search

monarch-initiative / monarch-legacy

Monarch web application and API
BSD 3-Clause "New" or "Revised" License
42 stars 37 forks source link

Anatomy pages #1271

Closed jmcmurry closed 7 years ago

jmcmurry commented 8 years ago

We need to introduce anatomy pages for classes in UBERON. Tabs would be:

As usual we would use the logical definitions to subsume subclasses; however, there is some question in my mind about the type of relationships that are valuable to traverse. Eg. what does the "Overview" tab look like. What hierarchies do we expose?

For instance, it would be nice to see all phenotypes associated with the brain and all its sub-parts. But it would also be nice to see all phenotypes associated with the circulatory system and its components. These are functionally related but positionally dispersed. These subtleties have been relevant before for phenotypes, but for anatomy it seems especially difficult and important. I'm not that familiar with the UBERON structure, so perhaps this is easier than I'm giving it credit for. From a UI standpoint, though, I don't see this as terribly straightforward.

Please discuss. @cmungall @mellybelly

cmungall commented 8 years ago

Depends on https://github.com/monarch-initiative/dipper/issues/147

As a first pass the anatomy page could have the generic ontology info/graph (note: we will need part-ofs in hierarchy), plus a single tab for gene. We may be able to spin separate tabs and populate via scigraph from some extensive metadata in the ontology, e.g. images, species.

The phenotype tab on the anatomy page will require more thought. We will probably want an affects closure in phenotype associations, populated by following P->has-part->inheres-in|inheres_in_part_of

jmcmurry commented 7 years ago

nice viz: http://genome.ucsc.edu/cgi-bin/hgGtexTrackSettings?hgsid=571665579_afGfPWeYJLcu0466KHATIRd3eI0e&c=chr9&g=gtexGene

cmungall commented 7 years ago

cc @fbastian

selewis commented 7 years ago

I'll be visiting this group on Thursday if you have specific questions you'd like me to ask.

On Tue, Dec 13, 2016 at 21:14 Chris Mungall notifications@github.com wrote:

cc @fbastian https://github.com/fbastian

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/monarch-initiative/monarch-app/issues/1271#issuecomment-266944733, or mute the thread https://github.com/notifications/unsubscribe-auth/ABcuEHKijCiKM9rgAqz2xwbkx84VkAP6ks5rH3sbgaJpZM4IwOlc .

cmungall commented 7 years ago

Looks like all the background work is done!

First pass

Here is an example command line query demonstrating this works:

qbiogolr.py MGI:2685856 -C anatomy
MGI:2685856     RO:0002206      UBERON:0003499;UBERON:0000922;UBERON:0005147;UBERON:0001068;UBERON:0001911;UBERON:0003220;UBERON:0000992;UBERON:0001844;UBERON:0000081;UBERON:0002116;UBERON:0006518;UBERON:0002048;UBERON:0001736;UBERON:0001343;UBERON:0001155;UBERON:0002197;UBERON:0008971;UBERON:0000982;UBERON:0000966;UBERON:0000947

Second pass

cmungall commented 7 years ago

I just realized I had the checklist here #1312

jmcmurry commented 7 years ago

Would love to see this happen. It would be important to demonstrate for Uberon proposal; when will this go in?

cmungall commented 7 years ago

Example of querying: https://github.com/biolink/ontobio/blob/master/notebooks/Gene_Expression.ipynb

(note: this uses an in-memory version of uberon so the lack of cached part-of closure in solr doesn't affect this)

kshefchek commented 7 years ago

Added with https://github.com/monarch-initiative/monarch-app/pull/1496

jmcmurry commented 7 years ago

When will this be live?

kshefchek commented 7 years ago

Hopefully in time for GSA (or COB today).