Annotating ontology terms with ISA alternative

[Francesco Vitali]

Dear all,

I am in the process of developing an ontolgoy for the nutritional field (ONS, which can be found here https://github.com/FrancescoVit/Ontology-for-Nutritional-Studies; still in an embrional stage I would say). 

The ontology right now, already imports a lot of terms from other ontologies and uses the BFO ontology for the top level terms and the OBI for hig-mid level terms. To improve the possibilities of cross-referencing ONS, I would like to include/map terms from the ISA-TAB format to the ontology.

I have seen that some terms in OBI are annotated with the "ISA alternative term" property (OBI_0001847), but I don't think that those terms cover the full spectrum of ISA terminology. On the other end, I have also seen the ISA-OBI-mapping spreadsheet file and the OBI-ISAview-subset.owl file as well as the isaterms.owl file.

I would like to have some guidance from you for completing this task. To my understanding, I was thinking of doing something like this:

1) import the whole "isaterms.owl" file, listing all the terms under a "ISA-Terms" class in my ontology

2) Import the whole "OBI-ISAview-subset.owl" to be sure to have all the OBI terms to be used for next step (actually, it seems that not all the necessary OBI classes would be imported like this, so I do not know if this is the correct way)

3) using the indication reported in the ISA-OBI-mapping spreadsheet file, connect each term to a term in OBI 

Do you think this is a good approach? Or, alternatively, would it be better to make more use of the "ISA alternative term" annotation property on OBI classes?

I am more than interested in having any other form of feedback from you on this matter.

Thank you for the attention,

Best wishes,

Francesco Vitali.

[Susanna-Assunta Sansone]

Dear Francesco,

Quick note to say that the team is currently away at different meetings and one of us will come back to you on this asap.

Regards

Susanna 

(Sent from my phone)

-- 
--

--
--
--
 
You received this message because you are subscribed to the Google
Groups "ISAforum" group.
To post to this group, send email to isaf...@googlegroups.com
To unsubscribe from this group, send email to
isaforum+u...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/isaforum?hl=en-GB
 
Visit the ISA tools website at http://isa-tools.org and http://isacommons.org
---
You received this message because you are subscribed to the Google Groups "ISAforum" group.
To unsubscribe from this group and stop receiving emails from it, send an email to isaforum+u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[PÅL ANDERS FRONTH NYHUS]

Will you be using protege for the .owls ? Xml forced mapping like xmlspy?

Sent from my iPhone

[Francesco Vitali]

Thank you Susanna for the note.

I am using Protege for all the ontology editing-managing. 

Francesco.

[Philippe]

Hello Francesco, Pål Anders

Thanks for getting in touch. I just had a look. Before delving into the issues and possibly making a call about overall approach, we would need to understand how you intend to you the ONS ontology in the context of ISA tools.
1. Do you need to create nutrition study specific ISA configurations for use with ISAcreator?
2. Do you need to convert existing nutrition studies to RDF triples?
3. Do you plan to use ONS to build web forms?

The reason for asking is the following: OBI-ISAview-subset.owl  as isaterms.owl have been developped to serve a very specific purpose.
With OBI-ISAview-subset, we carefully selected ontological artefacts which shared the same top level ontology. This meant that compatibility was almost built-in, modulo minor tweaks.

In the case of ONS, while the steps seems very similar, the fact that some resources do not share the same URI for same classe causes a number of duplications, dangling trees and so on.
(e.g anatomical_entity is not a child of entity)
'information content entity'  appears in many places on the current hierarchy , as does processes. These are manifestations of issues with imports.

Then, 'information Tab' appears as a child of Assay, which does not seem right, the same for 'omics tab'. In either case, 'citation' is a child of 'information tab' as it is a child of  'omics tab', which again is possibly not ideal.
(i just noted the comment, but again, the hierarchy and reasoner will not know about this)

On the other hand, you are  trying to address an number of cases which are not currently supporting by existing resources. One may also consider contribution to those resources  (e.g OBI) to cover your need to you benefit from the overall infrastructure and (hopefully) long term support.

All the best

Philippe

[Francesco Vitali]

Dear Philippe, thank you for the answer and indications. 

I'll try to answer to the points you raised:

1- The ONS ontology in the first place was developed has to support the needs of ENPADASI project, then as we were not able to find an ontology focused on the nutritional field, we commited to create a more general purpose ontology. To be really short, the project aims at creating an infrastructure for storing nutritional studies and enabling federated analysis, and uses two different database systems: istances of the Phenotype Database and istances of OPAL server. In this unfrastructure, the ontology serves as common annotation and for querying. In this context, it would be beneficial to be able to include a mapping to the ISA-TAB format to enhance standardisation of study related terms, to enhance interoperability, and because the Phenotype database inhers some of ISA-TAB structure. 

2- there are in fact some problems in the class hierarchy I am trying to fix. As you were suggesting there are some problems-repetition with imports. Nevertheless, those should be fixed soon.

3- the classes you are highlighting will disappear. Those classes were mainly linked to the initial ideal of applied ontology and doesn't really make sense for a general purpose ontology, as we want ons to be.

For the purpose highlighted in point 1), the higher benefit would be from a stand alone ontology, rather than from a contribution to existing resources. Nevertheless we tried to make the most use of those resources.

Best wishes,

Francesco.

isaforum+unsubscribe@googlegroups.com

For more options, visit this group at
http://groups.google.com/group/isaforum?hl=en-GB
 
Visit the ISA tools website at http://isa-tools.org and http://isacommons.org
---
You received this message because you are subscribed to the Google Groups "ISAforum" group.

To unsubscribe from this group and stop receiving emails from it, send an email to isaforum+unsubscribe@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--

Francesco Vitali - PhD

phone: +39 055 4574714

skype: francesco.vitali.fi