January 2013 - Microbe - mailman3.cels.anl.gov

Workspace Browser
by Paul Frybarger 30 Jan '13

30 Jan '13

Hey all, So here's the workspace browser, as it currently stands: http://140.221.92.49/ You can view all the workspaces you own, as well as the ones for which you have permission. Most of the workspace management functions are implemented, for instance changing the global permission, deleting, etc... There is a problem with giving other users permission that I hope to figure out today, so that feature is not working yet. You can also create new workspaces, and soon will be able to clone other's workspaces. For dealing with objects, right you can only view object metadata and download the object data in raw form. I'm not sure how much I'll be able to get done before the build next week, but I hope to have object management (delete, copy, move) and a better viewing system done by the demo. Please let me know about any questions or comments. I will send this prototype out to the dev list tomorrow, but wanted to see what you guys think first. There is still some work I need to do regarding styling, error handling, and workspace selection, but I think it's in decent shape. Thanks, Paul

1 0

Re: [Microbe] call today 11am P / 1pm C
by Christopher Henry 28 Jan '13

28 Jan '13

Hi all, I've got a conflicting call, so I can only attend today for the first 10 minutes. Chris On Jan 28, 2013, at 11:49 AM, Paramvir Dehal <psdehal(a)lbl.gov> wrote: > Everyone's really busy, let's have a quick call. If you need help on anything, now is a good time to ask. > > 866-740-1260 Code: 4958182 > _______________________________________________ > Microbe mailing list > Microbe(a)lists.kbase.us > https://lists.kbase.us/mailman/listinfo/microbe

1 0

call today 11am P / 1pm C
by Paramvir Dehal 28 Jan '13

28 Jan '13

Everyone's really busy, let's have a quick call. If you need help on anything, now is a good time to ask. *866-740-1260 Code: 4958182 *

1 0

Re: [Microbe] Modeling bootcamp
by Christopher Henry 25 Jan '13

25 Jan '13

Hello everyone, The first modeling bootcamp went pretty well I think, although I ended up talking the whole time so maybe others feel differently. I will strive to make this much more interactive and "workshop" like. This time, I was trying to cover as much of the tools and vision as possible. We made a video of my terminal and audio the entire time. Those videos, and the commented text from my terminal from the bootcamp are here: http://bioseed.mcs.anl.gov/~chenry/KBaseVideos/ModelingBootcamp/ I'll add the text from Sam's captionist as well, as soon as it's available. Again, please send comments and suggestions. It didn't go perfectly, and not everything is quite ready or a polished product, but this was a decent portrayal of the fulfillment of much of what I envisioned for the KBase microbes demo, so if I'm way off the mark on something, I want to know. We probably shouldn't share this first video with the community… I wasn't being careful enough with what I said to make this public. Also, there were bugs and flaws. The plan right now is to hold this bootcamp again on the following dates: 2/8: ~1-3PM 2/15: ~1-3PM 2/22: ~1-3PM The goal is to polish the tools, flag problems, identify areas for improvement, and improve the look, feel, and presentation of the workshop. This will evolve into the modeling demo we give at the GTL meeting. Sam already provided some great feedback. On the 15th or the 22nd, I think we'll produce videos and transcripts that are worthy of being shared and disseminated to the scientific community. For example, but then, we'll be running the whole thing in IRIS. I will open up the invitation for people to join the remaining bootcamps via google hangout. This time was a bit of a select crowd (microbial people who really need to see this workshop). Enjoy the videos and script (beware, it's two hours long). Hopefully we can build a tutorial from this. Best, Chris On Jan 24, 2013, at 10:33 PM, Christopher Henry <chrisshenry(a)gmail.com> wrote: > Hello all, > > I'm going to hold a little prototype modeling bootcamp tomorrow 1-3PM. We'll be going over the modeling tools in KBase. We'll hold it on google hangout. I'll also be recoding my screen in video, and Sam's captionist will be there (meaning we'll get a meeting transcript in writing). I'll also record the screen text of the demo. I'm hoping some of you can make it. This will be a documentation boon. > > This will be the first of many such bootcamps, which I will probably hold on Fridays around the same time every week until the demo (except next week, when I'll be at Notre Dame). > I expect this to evolve into our modeling demo on February 25th. > > I don't expect it to go smoothly. If you want to attend one that goes really smoothly, don't come to this one… or the next one. But I do hope a core of you will come with comments that will help us to polish these services to make them "production ready". > (strangely, the students in my thermo class this year didn't like it so much when I said the same thing to them) > > If you can't make it because of time conflicts, then let me know. I'll adjust the specific time of the bootcamp to accommodate people. > > If you plan to come, you will want to have the modeling and workspace scripts running on your machine before hand. These commands will go into IRIS, but I don't think we've got them quite yet (we are very very close… definitely by the next boot camp). > So you have two options. Follow these instructions to install my little client package (not the official KBase client, although it's been working well and I think there are things we can learn from this install): > https://github.com/ModelSEED/KBaseClient/blob/master/README.md > > If you really don't want to bother with an install, use this server: > login: kbase @128.135.250.215 > pw: fbaserv!1! > > Don't use that unless you are truly desperate. > So the plan for tomorrow is to run through using a workspace, loading a genome, building a model, running fba, gap filling a model, loading media, loading phenotype data, simulating phenotype data, reconciling a model. > > We got Pam's genomes (153 of those), the wild type shewy data, and b. subtilis data to play with. Got a lot of CDM models (100 right now). You should get a clear picture of what we've got emerging. Hopefully it's a picture that's nice for you. > If it's not, hopefully you got good constructive comments on things we can do quickly to make it better. > > Chris > > > >

1 0

Microbes Call today
by Paramvir Dehal 22 Jan '13

22 Jan '13

We have a Microbes Call today in 10minutes, 11am P / 1pm C. Call in if you can. 1-866-740-1260, 4958182#

1 0

Microbes Call Tuesday 1/22
by Paramvir Dehal 17 Jan '13

17 Jan '13

As Gavin pointed out, Monday is a holiday. Let's have our Microbes call on Tuesday at 11am Pacific / 1pm Central 1-866-740-1260 Code: 4958182# Paramvir

1 0

since next monday is a holiday, should we have a microbes meeting on Tues?
by Gavin A Price 16 Jan '13

16 Jan '13

A lot going on, seems like we should meet to me... -g

1 0

Re: [Microbe] experiment service implemented and proposed specs
by Gavin A Price 16 Jan '13

16 Jan '13

> So the question is whether this is/will be the way to query and > retrieve fitness data within KBase? Yes. > but its a bit odd to tackle these without having any generic > functionality in place. It comes down to that there's only so much time before the demo and I have a lot of things to do, so if it's not required for the demo I'm not doing it. > Based on the current structure I should expect a function in this spec > a la 'returnFitnessDataForBiclustering' and > 'returnExperimentDataForBiclustering'? Even 'returnTabData' or > 'returnDataMatrix' would be fantastic. Is this necessary for the demo? If so, if you have a use case write it up and I'll see what I can do. If not, do the same but don't bother sending it to me until Mar 1 :P -g On 1/16/2013 3:14 PM, Marcin Joachimiak wrote: > > >> Do you have a short spec in plain English? > I'm not convinced that'd be helpful. The point of a typespec is > that it describes the API with related free text docs. > > Really I need comments from the people that are going to be > using/representing this data - e.g. Matt, Ben, Chris, Bill, > Paramvir. I suppose I shouldn't really send it out to the Microbes > list as a whole, but that's what I said I'd do in the meeting... eh. > > > Oh, but I do need to use and represent fitness and expression data > within KBase which is why this service grabbed my attention. So the > question is whether this is/will be the way to query and retrieve > fitness data within KBase? I understand that you are targeting very > specific things for the demo, but its a bit odd to tackle these > without having any generic functionality in place. > > Based on the current structure I should expect a function in this spec > a la 'returnFitnessDataForBiclustering' and > 'returnExperimentDataForBiclustering'? Even 'returnTabData' or > 'returnDataMatrix' would be fantastic. > > p.s. Looks like I sent an empty msg by accident moments ago, gmail > can't undo those apparently.

1 0

Re: [Microbe] experiment service implemented and proposed specs
by Marcin Joachimiak 16 Jan '13

16 Jan '13

> > > Do you have a short spec in plain English? > > I'm not convinced that'd be helpful. The point of a typespec is that it > describes the API with related free text docs. > > Really I need comments from the people that are going to be > using/representing this data - e.g. Matt, Ben, Chris, Bill, Paramvir. I > suppose I shouldn't really send it out to the Microbes list as a whole, but > that's what I said I'd do in the meeting... eh. > Oh, but I do need to use and represent fitness and expression data within KBase which is why this service grabbed my attention. So the question is whether this is/will be the way to query and retrieve fitness data within KBase? I understand that you are targeting very specific things for the demo, but its a bit odd to tackle these without having any generic functionality in place. Based on the current structure I should expect a function in this spec a la 'returnFitnessDataForBiclustering' and 'returnExperimentDataForBiclustering'? Even 'returnTabData' or 'returnDataMatrix' would be fantastic. p.s. Looks like I sent an empty msg by accident moments ago, gmail can't undo those apparently.

1 0

Re: [Microbe] experiment service implemented and proposed specs
by Marcin Joachimiak 16 Jan '13

16 Jan '13

On Wed, Jan 16, 2013 at 2:54 PM, Gavin A Price <gavinaprice(a)gmail.com>wrote: > From the service name a naive user would assume something generic for > experiment data and/or meta-data but this seems customized to FBA/KO strain > data and analysis? Shouldn't that be in a more specific service aka > 'FitnessDataforFBA' or 'FunctionalGenomicsDataForFBA'? > > It will be the former, up until the demo it's basically the latter. > > > What is 'maxKnockouts'? The number of replicate KO strains? > > The maximum knockouts a strain can possess if it is to be returned as part > of a dataset. > > > Do you have a short spec in plain English? > > I'm not convinced that'd be helpful. The point of a typespec is that it > describes the API with related free text docs. > > Really I need comments from the people that are going to be > using/representing this data - e.g. Matt, Ben, Chris, Bill, Paramvir. I > suppose I shouldn't really send it out to the Microbes list as a whole, but > that's what I said I'd do in the meeting... eh. > > -g > > > > On 1/16/2013 2:44 PM, Marcin Joachimiak wrote: > > Hi Gavin, > I take it you'll be happy with any comments at this point ... > > Do you have a short spec in plain English? > > From the service name a naive user would assume something generic for > experiment data and/or meta-data but this seems customized to FBA/KO strain > data and analysis? Shouldn't that be in a more specific service aka > 'FitnessDataforFBA' or 'FunctionalGenomicsDataForFBA'? > > What is 'maxKnockouts'? The number of replicate KO strains? > > Beyond that I can't comment since I don't know what this is supposed to > be for exactly. > > marcin > > > On Wed, Jan 16, 2013 at 2:06 PM, Gavin A Price <gavinaprice(a)gmail.com>wrote: > >> So is everyone ok with these specs? I'm working on something else on the >> moment so I'm not losing time, but I'm a bit surprised that no one has >> commented. >> >> -g >> >> >> On 1/14/2013 12:16 PM, Gavin A Price wrote: >> >>> I should point out that the methods you're going to want to look at are >>> generateGrowmatchInput and generateAndStoreGMInput. The others aren't >>> really controversial afaik. >>> >>> -g >>> >>> On 1/14/2013 12:14 PM, Gavin A Price wrote: >>> >>>> As discussed in the meeting this morning. >>>> >>>> -g >>>> >>> _______________________________________________ >> Microbe mailing list >> Microbe(a)lists.kbase.us >> https://lists.kbase.us/mailman/listinfo/microbe >> > >

1 0