OBO vs. SNOMED CT domain coverage

classic Classic list List threaded Threaded
7 messages Options
Reply | Threaded
Open this post in threaded view
|

OBO vs. SNOMED CT domain coverage

Kristian Nicolaisen

Hi,

 

I am currently working on a report on the transition to ontology based terminologies in healthcare in Norway, in that regard I am writing about BFO in general and the OBO Foundry in particular. I have a question I hope you can help me with:

·         Are there any comparison between SNOMED CT and the OBO ontologies in regards to coverage in biomedical domains? A sort of map to see what domains SNOMED covers and what domains OBO ontologies cover, and where there are overlaps?

 

 

Sincerely,

 

Kristian Malm-Nicolaisen

Adviser/Project manager

Norwegian Centre for E-health Research

Tel: +47 950 59 001

www.ehealthresearch.no/en/

NSE en

 


______________________________________________
Tenk miljø - ikke skriv ut denne om det ikke er absolutt nødvendig

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Obo-discuss mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/obo-discuss
Reply | Threaded
Open this post in threaded view
|

Re: OBO vs. SNOMED CT domain coverage

Peter Robinson

Hi Kristian,

here is a comparison of the Human Phenotype Ontology with SNOMED and other UMLS terminologies:

https://lhncbc.nlm.nih.gov/system/files/pub8937.pdf

and here is some background information:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5210535/

hth, Peter


Peter Robinson

Professor of Computational Biology

The Jackson Laboratory for Genomic Medicine

10 Discovery Drive

Farmington, CT 06032

860.837.2095 t | 860.990.3130 m

[hidden email]

www.jax.org

Robinson lab: https://robinsongroup.github.io/

The Jackson Laboratory: Leading the search for tomorrow's cures


From: Kristian Nicolaisen <[hidden email]>
Sent: Thursday, July 20, 2017 3:45 AM
To: [hidden email]
Subject: [Obo-discuss] OBO vs. SNOMED CT domain coverage
 

Hi,

 

I am currently working on a report on the transition to ontology based terminologies in healthcare in Norway, in that regard I am writing about BFO in general and the OBO Foundry in particular. I have a question I hope you can help me with:

·         Are there any comparison between SNOMED CT and the OBO ontologies in regards to coverage in biomedical domains? A sort of map to see what domains SNOMED covers and what domains OBO ontologies cover, and where there are overlaps?

 

 

Sincerely,

 

Kristian Malm-Nicolaisen

Adviser/Project manager

Norwegian Centre for E-health Research

Tel: +47 950 59 001

www.ehealthresearch.no/en/

NSE en

 


______________________________________________
Tenk miljø - ikke skriv ut denne om det ikke er absolutt nødvendig
---

The information in this email, including attachments, may be confidential and is intended solely for the addressee(s). If you believe you received this email by mistake, please notify the sender by return email as soon as possible.
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Obo-discuss mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/obo-discuss
Reply | Threaded
Open this post in threaded view
|

Re: OBO vs. SNOMED CT domain coverage

Elvira Mitraka
Hello Kristian,

I will point you towards the EBI's Ontology Xref Service (OxO) http://www.ebi.ac.uk/spot/oxo/index
It is still in beta, but so far I didn't have any issues with functionality. 
Simply select SNOMED CT or any of the ontologies as the datasource on the dropdown box above the circular graph and you can see a list of ontologies and how many terms are mapped to it. Select one ontology and you will get the full  list of mappings.

Best,
Elvira

On Thu, Jul 20, 2017 at 9:38 AM, Peter Robinson <[hidden email]> wrote:

Hi Kristian,

here is a comparison of the Human Phenotype Ontology with SNOMED and other UMLS terminologies:

https://lhncbc.nlm.nih.gov/system/files/pub8937.pdf

and here is some background information:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5210535/

hth, Peter


Peter Robinson

Professor of Computational Biology

The Jackson Laboratory for Genomic Medicine

10 Discovery Drive

Farmington, CT 06032

<a href="tel:(860)%20837-2095" value="+18608372095" target="_blank">860.837.2095 t | <a href="tel:(860)%20990-3130" value="+18609903130" target="_blank">860.990.3130 m

[hidden email]

www.jax.org

Robinson lab: https://robinsongroup.github.io/

The Jackson Laboratory: Leading the search for tomorrow's cures


From: Kristian Nicolaisen <[hidden email]>
Sent: Thursday, July 20, 2017 3:45 AM
To: [hidden email]
Subject: [Obo-discuss] OBO vs. SNOMED CT domain coverage
 

Hi,

 

I am currently working on a report on the transition to ontology based terminologies in healthcare in Norway, in that regard I am writing about BFO in general and the OBO Foundry in particular. I have a question I hope you can help me with:

·         Are there any comparison between SNOMED CT and the OBO ontologies in regards to coverage in biomedical domains? A sort of map to see what domains SNOMED covers and what domains OBO ontologies cover, and where there are overlaps?

 

 

Sincerely,

 

Kristian Malm-Nicolaisen

Adviser/Project manager

Norwegian Centre for E-health Research

Tel: <a href="tel:+47%20950%2059%20001" value="+4795059001" target="_blank">+47 950 59 001

www.ehealthresearch.no/en/

NSE en

 


______________________________________________
Tenk miljø - ikke skriv ut denne om det ikke er absolutt nødvendig
---

The information in this email, including attachments, may be confidential and is intended solely for the addressee(s). If you believe you received this email by mistake, please notify the sender by return email as soon as possible.

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Obo-discuss mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/obo-discuss



------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Obo-discuss mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/obo-discuss
Reply | Threaded
Open this post in threaded view
|

Re: OBO vs. SNOMED CT domain coverage

Chris Mungall
In reply to this post by Peter Robinson

In the domain of anatomy, there are over 4000 mappings between uberon classes and snomed-ct. But if you want to look at coverage there are nuances.

  • uberon doesn't follow the SNOMED SEP model, so "X part" terms are not directly mapped, but are trivially axiomatized
  • uberon covers metazoa in general, snomed focuses on human, but there are various veterinary terms
  • coverage of things like major organs and bones will be similar but there are many different ways to subdivide, and these will vary by use case
  • it looks like we have over 10k anatomical terms potentially applicable to humans for which we don't have a snomed mapping but it's not clear what percentage is error of omission, the snomed mappings are not yet vetted as thoroughly as the mappings to open ontologies

There will be these nuances across each of the different domains so I'd urge for a thoughtful approach to this, the papers Peter cites below analyzing phenotypic coverage are a great example

On 20 Jul 2017, at 5:38, Peter Robinson wrote:

Hi Kristian,

here is a comparison of the Human Phenotype Ontology with SNOMED and other UMLS terminologies:

https://lhncbc.nlm.nih.gov/system/files/pub8937.pdf

and here is some background information:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5210535/

hth, Peter


Peter Robinson

Professor of Computational Biology

The Jackson Laboratory for Genomic Medicine

10 Discovery Drive

Farmington, CT 06032

860.837.2095 t | 860.990.3130 m

[hidden email]

www.jax.org

Robinson lab: https://robinsongroup.github.io/

The Jackson Laboratory: Leading the search for tomorrow's cures


From: Kristian Nicolaisen <[hidden email]>
Sent: Thursday, July 20, 2017 3:45 AM
To: [hidden email]
Subject: [Obo-discuss] OBO vs. SNOMED CT domain coverage
 

Hi,

 

I am currently working on a report on the transition to ontology based terminologies in healthcare in Norway, in that regard I am writing about BFO in general and the OBO Foundry in particular. I have a question I hope you can help me with:

·         Are there any comparison between SNOMED CT and the OBO ontologies in regards to coverage in biomedical domains? A sort of map to see what domains SNOMED covers and what domains OBO ontologies cover, and where there are overlaps?

 

 

Sincerely,

 

Kristian Malm-Nicolaisen

Adviser/Project manager

Norwegian Centre for E-health Research

Tel: +47 950 59 001

www.ehealthresearch.no/en/

 


______________________________________________
Tenk miljø - ikke skriv ut denne om det ikke er absolutt nødvendig
---

The information in this email, including attachments, may be confidential and is intended solely for the addressee(s). If you believe you received this email by mistake, please notify the sender by return email as soon as possible.

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot_______________________________________________
Obo-discuss mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/obo-discuss


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Obo-discuss mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/obo-discuss
Reply | Threaded
Open this post in threaded view
|

Re: OBO vs. SNOMED CT domain coverage

Bill Hogan
In reply to this post by Kristian Nicolaisen
In general:

1. Disease Ontology (DOID) covers diseases
2. PATO and HPO cover lots of signs / phenotypic abnormalities
3. Drug Ontology covers drugs
4. Ontology for Biomedical Investigations covers some labs/radiology, but would be very limited for, say, clinical labs.
5. Foundational Model of Anatomy covers human body (although like Chris said, no SEP triples thank goodness).

Also, though, quantity has traditionally had precedence over quality.  SNOMED is improving.  

OBO has been very use case driven.  So if you find gaps in coverage, it tends to be quite easy to collaborate to fill in those gaps.  

OBO is also Open (the first 'O'), meaning there are far fewer restrictions on what you can do with it.  

Has Norway joined IHTSDO?

Bill

On Thu, Jul 20, 2017 at 3:45 AM, Kristian Nicolaisen <[hidden email]> wrote:

Hi,

 

I am currently working on a report on the transition to ontology based terminologies in healthcare in Norway, in that regard I am writing about BFO in general and the OBO Foundry in particular. I have a question I hope you can help me with:

·         Are there any comparison between SNOMED CT and the OBO ontologies in regards to coverage in biomedical domains? A sort of map to see what domains SNOMED covers and what domains OBO ontologies cover, and where there are overlaps?

 

 

Sincerely,

 

Kristian Malm-Nicolaisen

Adviser/Project manager

Norwegian Centre for E-health Research

Tel: <a href="tel:+47%20950%2059%20001" value="+4795059001" target="_blank">+47 950 59 001

www.ehealthresearch.no/en/

NSE en

 


______________________________________________
Tenk miljø - ikke skriv ut denne om det ikke er absolutt nødvendig

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Obo-discuss mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/obo-discuss



------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Obo-discuss mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/obo-discuss
Reply | Threaded
Open this post in threaded view
|

Re: OBO vs. SNOMED CT domain coverage

Kristian Nicolaisen

Hi again.

 

Thank you all for the quick response, it is very interesting to read your thoughts!

 

Just a little background for the report we are working on: Norway is a member of IHTSDO as of January of 2017, there are plans to pilot part of the ontology in the dental domain, but in no way go as far as Sweden who translated the whole thing ab initio. The report is meant to give the responsible governmental agency knowledge in the decision processes regard the SNOMED trails. One of the aspects we wanted to investigate was the complexity of integration of different ontologies with different structures vs. for example OBO ontologies which shares structures and upper-level architectures in clinical ICT environments.

 

Again, thank you for your responses, greatly appreciated!

 

Sincerely,

 

Kristian Malm-Nicolaisen

Adviser/Project manager

Norwegian Centre for E-health Research

Tel: +47 950 59 001

www.ehealthresearch.no/en/

NSE en

 

Fra: Bill Hogan [mailto:[hidden email]]
Sendt: fredag 21. juli 2017 03.39
Til: [hidden email]
Emne: Re: [Obo-discuss] OBO vs. SNOMED CT domain coverage

 

In general:

 

1. Disease Ontology (DOID) covers diseases

2. PATO and HPO cover lots of signs / phenotypic abnormalities

3. Drug Ontology covers drugs

4. Ontology for Biomedical Investigations covers some labs/radiology, but would be very limited for, say, clinical labs.

5. Foundational Model of Anatomy covers human body (although like Chris said, no SEP triples thank goodness).

 

Also, though, quantity has traditionally had precedence over quality.  SNOMED is improving.  

 

OBO has been very use case driven.  So if you find gaps in coverage, it tends to be quite easy to collaborate to fill in those gaps.  

 

OBO is also Open (the first 'O'), meaning there are far fewer restrictions on what you can do with it.  


Has Norway joined IHTSDO?


Bill

 

On Thu, Jul 20, 2017 at 3:45 AM, Kristian Nicolaisen <[hidden email]> wrote:

Hi,

 

I am currently working on a report on the transition to ontology based terminologies in healthcare in Norway, in that regard I am writing about BFO in general and the OBO Foundry in particular. I have a question I hope you can help me with:

·         Are there any comparison between SNOMED CT and the OBO ontologies in regards to coverage in biomedical domains? A sort of map to see what domains SNOMED covers and what domains OBO ontologies cover, and where there are overlaps?

 

 

Sincerely,

 

Kristian Malm-Nicolaisen

Adviser/Project manager

Norwegian Centre for E-health Research

Tel: <a href="tel:&#43;47%20950%2059%20001" target="_blank">+47 950 59 001

www.ehealthresearch.no/en/

NSE en

 


______________________________________________
Tenk miljø - ikke skriv ut denne om det ikke er absolutt nødvendig


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Obo-discuss mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/obo-discuss

 


______________________________________________
Tenk miljø - ikke skriv ut denne om det ikke er absolutt nødvendig

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Obo-discuss mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/obo-discuss
Reply | Threaded
Open this post in threaded view
|

Re: OBO vs. SNOMED CT domain coverage

Melissa Haendel-2
for dental ontologies, there has been some work done here:
I haven’t been involved for some time, but some nice work was done by Bill Duncan and Alan Ruttenberg, they would be the folks to ask about this.

I should also note that there will soon be an NCIT OBO release for cancer, that has represent cancer disease, phenotypes, etc. which would be relevant for oral cancers.

Cheers,
Melissa





On Jul 20, 2017, at 11:51 PM, Kristian Nicolaisen <[hidden email]> wrote:

Hi again.
 
Thank you all for the quick response, it is very interesting to read your thoughts!
 
Just a little background for the report we are working on: Norway is a member of IHTSDO as of January of 2017, there are plans to pilot part of the ontology in the dental domain, but in no way go as far as Sweden who translated the whole thing ab initio. The report is meant to give the responsible governmental agency knowledge in the decision processes regard the SNOMED trails. One of the aspects we wanted to investigate was the complexity of integration of different ontologies with different structures vs. for example OBO ontologies which shares structures and upper-level architectures in clinical ICT environments.
 
Again, thank you for your responses, greatly appreciated!
 
Sincerely,
 
Kristian Malm-Nicolaisen
Adviser/Project manager
Norwegian Centre for E-health Research
Tel: +47 950 59 001
<a href="x-msg://166/www.ehealthresearch.no/en/" style="color: purple; text-decoration: underline;" class="">www.ehealthresearch.no/en/
<image003.jpg>
 
Fra: Bill Hogan [[hidden email]] 
Sendt: fredag 21. juli 2017 03.39
Til: [hidden email]
Emne: Re: [Obo-discuss] OBO vs. SNOMED CT domain coverage
 
In general:
 
1. Disease Ontology (DOID) covers diseases
2. PATO and HPO cover lots of signs / phenotypic abnormalities
3. Drug Ontology covers drugs
4. Ontology for Biomedical Investigations covers some labs/radiology, but would be very limited for, say, clinical labs.
5. Foundational Model of Anatomy covers human body (although like Chris said, no SEP triples thank goodness).
 
Also, though, quantity has traditionally had precedence over quality.  SNOMED is improving.  
 
OBO has been very use case driven.  So if you find gaps in coverage, it tends to be quite easy to collaborate to fill in those gaps.  
 
OBO is also Open (the first 'O'), meaning there are far fewer restrictions on what you can do with it.  

Has Norway joined IHTSDO?

Bill
 
On Thu, Jul 20, 2017 at 3:45 AM, Kristian Nicolaisen <[hidden email]> wrote:
Hi,
 
I am currently working on a report on the transition to ontology based terminologies in healthcare in Norway, in that regard I am writing about BFO in general and the OBO Foundry in particular. I have a question I hope you can help me with:

·         Are there any comparison between SNOMED CT and the OBO ontologies in regards to coverage in biomedical domains? A sort of map to see what domains SNOMED covers and what domains OBO ontologies cover, and where there are overlaps?

 
 
Sincerely,
 
Kristian Malm-Nicolaisen
Adviser/Project manager
Norwegian Centre for E-health Research
Tel: <a href="tel:&#43;47%20950%2059%20001" target="_blank" style="color: purple; text-decoration: underline;" class="">+47 950 59 001
<image001.jpg>
 

______________________________________________ 
Tenk miljø - ikke skriv ut denne om det ikke er absolutt nødvendig


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Obo-discuss mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/obo-discuss

 

______________________________________________ 
Tenk miljø - ikke skriv ut denne om det ikke er absolutt nødvendig 
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot_______________________________________________
Obo-discuss mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/obo-discuss

Melissa Haendel, PhD
Associate Professor
Library & Dept. of Medical Informatics and Clinical Epidemiology 
[hidden email]
503-407-5970

Appointments: Shanez De Silva 




------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Obo-discuss mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/obo-discuss