Further Investigation in Unmatched Phenotypes

[yuanzhou]

Disease page disease/OMIM:255800 doesn't display unmatched phenotypes for Human and Fish, but compared to fish, it seems we have more phenotypes listed for human by looking at the shaded area in overview map. Actually, fish has 60 phenotypes and human has 116. The monarch production server has the same results. Something might be wrong.

---

Side notes: Disease page disease/OMIM:255800 shows 190 phenotypes under the phenotype tab, but there are duplicated ones, Phenogrid removes all the duplicated phenotype IDs before sending the data to simsearch. In this case, there are 169 unique phenotypes.

By default, simsearch has a cutoff in returned JSON data, which is 100.

[yuanzhou]

Found another issue about the unmatched phenotypes. If you test page disease/OMIM:270400, the ajax calls will be sent out to grab the unmatched phenotype label every time you invert the axis. This should be cached and invert axis shouldn't trigger the ajax again. Will troubleshoot.

Also need to add a loading spinner indicator since I've noticed that when the monarch-app server runs slowly, we will see "No unmatched" first, wait for a while then the unmatched phenotype will show up there.

[yuanzhou]

Has fixed the issue regarding sending out repeated ajax calls after inverting axis in PR #193 . Now only change the selections of species will update the unmatched phenotypes, all other actions in options menu won't.

[yuanzhou]

Turned out that the current _getUnmatchedSources() is not returning the correct number of unmatched phenotype IDs, this explains why the unmatched + matched != provided phenotype list. It also turned out that in many cases we will have many (30+ or more) unmatched phenotypes, will play with the table view to display these unmatched phenotypes.

[yuanzhou]

Got this fixed. Also styled the unmatched phenotypes in a table view.