AWG-chat-20070703

From GO Wiki
Jump to: navigation, search
gwg: hello!
eurie: morning
j-lo has joined the channel
gwg: comment va la bataille avec les limaces?
8:30 AM
pascale: plutot bien! les oiseaux ont gagne et mange toutes les fraises
gwg: ah non! sacre bleu
gwg: hier soir, il pleuvait, et il y avait beaucoup de limaces 
rama has joined the channel
gwg: je les ai tue sous mes chaussures
pascale: ah oui> j'en ai attrape quelques unes, et je les ai noyees dans de l'eau salee
rama: gwg- please translate
gwg: last night it was raining and there were loads of slugs around
gwg: I killed them under my shoes
j-lo: yeah, it's been a good year for slugs and snails so far 
8:35 AM
j-lo: i feed them to my chickens
gwg: il ya avait aussi des escargots, et je les ai lancee dans la rue
eurie: does anyone know if donghui and val are around?
eurie: ben is out of town
eurie: what about chris?
sjcarbon: talked to him last night before he left
sjcarbon: not in yet
sjcarbon: lemme take a look
j-lo: you throw the snails into the road?
gwg: or into my neighbours garden, depending on my mood 
pascale: that's really evil ! 
cjm has joined the channel
gwg: well, they only have flowers and stuff, so they don't mind, I'm sure
j-lo: I saw Val earlier - lemme email her...
gwg: I have to protect my crops!
pascale: bien sur! 
8:40 AM
eurie: i'm sure this isn't the last time we'll talk about mapper and finder.  
eurie: nice work in getting those set up and integrated into the rest of AmiGO, Seth and Amelia
j-lo: yeah - looks really good
gwg: that was Seth, not me
sjcarbon: stole all of amelia's front stuff
sjcarbon: some work still needs to be
eurie: ok, i take that back then - no compliments for you amelia.  
sjcarbon: done in back to fully integrate it
gwg: I feel much better now Eurie! 
gwg: who is chairing today?
eurie: so should we talk about some features we'd like implemented on finder and mapper?
eurie: me
sjcarbon: let's talk!
eurie: Seth, do you want to start out with finder or mapper?
sjcarbon: I'd rather fish for what people like in other implementations
sjcarbon: oops
sjcarbon: parsed that wrong
sjcarbon: mapper
8:45 AM
gwg: as in map2slim?
sjcarbon: yes
eurie: I like the ability to choose one ontology and the specific organism annotation set
eurie: I like the options you have now - choose an existing slim or upload your own slim as well as uploading your own gene_assocation file
gwg: sorry, do you have the URLs of those two other implementations? I've lost the email between here and my work computer
rama: I also like the ability to upload a slim in OBO format
rama: http://go.princeton.edu/cgi-bin/GOTermMapper
rama: SGD's -
rama: http://db.yeastgenome.org/cgi-bin/GO/goTermMapper\
rama: Sorry, there is a slash at the end. 
gwg: thanks Rama!
8:50 AM
eurie: Are all the slim sets that are in the OBO file (using the category tag, I think?) loaded into AmiGO?
gwg: they're all in the database, yes
eurie: cool - so could all those be available from the mapper?
sjcarbon: i suppose once the backend is cleaned up from where it is now
j-lo: we should definitely do that
gwg: the current mapper doesn't work in a very efficient way, it parses an ontology file plus whatever goslim you upload
gwg: so it doesn't run off the database at all
gwg: did anyone have a look at the options available here: http://toy.lbl.gov:9006/cgi-bin/amigo/goslimmer.cgi
eurie: this wasn't working for me this morning
gwg: no, I am half-way through uploading various files at the moment unfortunately!
eurie: never mind - i can see it now
8:55 AM
gwg: the main difference is that you can upload a list of terms or an ontology file
gwg: you can also upload a list of gene products and it will use the data from the database
gwg: or you can upload your own association file
gwg: It's only a partial implementation at the moment because I didn't know how useful or otherwise that would be 
gwg: sorry, I think I've just replaced an important template file so the page isn't working right now!
9:00 AM
j-lo has quit the server saying: Ping timeout
jane has joined the channel
pascale has quit the server saying: Ping timeout
jane: (sorry - got booted out)
jane: I like the ability to enter a flat list of GO terms as an alternative to choosing or uploading a full slim file
jane: (sorry if we've already moved on to another topic!)
gwg: yes, we were on to the metaphysics of ontologies
gwg: anyway, I need to discuss with Chris and Seth about it, but I think ideally it's a lot easier for users not to have to have an OBO format GO slim to upload
pascale has joined the channel
eurie: I like how you've broken it up into sections.  There are three essential inputs for mapper: (1) input list of genes to be analyzed.  (2) go slim to be used as the terms to map to (3) the associations that are to be used for the mapping.  Maybe the page should be split into these sections?
jane has quit the server saying: Read error: Connection reset by peer
gwg: I was also wondering about other possible presentation methods, e.g. a list of tasks
gwg: so you could have 'create a mapping for all ontology terms using my input terms'
gwg: 'map my GA file terms to a goslim'
gwg: ... well, things like that, perhaps expressed more clearly
9:05 AM
gwg: the thing with map2slim (and similar scripts / amigo options) is that there are a number of options available, but presenting them to the user in a comprehensible manner isn't always easy
gwg: I think a list of 'tasks' would appeal to newbies, because it would make it clear for them, but then it might get annoying for seasoned users to have to take 'baby steps'
gwg: what do people think?
sjcarbon: i recall that people didn't much care for the ORB 'wizard'
eurie: For this first release, I don't think we necessarily need to shoot for the moon - just present simple instructions
rama: I agree
jane has joined the channel
gwg: it's trying to find a way to make it simple that's tricky, though!
eurie: Section 1: required, input a list of genes to be mapped
jane: (that was a power cut that time!)
eurie: Section 2: Pick a slim set (default: one of the go slims in the OBO file; advanced, your own list of GOID)
eurie: Section 3: Pick a gene association file to use for the analysis (default: the organisms in the database; advanced, upload your own gene_associations file)
rama: Did we decide on one ontology vs all three at a time?
jane: Section 1: or select an annotation file?
9:10 AM
jane: people might want just an overview of annotation for a given species?
eurie: yup, both rama's and jane's suggestions sound good too
rama: Jane- your idea is great. won't it take a long time to compute it on the fly?
rama: should we precompute and have that available?
eurie: maybe that's a summary page we couldl have pre-computed each time the database is loaded?
jane: GOA do that - they provide a pre-slimmed annotation set
rama: Regarding users entering terms- what if an user enters a bunch of terms from all three ontologies
jane: then it slims over all 3 ontologies?
eurie: maybe we should limit to 1 ontology per run
9:15 AM
rama: I don't know if it is going to be hard on the servers to do all 3 at a time
rama: I agree with Eurie
eurie: We've got ~15 minutes left.  Is this enough initial thoughts for the next rendition mapper?  should we move onto finder?  Seth/Amelia?
sjcarbon: moving on might be good for the time left
eurie: ok - amigo's finder: http://toy.lbl.gov:9002/cgi-bin/amigo/termfinder.cgi
eurie: princeton: http://go.princeton.edu/cgi-bin/GOTermFinder
eurie: sgd: http://db.yeastgenome.org/cgi-bin/GO/goTermFinder
eurie: i actutally really like princeton's "basic input" list 
rama: me too.
9:20 AM
gwg: re: slimming over 1 vs 3 ontologies - it is quicker just to do it over 1 ontology, but it doesn't take all that long to do a slim down on all the terms in the ontology - if you choose the bottom option of the list at http://toy.lbl.gov:9006/cgi-bin/amigo/goslimmer.cgi, you'll see it is quick(ish)
gwg: (sorry, took ages to type that reply!)
rama: we should also change the title to Term enrichment analysis or something meaningful
sjcarbon: choose a title and it will be done
eurie: but go term finder is so catchy.  
gwg: but it sounds like a search engine!
jane: How about 'find over-represented GO terms'
jane: or 'enriched GO term finder'
gwg: I think that sounds like areas of the ontology where we've got loads of detail
rama: okay settling on a title is easy. Lets get to the options on the interface (sorry!)
jane: I like the basic v/s advanced input sections too
eurie: I like having the evidence code options - but maybe checkboxes?  so you can see the full list?  maybe people will start actually paying attention to them!
9:25 AM
rama: checkboxes is a good idea
jane: I think we should just have maybe three evidence codes options
eurie: experimental, sequence-based, computational?
jane: e.g. experimental, all manual, manual + electronic
jane: yeah - exactly
pascale: I like that too; I think AmiGO should also have that 
rama: can we say experimental and Non-experimental
gwg: if we can decide on what to use, I can implement that pretty quickly
gwg: in the main AmiGO, that is
rama: we are combining type of evidence and type of annotation when we say experimental and manual etc. 
gwg: doesn't the new evidence code documentation split the evidence codes into three main groups?
eurie: Possibly the separation of evidence codes into major sections should be brought up tot he greater GOC in order to make it consistent across the board
rama: Yeap.
9:30 AM
jane: how about we just have 'manual v/s manual+electronic' and then in advanced options allow people to choose specific evidence codes?
rama: How about the results page for these interfaces
rama: Sorry I moved on since we are out of time
pascale: right, that seems like a big change
jane: has anyone seen the new evidence code docs?
rama: I have
gwg: am looking at them right now
gwg: they split evidence codes up into several categories:
gwg: experimental evidence codes
gwg: computational analysis
gwg: author statement
gwg: curator statement
gwg: and automatically assigned
jane: hmm - not very intuitive
gwg: ( http://www-dev.yeastgenome.org/draftGO/go/www/GO.evidence.new.shtml )
gwg: no - I wouldn't have thought the author/curator divide would be all that interesting to people outside GO
jane: I don't think those are good categories for AmiGO - they're almost as confusing as the evidence codes!
gwg: how about experimental / computational / statement?
eurie: ok, maybe this is a conversation we should have on the greater list - evidence code breakdown on evidence code docs and possibly transferring them to AmiGO.
eurie: and we can influence some change.  
rama: I agree
jane: me too
gwg: moi aussi
9:35 AM
eurie: sounds like there is enough food for thought for the next interface versions of finder and mapper?
eurie: any specific actions items we want to extract  from this meeting?
gwg: yeah
rama: How about results page?
eurie: maybe that can be a topic for discussion for the next awg?
rama: okay
gwg: I did a little mock up of the mapper front page from Eurie's description: http://toy.lbl.gov:9006/cgi-bin/amigo/goslimmer.cgi
gwg: or maybe we can discuss it on the mailing list (if it's not too controversial!)
eurie: Sending it back around to the mailing list sounds good
gwg: OK, perhaps I will email about it tomorrow?
gwg: I also want people to test the new search to see if the relevant results it produces are relevant or not
eurie: Since these tools are major leaps forward, Seth and Chris suggested sending these out to the greater GOC for comments once the AWG has come up with a version we like
gwg: good idea
9:40 AM
eurie: ok, we're all done here then?  look forward to the discussion of mapper over email then
gwg: I'm excited about it already!
jane: ok
gwg: before we all go
eurie: yes?
gwg: it's Jane's birthday tomorrow, so I'm sure you'd all like to join me in a (silent) rendition of 'Happy Birthday'!
eurie: HAPPY BIRTHDAY!!!
pascale: joyeux anniversaire jane! 
rama: Happy birthday Jane
gwg: perhaps we can give her the bumps, too! 
rama: Bye
gwg: see y'all on the mailing list
pascale: ciao 
eurie: bye
pascale has quit the server
sjcarbon: bye
gwg has left the channel
sjcarbon has left the channel
jane: thank you 
jane: (sorry, wasn't paying attention 
jane: see ya!
rama has left the channel
9:45 AM