Time
S
Nick
Message
00:26 sivoais joined #dataverse
02:09 djbrooke joined #dataverse
02:48 bjonnh
is anyone aware of any project to document 96 well plates?
03:06 djbrooke joined #dataverse
10:54 pdurbin
bjonnh: are you trying to inventory what's in the wells? Or are you storing assay data? Or something else? (Hat tip to my wife for telling me what to ask!)
10:55 pdurbin
She's the one who's always playing God: https://twitter.com/philipdurbin/status/664645552353755136
11:50 djbrooke joined #dataverse
12:04 donsizemore joined #dataverse
13:08 bjonnh
pdurbin: both
13:11 bjonnh
I wanted to make something that allows to export a JSONish description of the plate to svg/html/so on
13:11 bjonnh
and manipulate it
13:11 pdurbin
bjonnh: I believe what you're looking for is called an electronic lab notebook (ELN) or a laboratory information management system (LIMS).
13:11 bjonnh
pdurbin: so God is using android… :p
13:11 bjonnh
pdurbin: yes but no ;)
13:12 bjonnh
pdurbin: I want a simple system that you can connect to other things easily
13:12 bjonnh
not a big mammuth
13:12 pdurbin
pameyer: know any biologists or biochemists?
13:13 bjonnh
especially something that could be integrated with systems such as isatools
13:13 pdurbin
bjonnh: my wife suggested using Excel for starters
13:13 bjonnh
easily in websites
13:13 bjonnh
pdurbin: yeah I know that, that's exactly what I want to move people from
13:13 bjonnh
:p
13:14 pdurbin
heh
13:14 bjonnh
Sorry I was not clear enough :p
13:14 pdurbin
bjonnh: have you ever heard of ResearchSpace or RSpace? https://github.com/IQSS/dataverse/blob/71b2c33b1261691a839fbb6b13f2c9970feb245e/doc/sphinx-guides/source/api/client-libraries.rst#java
13:16 bjonnh
pdurbin: I'm more talking low level
13:16 bjonnh
pdurbin: format/protocol, manipulation and then visualization
13:17 bjonnh
Also, I want people to still be able to use their data in some years from now
13:18 bjonnh
my problem is exactly all those ELN systems
13:18 bjonnh
with all their different formats, different approaches, different protocols
13:19 bjonnh
(note: my job is partly to make ontologies)
13:21 pdurbin
bjonnh: but wouldn't the goal be to have Dataverse and RSpace (or whatever ELN or LIMS) use the same ontology and formats?
13:22 bjonnh
yes
13:23 bjonnh
I need to see if displaying plates on dataverse would be useful though
13:23 bjonnh
not sure people are still using plate-level data at that point
13:24 pdurbin
what level are they using?
13:24 bjonnh
I see mostly lists
13:24 bjonnh
but I'm biased
13:25 bjonnh
that's part of my project to look in a more statistically relevant way
13:25 pdurbin
is any of what you're looking at public?
13:26 bjonnh
well there aren't plenty
13:27 bjonnh
let me find one
13:28 bjonnh
if I can
13:32 bjonnh
ok so maybe I can't see one in dataverse
13:33 pdurbin
bjonnh: go ahead and upload something :)
13:33 bjonnh
well
13:33 bjonnh
that's the plan at some point
13:33 bjonnh
:p
13:34 bjonnh
but what people do usually is that they convert the rectangular format to a list format
13:39 bjonnh
If I was dishonest I would say that this explain why what I want to do is important ;)
13:39 bjonnh
because people cannot do it easily :p
13:41 pdurbin
bjonnh: maybe for now you could just upload it to https://demo.dataverse.org
13:42 bjonnh
I don't have data I can upload right now
13:42 bjonnh
not mine
13:42 bjonnh
at least
13:42 pdurbin
oh, bummer
13:42 pdurbin
It's hard to know what you're talking about then.
13:42 bjonnh
yes
13:42 bjonnh
I'm trying to find an open dataset somewhere else
13:43 pdurbin
ah, good
13:45 bjonnh
but it is hard to find
13:45 bjonnh
because most of the time people reduce the data to results
13:46 bjonnh
raw plate data is usually not published
13:46 bjonnh
pdurbin: http://dev.lincs.hms.harvard.edu/wordpress/wp-content/uploads/2011/11/HMS-LINCS-ISA-Tab-201111181-e1441295598409.png
13:47 pdurbin
oh! LINCS!
13:47 pdurbin
pameyer: we've heard of them!
13:49 bjonnh
pdurbin: so many projects out there
13:50 djbrooke joined #dataverse
13:51 djbrooke joined #dataverse
13:52 djbrooke_ joined #dataverse
13:52 pdurbin
bjonnh: please check out the "Biomedical Dataverse: Datasets from the NIH LINCS Project" slides at http://projects.iq.harvard.edu/files/dcm2016/files/shamu_dataversemtg_7-11-16_copy.pptx by Caroline Shamu from http://projects.iq.harvard.edu/dcm2016/meeting-agenda
13:55 bjonnh
https://www3.appliedbiosystems.com/cms/groups/mcb_marketing/documents/generaldocuments/cms_039767.pdf
13:56 bjonnh
most machine software update data as a CSV file
13:56 bjonnh
with
13:56 bjonnh
A1, 0.1
13:56 bjonnh
A2, 0.3
13:56 bjonnh
etc
13:56 bjonnh
(can output many more values)
13:56 pdurbin
ok, are those slides helpful?
13:57 bjonnh
well they don't talk about the well-format to treated format
13:58 pdurbin
bjonnh: I think we should start with your real world data.
13:59 bjonnh
well that's the thing
13:59 pdurbin
without a user story...
14:00 bjonnh
we have to make well plans (for now it is a Spreadsheet that look like the last pdf I sent)
14:00 bjonnh
each well has a lot of data associated
14:01 bjonnh
I'll try to get an anonimized dataset from this person
14:02 pdurbin
bjonnh: perfect! Thanks!
14:02 bjonnh
https://github.com/craic/plate_maps
14:03 bjonnh
https://github.com/craic/plate_maps/blob/master/test/data/map_one_plate_384.csv
14:03 bjonnh
that's one of the data type
14:03 bjonnh
that associates sample name to physical position in the plate
14:04 bjonnh
then you obtain one or more results by well in a format similar to what you see here: https://qbrc.swmed.edu/download/SbacHTS_Manual.pdf
14:08 pdurbin
cool but where is the cvs with the T1Raw, T2Raw, T3Raw values?
14:09 bjonnh
what do you mean?
14:10 pdurbin
from that last pdf, the actual data
14:10 pdurbin
that biochemists care about
14:11 bjonnh
there are many
14:11 bjonnh
the output values are of many different type
14:11 bjonnh
and dimensions
14:12 bjonnh
sometimes reference to picture or videos, sometimes one float, sometimes hundreds of dimensions
14:12 bjonnh
there is no "one format" nor "one standard"
14:13 bjonnh
most of the time produced by machines, but sometimes by humans (visual analysis)
14:14 bjonnh
so instead of creating yet another format for describing what we are doing in the lab, I wanted to make something generic
14:14 bjonnh
that can be reused
14:15 pdurbin
cool
14:15 pdurbin
it sounds like a worthy project
14:15 bjonnh
maybe, I have to finish my study of what is existing first
14:16 pdurbin
I think the only way I can help is by getting you in touch with the RSpaces and researchers I know. I've invited some folks to read these logs and jump in. :)
14:16 bjonnh
great, thanks
14:17 bjonnh
pdurbin: https://upload.wikimedia.org/wikipedia/commons/5/5e/Microplates.jpg
14:17 bjonnh
I don't know if you have seen what it looks like in real
14:18 pdurbin
I've been to my wife's lab before. :)
14:19 bjonnh
what is she doing?
14:20 pdurbin
https://www.linkedin.com/in/teresa-broering-2464242
14:21 bjonnh
by the way did I told you that my lab now requires mandatory dataverse deposit of data involved in publications?
14:22 bjonnh
neat
14:22 bjonnh
pdurbin: I suppose they are using ELN in her lab?
14:23 pdurbin
yeah, it's in that tweet. Benchling
14:23 pdurbin
the big news is that the FDA just approved her first drug: http://www.mercknewsroom.com/news-release/corporate-news/fda-approves-mercks-zinplava-bezlotoxumab-reduce-recurrence-clostridium-
14:24 bjonnh
oh I was thinking it was just a plasmid maker
14:24 bjonnh
oh congrats then !
14:26 pdurbin
thanks, I'll pass it along. I already showed pameyer the PDB entry: http://www.rcsb.org/pdb/explore.do?structureId=4NP4 :)
14:32 donsizemore joined #dataverse
14:34 pameyer
bjonnh - sorry I'm a bit late to the conversation. it's sounding like you've got a mix of input formats for the plates (aka - sometimes instrument, sometimes experimenter/analysis tool/visualization tool).
15:58 rob-researchspac joined #dataverse
16:00 jhand joined #dataverse
16:02 dataverse-user joined #dataverse
16:03 rob-researchspac joined #dataverse
16:05 pdurbin
rob-researchspac: hi! welcome!
16:06 rob-researchspac
RSpace uses totally standard formats: pdf, html and xml. Pdurbin, One click export would make it very easy to get your plate data out. Export to dataverse is already built in. You can try it free at community.researchspace.com or I can give you a live demo to show you how easy it would be to use with 96 well plate data.
16:06 rob-researchspac
Its a really simple system
16:07 pdurbin
rob-researchspac: do you have public examples of plate data already?
16:07 rob-researchspac
email me if you like and I'll give you a webcast demo right now if you are available. rob
16:08 pdurbin
rob-researchspac: sorry, I'm kind of heads down in fixing bugs for pameyer but maybe bjonnh would like a demo
16:11 rob-researchspac
Anyone here is welcome to contact me
16:13 pdurbin
rob-researchspac: much appreciated! And you're saying the integration with Dataverse is already live?
16:16 rob-researchspac joined #dataverse
16:18 rob-researchspac
dataverse integration is technically in advanced prototype stage, but will be available for all users with next release. RSpace is evolving at very fast pace, with new releases and features every month.
16:21 pdurbin
rob-researchspac: is sign in via Google working? I just got this: Signin could not complete: Status:500 - server error There was an error responding to this request, which has been logged. Something went wrong: No account found for email philipdurbin
16:23 karissa joined #dataverse
16:34 rob-researchspac
I think you need to create an account first and list a gmail address as your official email for that account, THEN you can subsequently login with google
16:35 rob-researchspac
ah, nope, that's not true! you can just use whatever google account you are currently logged in with
16:36 rob-researchspac
I just tested it ... it works fine even for a totally new google accont
16:36 rob-researchspac
no "pre-register" is needed.
16:36 rob-researchspac
I suggest rechecking your google credentials,
16:38 pdurbin
rob-researchspac: I get the same thing for a different google account: "Something went wrong: No account found for email pdurbin
16:38 rob-researchspac
so... you have TWO login options.... you can manually make a new account with any email address you like, OR just click "login with google" and it will create a new account for you based on whatever google ID your browser is currently using.
16:43 pdurbin
ok
16:45 bjonnh joined #dataverse
16:46 skay joined #dataverse
16:50 pdurbin
karissa: hi! Welcome! Did you catch some of the chatter at https://botbot.me/freenode/dat/msg/75629136/ ?
16:50 pdurbin
chatter with jhand. Welcome to you too! :)
17:17 donsizemore joined #dataverse
17:22 djbrooke joined #dataverse
17:39 djbrooke joined #dataverse
17:39 bret joined #dataverse
18:30 bjonnh
pameyer: yep
18:30 bjonnh
oh rob left
18:32 pameyer
I'm guessing there aren't format standards for "put complex, possibly structured, data into a 2d grid"
18:43 bjonnh
so I may have to make one
18:43 bjonnh
but that's what I'm trying to figure out
18:43 bjonnh
if I need to make one or if something akin to it exist
18:45 pameyer
I haven't done much w\ 96 well plates for a few years, and then it was liquid handling robots instead of experimental measurements. but there wasn't anything standardized then
18:45 djbrooke joined #dataverse
19:41 djbrooke joined #dataverse
19:41 pdurbin
mmm, robots
19:42 pdurbin
jhand: nice presentation! I'm looking forward to your follow up talk on Monday! http://www.iq.harvard.edu/event/tots-tip-4
19:42 pdurbin
pameyer: you should come to this. It has DCM and DLM written all over it
19:46 djbrooke joined #dataverse
19:50 djbrooke joined #dataverse
19:54 pameyer
pdurbin: the tots-tip thing?
19:55 pameyer
DATS has come up across my radar a few times; if things aren't too crazy I'll try to make it
19:57 pdurbin
pameyer: cool. and it's just "Dat"
19:57 pdurbin
pameyer: I'm gonna show jhand https://data.sbgrid.org
19:57 pameyer
cool
19:58 pameyer
pdurbin (/jhand) : do you have a link for "Dat"?
19:58 pameyer
nm - found it
19:59 pameyer
I'll be back in a bit (if there are data.sbgrid.org questions I should field)
20:03 pdurbin
pameyer: I should know this but is the DLM powered by rsync?
20:11 djbrooke joined #dataverse
20:13 djbrooke joined #dataverse
20:14 djbrooke joined #dataverse
20:17 djbrooke joined #dataverse
20:22 jhand
pameyer: hopefully I'll see you Monday. Happy to answer any questions in meantime, I'm in here or in #dat.
20:28 djbrooke joined #dataverse
20:34 djbrooke joined #dataverse
20:38 djbrooke joined #dataverse
20:45 djbrooke joined #dataverse
21:34 djbrooke joined #dataverse
