Perl 6 - the future is here, just unevenly distributed

IRC log for #opentreeoflife, 2014-07-15

| Channels | #opentreeoflife index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
02:13 towodo joined #opentreeoflife
03:46 jimallman joined #opentreeoflife
05:07 scrollback joined #opentreeoflife
10:08 scrollback joined #opentreeoflife
10:49 towodo joined #opentreeoflife
11:00 kcranstn joined #opentreeoflife
11:29 kcranstn joined #opentreeoflife
11:57 scrollback joined #opentreeoflife
12:05 scrollback joined #opentreeoflife
12:19 kcranstn joined #opentreeoflife
12:29 jimallman joined #opentreeoflife
13:27 kcranstn quiet here
13:31 * jimallman sees a tumbleweed rolling by...
13:39 kcranstn hears Morricone theme song
13:43 towodo joined #opentreeoflife
13:46 josephwb joined #opentreeoflife
13:52 * jimallman spits into the dust   #nocontext
13:54 josephwb @jimallman: [I don't expect you to know this but] do you know of a study where an exemplar individual from a set of duplicated taxa has been designated?
14:01 jimallman hm, i don’t know of one off-hand, but i might have a searchable metatag for you… pls stand by...
14:06 jimallman josephwb: yes, search for “^ot:isTaxonExemplar” in the docstore
14:06 jimallman https://github.com/OpenTreeOfLife/phylesystem-1/search?q=%5Eot%3AisTaxonExemplar&ref=cmdform
14:06 jimallman looks like ~ 20 so far...
14:07 jimallman (studies, that is)
14:09 josephwb great, thanks. we haven't supported this yet in TM (we select an arbitrary exemplar), but I guess we should take this into account
14:13 kcranstn @jimallman - thinking about mocking up a page layout for each version of the synthetic tree
14:13 kcranstn download links, metrics, etc
14:14 jimallman kcranstn: so these would be available on the site, like a history of past releases? i like it!
14:15 kcranstn let me sketch out what I am thinking
14:15 jimallman that could tell a nice story over time, if succeeding versions of the tree get more use/downloads
14:15 kcranstn we should start tracking downloads
14:15 jimallman josephwb: yes, please use these exemplars!
14:16 jimallman kcranstn: for sure. i’ll add a ticket for this.
14:16 kcranstn https://trello.com/c/lLA6b5kt/260-statistics-metrics
14:18 jimallman thanks. i suppose this will become a proper ticket in the Great Trello Migration of 2014.
14:19 kcranstn yes
14:19 jimallman speaking of which, i wonder if we could do that as a sort of “barn raising”… pick an afternoon and get a group together to swarm the trello board, we could have it done pretty quickly.
14:20 kcranstn @towodo and I are doing that later this week
14:26 kcranstn is there a github issue for treemachine to use the ot:isTaxonExemplar attribute?
14:27 jimallman kcranstn: i don’t see one, no.
14:27 * jimallman is adding an issue now...
14:29 kcranstn link when you have it, please
14:33 towodo joined #opentreeoflife
14:35 jimallman https://github.com/OpenTreeOfLife/treemachine/issues/107
14:35 jimallman three minutes old and already a hot topic!
14:36 kcranstn are there other nexson attributes that treemachine should be reading but is not? what about the “should contribute to synthesis” flag?
14:37 kcranstn looking at the exemplar nodes in http://tree.opentreeoflife.org/curator/study/view/ot_48/?tab=trees
14:37 kcranstn these are sister taxa, so do we really need to pick one?
14:37 kcranstn seems like requiring extra curation effort for no reason
14:40 jimallman kcranstn: i believe i handled that in code (might be an automatic selection), checking now…
14:41 kcranstn not for that example
14:42 jimallman right you are. i cleared an exemplar and had to choose it again. i’ll put this on my TODO list…
14:42 kcranstn I’ll re-open the issue
14:43 jimallman assuming there’s no possible “interesting” difference between these siblings, i’m inclined to choose automatically. Or we can simply ignore these and let treemachine sort them out…
14:43 jimallman link to issue?
14:44 kcranstn https://github.com/OpenTreeOfLife/opentree/pull/270#issuecomment-49041484
14:44 jimallman thanks!
14:45 kcranstn although I am not sure where the issue went (as opposed to the pull request)
14:46 josephwb so, Dail just raised a concern over the use of "Biologically correct" for the root.
14:47 josephwb He thinks it connotes "the TRUTH", and so has not been setting it, because further sampling may find a different root.
14:47 josephwb I don't quite understand the confusion, but he seems to think people will be confused
14:47 kcranstn ah, semantics
14:48 brucewayne joined #opentreeoflife
14:48 kcranstn did he have better wording?
14:48 josephwb asking now
14:49 josephwb hey brucewayne
14:50 josephwb wandered off from Gotham, didja?
14:51 josephwb kcranstn: Dail doesn't seem to be around. Maybe "experimentally determined"?
14:52 kcranstn inferred?
14:52 josephwb "publication specified"
14:52 josephwb yes, much better
14:52 kcranstn sounds like we also need some help text in the UI
14:54 josephwb this language doesn't make it into the nexson, so this should just be a UI change (if necessary)
14:54 josephwb I'm happy with "Inferred". What do you think jimallman?
14:55 josephwb maybe the validator uses such "biologically correct" language, tho
14:56 jimallman practically speaking, this mostly means the root has been explicitly chosen (or more likely, *confirmed*) by the curator, versus the inevitable root node because of NexSON’s hierarchical trees.
14:58 jimallman to me,  “inferred” conveys the opposite. perhaps “confirmed by curator” or “intentional” ..? “matches publication”?
14:59 kcranstn sounds like we may want broader opinion on this one
15:08 17SAAF7BZ joined #opentreeoflife
15:22 josephwb "Confirmed" sounds good. But, yeah, solicit suggestions.
15:25 josephwb "Specified" or "Designated" might work too
15:26 hdliv joined #opentreeoflife
15:26 josephwb "Curator designated"?
15:27 hdliv good day to all
15:27 josephwb hdliv: discussing possible name changes to "Biologically correct"
15:33 kcranstn suggestions so far: ‘confirmed by curator’, ‘matches publication’, ‘specified’, ‘designated'
15:33 kcranstn 'inferred'
15:34 kcranstn (hi dail!)
15:40 josephwb I kinda like "Curator designated" or "Curator confirmed. Often, trees are inferred as unrooted (RAxML, MrBayes). Curators often reroot the tree to match what the researcher intended (i.e. with explicitly sampled outgroups)
15:40 josephwb But I have no strong feelings on this.
15:47 hdliv I like the "Curator designated" or "Curator confirmed
15:48 hdliv but have not strong feelings.  my feelings are more against: "Biologically correct" — since I don't believe the tree i put in there is ' thetruth'
15:56 kcranstn yeah, I am convinced that Biologically correct is not the best wording
16:17 jimallman perhaps “confirmed by curator” .. ? This would read smoothly, since the UI text (from left to right) is
16:18 josephwb +1
16:18 jimallman “Tree root is  [  ]  confirmed by curator  [  ]  arbitrary (for display only)”
16:34 hdliv joined #opentreeoflife
16:41 kcranstn I would get rid of “for display only"
16:41 kcranstn that suggests that the root doesn’t matter, since it is for display onle
16:41 kcranstn only
16:52 jimallman kcranstn: this was intended, in the sense that absent confirmation, the current root could be just a consequence of the NexSON format. in other words, this might be an un-rooted tree. we don’t currently have the means to show it as an unrooted tree (a loose network), so in that sense it’s “for display only”.
16:52 kcranstn ah, I see.
16:52 jimallman i agree that it’s awkward wording, but that was the intent. still, maybe “arbitrary” makes the point more succinctly.
16:53 jimallman fyi - i have an improved MRCA test online for devtree: http://devtree.opentreeoflife.org/curator/study/edit/pg_2823?tab=trees&tree=tree6570
16:53 jimallman this sidesteps the show-subtended-tree requirement by linking to the MRCA in the synth-tree viewer..
16:54 kcranstn got a message in the status bar about the tree not found
16:54 kcranstn but then it loaded
16:54 kcranstn “Click any node below to set the ingroup clade. Click any node or edge to re-root the entire tree. N.B. An arbitrary root (see options above) will only be used to draw the tree and correctly set the ingroup clade. It will not be considered biologically correct. “
16:54 jimallman ! good to know, thanks. sounds like a race condition
16:55 kcranstn suggested rewording
16:55 kcranstn coming...
16:57 jimallman thanks (off to make a sandwich, back in a few....)
16:57 kcranstn actually, I have some thoughts about this whole rooting workflow. more soon (meetings all afternoon)
17:13 josephwb kcranstn and jimallman: (I know you may not be around at the moment) one issue I've come up against is that the curator only lets you root by node, whereas, say, FigTree lets you root by edge (placing a node on that edge with some sort of mid-rooting method).
17:13 josephwb That is usually what we want e.g. the root of the tree lies *between* crocodiles and birds, not at one of the actual nodes in the tree
17:14 kcranstn the help text says node or edge, but you can only do node?
17:14 josephwb really? let me try...
17:14 kcranstn yup - edges work
17:16 josephwb whoops
17:16 kcranstn these are my favourite kind of bug reports
17:20 jimallman agreed!
17:56 hdliv joined #opentreeoflife
18:06 kcranstn hey @josephwb - don’t see a hackathon application from you
18:06 kcranstn was that intentional
18:06 kcranstn ?
18:06 josephwb no, just haven't got to it yet. thanks for the reminder
18:06 kcranstn too late...
18:07 kcranstn maybe
18:07 kcranstn we are reviewing right now
18:11 josephwb really?
18:13 josephwb kcranstn ^
18:13 josephwb should i bother?
18:13 kcranstn don’t worry about it - we are assuming you will be there ;)
18:15 josephwb oh, just did
18:34 hdliv left #opentreeoflife
18:39 jimallman kcranstn: regarding the hackathon, i’m in a similar boat (don’t know if i should attend)… thoughts?
18:41 kcranstn participate remotely?
18:42 jimallman that certainly works for me, and was my first guess.
18:42 jimallman we could even keep a standing hangout, if that would help.
18:42 jimallman for remote folk..
18:43 jimallman s/guess/thought
18:45 jimallman hm, looks like someone wants to revive the PhyloStyloTastic project there: https://github.com/OpenTreeOfLife/hackathon/issues/17
19:34 kcranstn joined #opentreeoflife
20:06 josephwb jimallman: i think i am missing something obvious (again): for trees uploaded with node support values, where is that information communicated in the curator app?
20:07 josephwb i don't see a way to specify what those values mean
20:07 josephwb an example is study pg_2924
20:19 josephwb jimallman ^
20:20 kcranstn joined #opentreeoflife
20:58 kcranstn joined #opentreeoflife
21:07 * jimallman will follow up w/ josephwb by email to try to understand his question
21:14 josephwb joined #opentreeoflife
21:17 josephwb jimallman: i got your email
21:17 josephwb node support will be part of the tree string
21:17 josephwb e.g.:
21:18 josephwb ((A:0.01,B:0.03)99:0.04,C:0.07);
21:18 josephwb here, 99 is the node support
21:18 josephwb you know this though
21:19 josephwb I uploaded the tree here with node support for all nodes: http://tree.opentreeoflife.org/curator/study/view/pg_2924/?tab=trees
21:19 josephwb (in PG, obviously)
21:19 josephwb they were bootstrap values (instead of probabilities)
21:20 josephwb erg, i will just email you back
21:29 kcranstn joined #opentreeoflife
21:43 josephwb joined #opentreeoflife
21:58 kcranstn joined #opentreeoflife
22:15 kcranstn joined #opentreeoflife
23:15 josephwb joined #opentreeoflife

| Channels | #opentreeoflife index | Today | | Search | Google Search | Plain-Text | summary