fephyfofum / otu Goto Github PK

It would be nice to be able to say X% complete based on some set of criteria. At the very least, we have a placeholder for a status message. With status in place, we can search for things that need updating or browse by level of completeness, etc.

Currently the width of the rects for the otu labels on tree view is calculated from the number of characters in the name. But we are not using a fixed-width font so the actual dimensions of words containing the same number of characters vary. We compensate by making the boxes wider than they need to be for many words so that they are always (hopefully) wide enough for any word. There should be a way to select the text element of each node and get its bounding box so we can set the rects to that.

Alternatively, we could color the background of the text elements, but I haven't been able to get this to work.

A lot of ambiguity between treeJsons and sourceJsons.

Need to be able to add new properties to sources and trees in from the source view screen. Basically, an "add property" button should appear when in edit mode at the bottom of the table, which will add a row with a dropdown and a blank field. Set the property name in the dropdown and the value in the field and save changes to record the edits to the db. The dropdown is so we can restrict the keys to the set of approved ot: properties, but we can could allow custom properties, we just want to give people the set of known properties so they don't have to type them by hand...

This will require making property lists available via services. Need to add all the ot: properties to the enums to do this. See the nexson documentation page for the list of available properties: http://opentree.wikispaces.com/NexSON

on this one
http://localhost:8000/cgi-bin/tree_view.py?treeId=2598_6020
tree goes off the page

without the tree id listed on the study view there is no way to disambiguate the trees (and to refer to the id anywhere else).

Would be cool to have checkboxes for has ingroup, rooted, etc. type criteria on the browse/search page.

We should be able to manually annotate studies and trees. Waiting to hear how the annotations get worked out for the nexson spec.

not sure if it is best to have "edit tree" or "view tree" or "edit/view" or "view" and then when you get to the view you can click "edit". either way, I wanted to view the tree and figured edit was the button for me, but there may be a better way to present that.

doesn't work. nested cgi-bin problem.

This will facilitate editing without inducing irreversible changes. Working copies can be "saved" to replace local copies, or edits can be discarded. On push, any un-saved edits (i.e. working copies of trees) will be ignored.

This will solve the other issue of allowding things like rotation, rerooting, etc. without having to make major changes.

It looks like the taxomachine neo4j on the dev server has a maximum connection timeout of about 3 minutes. I am not sure how to increase this, but it definitely needs to be much longer if we want to get TNRS results for large trees.

An alternative would be to use a local instance of taxomachine to do the TNRS. This would require downloading the taxonomy. We could use/provide a snapshot of the taxonomy db so people would not have to build it. This is related to the issue of bringing taxonomy into OTU. See #14.

For tree 2407_5075, if I click on the tree root node it appears to go to some unrelated taxon. This seems to be a problem only of the bottom-most node in the tree. Not sure if this is actually deeper than the root, or if the root properties are coming through strangely...

Add the taxonomy to facilitate things like:

• annotating internal nodes with taxon names
• cross-referencing nodes from different trees on the basis of taxon assignment
• identifying percent taxonomic composition of trees and studies
• include more detailed taxonomic information in queries and reports

See also #13.

curl -X POST http://localhost:7474/db/data/ext/sourceJsons/graphdb/putSourceNewickSingle -H "Content-type:Application/json" -d '{"sourceId": "1379096234", "newickString": "(A,(B,((C,(D,E)),(F,(G,H)))))"}'

Currently the indexes are not updated when a property or a taxon assignment is changed. Fixing this should be as simple as removing the edited node and then re-indexing the edited source or tree.

Hopefully Peter will have nexson import in phylografter soon so we can test the capabilities. I'm not excited about validating nexson by hand... Have we tested it in JSON validation?

We should probably also have a nexson serialization method that works with the Representation framework for neo4j, so we can send nexson via neo4j's server plugin web services. This may belong in a separate issue.

need to do some transactions during the indexing. my laptop (4GB RAM) got overwhelmed. it finished, but wouldn't take much more.

Use d3 for this?

Browse should contain search + list local sources. This is pretty simple. We should change the standard list view of imported sources though to be more informative. It should mirror the search results view.

Add new source would then just be import from local file. This could get complex for loading multiple trees from a newick file, etc. E.g. we might not want to send things to the db until we know they are not total garbage. Not sure.

Not a high priority, just documenting thoughts as they occur.