Document design pattern strategy for all phenotype ontologies

cmungall commented 8 years ago

Desiderata

[ ] All pheno ontologies (at least MP and HP) should use the same design pattern approach
[ ] design patterns in MP and HP should be coordinated
- they may diverge in terms of textual definition pattern, synonym pattern, etc
- we want to be able to easily compare and reuse the DPs
[ ] The DPs should have clear documentation that is understandable to a non-ontologist user
[ ] The DPs should be computable and in a standard syntax
[ ] The DPs should allow the option of generating ontology classes from simple tables
- this may be used for generating portions of the ontology, a la TermGenie
- this can also be used for QC purposes. E.g. comparing what an autogenerated subset looks like compared with the manually asserted subset
[ ] Should allow for reverse engineering analysis
- E.g take the ontology, automatically partition classes into their explicit or implicit DPs
[ ] Should fit easily into ontology development workflow
Proposal

Use DOSDPs to document patterns as YAML. As well as serving as computable documentation, these can be used

in a generative capacity: see ECTO or OBA for examples
- for trivial compositional classes, the ontology author fills in values in CSVs
- even this step can be partially automated. E.g. search the literature for all X atresia, map X to uberon, create CSV, manually vet and remove unlike entries
in a test/parse capacity
- I have done this for uberon and HPO, will post more here later
  TBD

How does this relate to the DP docs on the google doc?

These should be converted to YAML

How does this related to the new docs on the wiki?

The wiki can provide high level documentation that does not boil down to the compositional patterns expressible via DOSDPs. These can link to the DOSDPs for specific details.

mellybelly commented 8 years ago

I would love to see this happen. There is a lot of design pattern documentation and much has been done already to manually align. We have some checks but not enough, this seems very timely.

@ybradford This may also be a good time to plan future ZP and/or coordination with ZFIN in any case too.

@cmungall once we have the design patterns perhaps we can use them to QA and/or enhance some of the other phenotype ontologies too, making for a better and more extensive upheno?

I would also like to see these implemented incrementally in TermGenie, as well as having some reporting on potentially missing classes based upon the existing classes & axioms. The reason being, we seem to continue to need high level grouping classes - we should be able to see that these are missing.

balhoff commented 8 years ago

I've created a query generator which takes a DOSDP and creates a SPARQL query for finding terms which meet the design pattern in an ontology: https://github.com/balhoff/dosdp-scala

I need to create a command line so that it's easy to run. It will use Owlet for expanding the variable constraints in the case that the constraint is an expression.

sbello commented 8 years ago

There is existing documentation for patterns that MP and HPO have worked on: https://docs.google.com/document/d/1LOYQ6pOlvapzJhHbSA_Rb3aZYylXZ564B78jiysvB-U/edit?usp=sharing

pnrobinson commented 8 years ago

Hi everybody,

would it be ok for everybody if I put a link to this google doc onto the HPO wiki and we then discuss the best place to host this data in the future? It may be an idea to set up a common wiki for both MP and HP where there are subsites for each ontology or it would also be fine to have two websites, Cindy, what do you think?

In any case, I would love to start making some of this material more public and also would like to write down explicitly some of the background information for the medical terminology (to keep me from having to look up the same information multiple times!)

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

peter.robinson@jax.orgmailto:peter.robinson@jax.org

www.jax.org

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

From: sbello notifications@github.com Sent: Wednesday, August 31, 2016 2:18 PM To: obophenotype/upheno Subject: Re: [obophenotype/upheno] Document design pattern strategy for all phenotype ontologies (#168)

There is existing documentation for patterns that MP and HPO have worked on: https://docs.google.com/document/d/1LOYQ6pOlvapzJhHbSA_Rb3aZYylXZ564B78jiysvB-U/edit?usp=sharing

You are receiving this because you are subscribed to this thread.

Reply to this email directly, view it on GitHubhttps://github.com/obophenotype/upheno/issues/168#issuecomment-243853499, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AEtuPHVEuZdtNdnzDfw5NX2wWTlkESlHks5qlcVtgaJpZM4Ju0CO.

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.

pnrobinson commented 8 years ago

Fine with me! ☺

Thanks Peter!

Nicole Vasilevsky, PhD  Senior Biocurator Ontology Development Group, Library Oregon Health & Science University vasilevs@ohsu.edumailto:vasilevs@ohsu.edu  503-806-6900  skype: nicolevasilevsky

From: phenotype-ontologies-editors@googlegroups.com on behalf of Peter Robinson Peter.Robinson@jax.org Date: Wednesday, August 31, 2016 at 11:45 AM To: obophenotype/upheno upheno@noreply.github.com, obophenotype/upheno reply@reply.github.com Cc: "phenotype-ontologies-editors@googlegroups.com" phenotype-ontologies-editors@googlegroups.com Subject: Re: [obophenotype/upheno] Document design pattern strategy for all phenotype ontologies (#168)

Hi everybody,

would it be ok for everybody if I put a link to this google doc onto the HPO wiki and we then discuss the best place to host this data in the future? It may be an idea to set up a common wiki for both MP and HP where there are subsites for each ontology or it would also be fine to have two websites, Cindy, what do you think?

In any case, I would love to start making some of this material more public and also would like to write down explicitly some of the background information for the medical terminology (to keep me from having to look up the same information multiple times!)

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

peter.robinson@jax.orgmailto:peter.robinson@jax.org

www.jax.org

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

From: sbello notifications@github.com Sent: Wednesday, August 31, 2016 2:18 PM To: obophenotype/upheno Subject: Re: [obophenotype/upheno] Document design pattern strategy for all phenotype ontologies (#168)

There is existing documentation for patterns that MP and HPO have worked on: https://docs.google.com/document/d/1LOYQ6pOlvapzJhHbSA_Rb3aZYylXZ564B78jiysvB-U/edit?usp=sharing

— You are receiving this because you are subscribed to this thread.

Reply to this email directly, view it on GitHubhttps://github.com/obophenotype/upheno/issues/168#issuecomment-243853499, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AEtuPHVEuZdtNdnzDfw5NX2wWTlkESlHks5qlcVtgaJpZM4Ju0CO.

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.

You received this message because you are subscribed to the Google Groups "phenotype-ontologies-editors" group. To unsubscribe from this group and stop receiving emails from it, send an email to phenotype-ontologies-editors+unsubscribe@googlegroups.commailto:phenotype-ontologies-editors+unsubscribe@googlegroups.com. For more options, visit https://groups.google.com/d/optout.

pnrobinson commented 8 years ago

Fine by me. It would be great to have this in one place. The link I put in the github thread allows people to comment but not edit. Sue

From: phenotype-ontologies-editors@googlegroups.com [mailto:phenotype-ontologies-editors@googlegroups.com] On Behalf Of Nicole Vasilevsky Sent: Wednesday, August 31, 2016 2:46 PM To: Peter Robinson; obophenotype/upheno; obophenotype/upheno Cc: phenotype-ontologies-editors@googlegroups.com Subject: Re: [obophenotype/upheno] Document design pattern strategy for all phenotype ontologies (#168)

Fine with me! ☺

Thanks Peter!

Nicole Vasilevsky, PhD  Senior Biocurator Ontology Development Group, Library Oregon Health & Science University vasilevs@ohsu.edumailto:vasilevs@ohsu.edu  503-806-6900  skype: nicolevasilevsky

From: phenotype-ontologies-editors@googlegroups.com on behalf of Peter Robinson Peter.Robinson@jax.org Date: Wednesday, August 31, 2016 at 11:45 AM To: obophenotype/upheno upheno@noreply.github.com, obophenotype/upheno reply@reply.github.com Cc: "phenotype-ontologies-editors@googlegroups.com" phenotype-ontologies-editors@googlegroups.com Subject: Re: [obophenotype/upheno] Document design pattern strategy for all phenotype ontologies (#168)

Hi everybody,

would it be ok for everybody if I put a link to this google doc onto the HPO wiki and we then discuss the best place to host this data in the future? It may be an idea to set up a common wiki for both MP and HP where there are subsites for each ontology or it would also be fine to have two websites, Cindy, what do you think?

In any case, I would love to start making some of this material more public and also would like to write down explicitly some of the background information for the medical terminology (to keep me from having to look up the same information multiple times!)

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

peter.robinson@jax.orgmailto:peter.robinson@jax.org

www.jax.org

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

From: sbello notifications@github.com Sent: Wednesday, August 31, 2016 2:18 PM To: obophenotype/upheno Subject: Re: [obophenotype/upheno] Document design pattern strategy for all phenotype ontologies (#168)

There is existing documentation for patterns that MP and HPO have worked on: https://docs.google.com/document/d/1LOYQ6pOlvapzJhHbSA_Rb3aZYylXZ564B78jiysvB-U/edit?usp=sharing

— You are receiving this because you are subscribed to this thread.

Reply to this email directly, view it on GitHubhttps://github.com/obophenotype/upheno/issues/168#issuecomment-243853499, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AEtuPHVEuZdtNdnzDfw5NX2wWTlkESlHks5qlcVtgaJpZM4Ju0CO.

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.

You received this message because you are subscribed to the Google Groups "phenotype-ontologies-editors" group. To unsubscribe from this group and stop receiving emails from it, send an email to phenotype-ontologies-editors+unsubscribe@googlegroups.commailto:phenotype-ontologies-editors+unsubscribe@googlegroups.com.

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

You received this message because you are subscribed to the Google Groups "phenotype-ontologies-editors" group. To unsubscribe from this group and stop receiving emails from it, send an email to phenotype-ontologies-editors+unsubscribe@googlegroups.commailto:phenotype-ontologies-editors+unsubscribe@googlegroups.com.

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

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.

cindyJax commented 8 years ago

I do not have a preference for one site with sub pages, or two interconnected sites. As long as we all agree that this is/these are the definitive go-to sites. Cindy

From: Peter Robinson notifications@github.com<mailto:notifications@github.com> Reply-To: obophenotype/upheno reply@reply.github.com<mailto:reply@reply.github.com> Date: Wednesday, August 31, 2016 at 2:48 PM To: obophenotype/upheno upheno@noreply.github.com<mailto:upheno@noreply.github.com> Subject: Re: [obophenotype/upheno] Document design pattern strategy for all phenotype ontologies (#168)

Fine by me. It would be great to have this in one place. The link I put in the github thread allows people to comment but not edit. Sue

From: phenotype-ontologies-editors@googlegroups.commailto:phenotype-ontologies-editors@googlegroups.com [mailto:phenotype-ontologies-editors@googlegroups.com] On Behalf Of Nicole Vasilevsky Sent: Wednesday, August 31, 2016 2:46 PM To: Peter Robinson; obophenotype/upheno; obophenotype/upheno Cc: phenotype-ontologies-editors@googlegroups.commailto:phenotype-ontologies-editors@googlegroups.com Subject: Re: [obophenotype/upheno] Document design pattern strategy for all phenotype ontologies (#168)

Fine with me! ☺

Thanks Peter!

Nicole Vasilevsky, PhD  Senior Biocurator Ontology Development Group, Library Oregon Health & Science University vasilevs@ohsu.edumailto:vasilevs@ohsu.edumailto:vasilevs@ohsu.edu  503-806-6900  skype: nicolevasilevsky

From: phenotype-ontologies-editors@googlegroups.com<mailto:phenotype-ontologies-editors@googlegroups.com> on behalf of Peter Robinson Peter.Robinson@jax.org<mailto:Peter.Robinson@jax.org> Date: Wednesday, August 31, 2016 at 11:45 AM To: obophenotype/upheno upheno@noreply.github.com<mailto:upheno@noreply.github.com>, obophenotype/upheno reply@reply.github.com<mailto:reply@reply.github.com> Cc: "phenotype-ontologies-editors@googlegroups.commailto:phenotype-ontologies-editors@googlegroups.com" phenotype-ontologies-editors@googlegroups.com<mailto:phenotype-ontologies-editors@googlegroups.com> Subject: Re: [obophenotype/upheno] Document design pattern strategy for all phenotype ontologies (#168)

Hi everybody,

would it be ok for everybody if I put a link to this google doc onto the HPO wiki and we then discuss the best place to host this data in the future? It may be an idea to set up a common wiki for both MP and HP where there are subsites for each ontology or it would also be fine to have two websites, Cindy, what do you think?

In any case, I would love to start making some of this material more public and also would like to write down explicitly some of the background information for the medical terminology (to keep me from having to look up the same information multiple times!)

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

peter.robinson@jax.orgmailto:peter.robinson@jax.orgmailto:peter.robinson@jax.org

www.jax.org

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

From: sbello notifications@github.com<mailto:notifications@github.com> Sent: Wednesday, August 31, 2016 2:18 PM To: obophenotype/upheno Subject: Re: [obophenotype/upheno] Document design pattern strategy for all phenotype ontologies (#168)

There is existing documentation for patterns that MP and HPO have worked on: https://docs.google.com/document/d/1LOYQ6pOlvapzJhHbSA_Rb3aZYylXZ564B78jiysvB-U/edit?usp=sharing

— You are receiving this because you are subscribed to this thread.

Reply to this email directly, view it on GitHubhttps://github.com/obophenotype/upheno/issues/168#issuecomment-243853499, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AEtuPHVEuZdtNdnzDfw5NX2wWTlkESlHks5qlcVtgaJpZM4Ju0CO.

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.

You received this message because you are subscribed to the Google Groups "phenotype-ontologies-editors" group. To unsubscribe from this group and stop receiving emails from it, send an email to phenotype-ontologies-editors+unsubscribe@googlegroups.commailto:phenotype-ontologies-editors+unsubscribe@googlegroups.commailto:phenotype-ontologies-editors+unsubscribe@googlegroups.com.

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

You received this message because you are subscribed to the Google Groups "phenotype-ontologies-editors" group. To unsubscribe from this group and stop receiving emails from it, send an email to phenotype-ontologies-editors+unsubscribe@googlegroups.commailto:phenotype-ontologies-editors+unsubscribe@googlegroups.commailto:phenotype-ontologies-editors+unsubscribe@googlegroups.com.

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

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.

— You are receiving this because you are subscribed to this thread.

Reply to this email directly, view it on GitHubhttps://github.com/obophenotype/upheno/issues/168#issuecomment-243862526, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AEXFtIihFHjckl4y6dLT2W6cnONbmLA-ks5qlcyMgaJpZM4Ju0CO.

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.

obophenotype / upheno