[Obo-anatomy] where to begin?

[Sofia Robb]

Hello All,

I am new to constructing ontologies. I am embarking on an anatomy ontology for the fresh water flatworm planaria. I found that the C. elegans anatomy ontology has a lot of terms that could also work for planaria. I have also downloaded OBOedit2. Are there any tutorials or tips and hints that would be useful to me? 

Do I just load up the C.elegans ontology and start clicking on terms and modifying definitions as needed. Do I change the C. elegans IDs to my own?

Thank you for any help and information that you will share :)

Sofia

[Chris Mungall]

Hi Sofia,

Yes, as it happens, there was an entire course on building anatomy
ontologies, run by Melissa Haendel

https://academy.nescent.org/wiki/Ontologies_for_evolutionary_biology

(Hilmar - this is hosted on the Nescent wiki - are there plans to
migrate the wiki after Nescent goes away?)

We also ran a course for the GO editors
http://wiki.geneontology.org/index.php/Hinxton_OBO-Edit/Protege_4_workshop_Jan_2012

All of the course material is here (oops, google code):
http://oboformat.googlecode.com/svn/docs/tutorial/

Some basic tips before you get started:

No offence to Wao, but cloning it and making modifications is probably
not the way to go. Others have gone this route with other ontologies in
the past and can share their experience. Raymond may have suggestions
too.

Before going any further I'd stop and write down what some of the short
and long term goals of the ontology are. If you just need a semi-flat
list of terms for doing some image annotation then just dive in with
obo-edit and learn as you go along as most people do. However, if the
ontology is to have a more central role in your research, is likely to
be around a while, or have multiple users, or be integrated with
ontologies of either closely or distantly related organisms, modeling
neural networks, etc I'd really recommend dedicating some time upfront
to learning some modern OWL-based ontology development practices,
planning things out strategically, as this will all pay off in the
longer term. Everything will be more maintainable, amenable to automated
tooling, etc. I'll admit upfront that this can seem a bit daunting and
the tools are far from perfect.

Whichever route you take, what I'd really recommend is starting out with
a github project. It can be empty at first and just have a few files or
notes in the wiki sketching out some ideas and requirements. I know
you're familiar with github and you might be pleasantly surprised about
how much of your way of thinking for coding and bioinformatics
translates to ontology development - for example using our tools you can
set up a travis job to auto-check every commit on your repo, or pull
requests. I wrote this on setting up your gh project, but it's a bit out
of date now:
http://douroucouli.wordpress.com/2014/01/08/creating-an-ontology-project/
- after you set it up others here who are interested can follow tickets
etc.

Not really related to your question but you may want to look into some
of the work done by Daniel Lobo on planaria modeling?

> ------------------------------------------------------------------------------
> _______________________________________________
> Obo-anatomy mailing list
> Obo-a...@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/obo-anatomy

------------------------------------------------------------------------------
_______________________________________________
Obo-anatomy mailing list
Obo-a...@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/obo-anatomy

[Sofia Robb]

Thanks for the information. You pointed out some great places for me to start reading to figuring this all out.

Thank you!

Sofia

[Sofia Robb]

Hi Chris,

Is the material from the Nescent course available? I do not see any links to the material on this site https://academy.nescent.org/wiki/Ontologies_for_evolutionary_biology

Your post on creating an ontology project looks very useful.

Thanks Again,

Sofia

[Chris Mungall]

It looks like it's linked from the phenotype RCN blog:
http://www.phenotypercn.org/?page_id=1858

(The linked course material is to google code, arghh, need to migrate
that soon)

[Sofia Robb]

Great! I have some studying to do :) I am excited about this project. I am sure I will have more questions as I proceed.

Thanks,

Sofia

[Melissa Haendel]

Sofia-

let us know when you’ve digested things and we’d be happy to have a call with you to assist.

Thanks Chris for replying (on vacation). 

cheers

melissa

Dr. Melissa Haendel

Associate Professor
Ontology Development Group, OHSU Library
www.ohsu.edu/library/ontology
Department of Medical Informatics and Clinical Epidemiology
Oregon Health & Science University
hae...@ohsu.edu
skype: melissa.haendel
503-407-5970

[Sofia Robb]

Hi Melissa,

A call would be great! I was reading through the material that Chris sent yesterday. I understand a lot of it, but have definitely found some points that I am still confused about. 

Thanks,

Sofia

P.S. Sorry for the double email. I often forget to change to the email address I use for mailing lists.

[Sofia Robb]

Hi Melissa,

A call would be great! I was reading through the material that Chris sent yesterday. I understand a lot of it, but have definitely found some points that I am still confused about. 

Thanks,

Sofia

On Fri, Aug 21, 2015 at 5:45 AM, Melissa Haendel <hae...@ohsu.edu> wrote:

[Hilmar Lapp]

On Aug 19, 2015, at 6:15 PM, Chris Mungall <CJMu...@lbl.gov> wrote:

Yes, as it happens, there was an entire course on building anatomy ontologies, run by Melissa Haendel

https://academy.nescent.org/wiki/Ontologies_for_evolutionary_biology

(Hilmar - this is hosted on the Nescent wiki - are there plans to migrate the wiki after Nescent goes away?)

Just for the record, as we’ve meanwhile established that the actual course content isn’t on that wiki, but in case this thread shows up later in a someone’s search, we do not currently have plans to maintain or migrate this wiki.

However, I have now archived it on Zenodo for perpetuity in a form that others can re-instantiate it:
http://dx.doi.org/10.5281/zenodo.29430

  -hilmar

-- 

Hilmar Lapp -:- informatics.nescent.org/wiki -:- lappland.io