools. Only some of these are from NCBC minimal (URL, description = purpose, keywords = SoftwareFunction, Type of resource is implicit from the class atomic) and it's not clear where the other minimal requirement go (name, stage, license, Organization).

So, apart from the protege workings, presumably, if there are things missing, they can be added, and extra things can be present, and they are available for greater than minimal compliance, and can be consistent with the observation by Ivo earlier that resources like CCB and NITRC may indeed capture more than minimal information.

So, the next topic is softwarefunction keywords. Lets start from the higher levels, particularly with respect to image processing types of tools (a given tool may have one, or many of these function keywords): (+ means some sort of high-level NCBC -> NITRC mapping, see below)

* Computational Geometry +
* Genomic & Phenotypic Analysis
* Image Processing
o Atlas Generation +
o Cortical Modeling +
o EDA ??? What's this?
o Pre-Processing +
+ Data Transforms +
# Spectral Transforms +
* Fourier Transform +
* Wavelet Transform +
+ Filtering +
# Skull Stripping
* Inhomogeneity Correction
o Registration +
o Segmentation +

* Interaction Modeling

* Natural Language Processing

* Numerical Methods
o ComputationalGeometry +
o LinearAlgebraTools
o MonteCarloSimulation
o NumericalIntegrators
o Optimizers
o PDESolvers
o RootFinders

* Protein Modeling and Classification

* Software Engineering and Development Tool
o Cross-LanguageWrapping
o Cross-Platform Tools
o Document Generation
o Integration
o Ontology Development and Management
o Source Control
o Testing Tools

* Statistical Analysis +
o Statistical Package

* Visualization +
o Clinical Charts e.g., Demographics
o Graph Viewers
+ Hyperbolic Graphs
# Hierarchical Trees
o Imaging
+ Cross-Sectional Viewers
# Manifold Viewers 2D, 3D, 4D, ND
+ Molecular Structure Viewers
o Sequences

NITRC High-level software function descriptors:

* Atlas => NCBC: Image Processing: Atlas Generation
* Database => no mapping, perhaps because database would be of the Data Resource class of entry
* Information Theory => no mapping?
* Modeling => NCBC: Interaction Modeling?, NCBC: Protein Modeling & Classification?, NCBC: Image Processing: Cortical Modeling
* Quantification => no mapping?
* Surface Analysis => NCBC: Computational Geometry, NCBC: Numerical Methods: Computational Geometry
* Tensor Metrics => no mapping?
* Segmentation => NCBC: Image Processing : Segmentation
* Shape Analysis => no mapping?
* Image Reconstruction => NCBC: Image Processing: pre-processing
* Spatial Transformations => NCBC: Image Processing: pre-processing
o Frequency Domain
o Intensity Operations
o Quality Metrics
o Registration => NCBC: Image Processing: Registration
o Resampling
o Spatial Convolution - Deconvolution
o Warping
o Wavelet Transformation
* Statistical Operations => NCBC: Statistical Analysis
* Temporal Transformations => NCBC: Image Processing: pre-processing: Spectral transformations
* Time Domain Analysis => NCBC: Image Processing: pre-processing: Spectral transformations
* Tractography => no mapping?
* Visualization => NCBC: Visualization
* Workflow => no mapping?

So, in summary, most of the imaging related NCBC functionality is in the NITRC hierarchy as well; lots of the 'other NCBC' functionality is not also represented in NITRC (proteins, engineering, interaction, NLP, etc.). Some of the NITRC entries are not included in NCBC (tensor/tractography, workflow, shape, etc.). he things that do match can use some further discussion (NCBC uses pre-processing at a level over some functions that I'm not sure is necessary, NITRC includes Temporal Transformations and Time domain analysis, which are probably synonymous... etc.), but I contend that a 'small' number of folks can still hash out a reasonable position on these issues.