SEP 016 -- Experimental implementations and replicates

[graik]

We propose to introduce a new class, tentatively called Implementation, to represent different replicates of physical design realizations. In practice, Implementation will identify cell clones or experimental batches of material. Implementation allows to collate experimental results as they are recorded for one particular clone or biological replicate. Implementation will serve as the connection point between experimental results with theoretical design and a, often more detailed, description of the system as it was actually built.

The SEP is being worked out in a fork at https://github.com/graik/SEPs/blob/master/sep_016.md

[MadraghRua]

Hi Guys
This looks like a pretty worthy SEP. So some clarifying questions:

1. As I move from a wild type cell line to an edited cell line are you tracking validated clones or are your trying to track everything. So as an example: I hit a population of cells with Cas9/gRNAs. My goal is to get a defined clonal population. So I flow sort, grow, assay and work out the heterogeneity of the well population. As I then find high efficiency wells I will resort, replate and then do sequence verification. Eventually I might identify three or four clonal populations with slightly different sequence verified changes. I will probably verify the phenotype changes - eg mRNA depressed, protein not found, etc. How much of this does this SEP intend to track?
2. As I work with developing several clones as possible circuit designs, I would track assembly information, sequencing information, expression information, circuit characterization maybe metabolic modeling if appropriate. I would probably also be tracking the starting parts and intermediate and final clones with bar codes. How much of this do you intend the SEP to track?
   Please let me know how I might help/contribute.
   kevin

[graik]

Hi Kevin,
there is not yet a clear answer to (1). Intuitively I would say it would make sense to track the final 4 clonal populations (each of them one Implementation instance, all pointing to the original design) because this is where you start accumulating clone-specific experimental results. But the level of detail and granularity captured will depend on software tools and users (and community norms?).

(2) Great example. I think the aim is to extent SBOL to the point where all this can be tracked. We want to keep this particular SEP minimal though and rather approach this in several SEPs. Bryan is working on how to follow a complete design-build-test cycle. The other missing piece is an attachment class for Experimental data (files).

Your help is very welcome!
Raik

[MadraghRua]

OK, where would you like me to start?

[nroehner]

Hi Kevin, I think providing examples like those in your first mail (but with some additional details to flesh them out) would help. I'm working on explaining the class diagram for Implementation and the object diagrams for existing examples provided by Jake in the sep16 branch of Raik's repo (here <https://github.com/graik/SEPs/blob/sep16/sep_016.md>). Best, Nic

…

On Tue, Sep 26, 2017 at 4:06 PM, MadraghRua ***@***.***> wrote: OK, where would you like me to start? — You are receiving this because you were assigned. Reply to this email directly, view it on GitHub <#35 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AApI_7-wgKfHS9VTpPrdU5sVy_3NXtOMks5smVkygaJpZM4PL2AS> .

[MadraghRua]

Have you folk looked at OBI for help with describing experiments? I was reading through Bryan's SEP14 and the ensuing discussions. It looks to me like OBI could be a viable way of providing Ontology terms for experiments, protocols and samples that would complement what the PROVO ontology would bring in. Just a thought.
OK, so I'll send in some examples of experiments I would want to track over time and submit these to you. text/pictures/both?

[cjmyers]

Please be aware the SEP14 is discontinued. It has been replaced with SEPs 16, 17, and 18 that split the concerns into three separate proposals: one for the implementation class, one for the PROV-O annotation of design-build-test, and one for the attachment class. We really need to mark SEP14 as abandoned somehow. I regret that we did not replace it with 17 in the first place, since this would have been more clear, I think. Nic: please pull in the drafts for 17 and 18, so others can see them.

…

On Sep 26, 2017, at 2:28 PM, MadraghRua ***@***.***> wrote: Have you folk looked at OBI for help with describing experiments? I was reading through Bryan's SEP14 and the ensuing discussions. It looks to me like OBI could be a viable way of providing Ontology terms for experiments, protocols and samples that would complement what the PROVO ontology would bring in. Just a thought. OK, so I'll send in some examples of experiments I would want to track over time and submit these to you. text/pictures/both? — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#35 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ADWD9wLWlMUv228VW1IEKWJ2tsvPWZ-Pks5smV5fgaJpZM4PL2AS>.

[graik]

I've labelled SEP 14 in the issue tracker as "withdrawn". We should also change the status in the document itself. It is in my view preferable to also conserve proposals that ultimately turn out to be not successful because there is something to be learned from those as well. Greetings Raik

…

On Wed, Sep 27, 2017 at 1:41 AM, cjmyers ***@***.***> wrote: Please be aware the SEP14 is discontinued. It has been replaced with SEPs 16, 17, and 18 that split the concerns into three separate proposals: one for the implementation class, one for the PROV-O annotation of design-build-test, and one for the attachment class. We really need to mark SEP14 as abandoned somehow. I regret that we did not replace it with 17 in the first place, since this would have been more clear, I think. Nic: please pull in the drafts for 17 and 18, so others can see them. > On Sep 26, 2017, at 2:28 PM, MadraghRua ***@***.***> wrote: > > Have you folk looked at OBI for help with describing experiments? I was reading through Bryan's SEP14 and the ensuing discussions. It looks to me like OBI could be a viable way of providing Ontology terms for experiments, protocols and samples that would complement what the PROVO ontology would bring in. Just a thought. > OK, so I'll send in some examples of experiments I would want to track over time and submit these to you. text/pictures/both? > > — > You are receiving this because you are subscribed to this thread. > Reply to this email directly, view it on GitHub < #35 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ ADWD9wLWlMUv228VW1IEKWJ2tsvPWZ-Pks5smV5fgaJpZM4PL2AS>. > — You are receiving this because you were assigned. Reply to this email directly, view it on GitHub <#35 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABxs3R696E64fPSXnloA5AsBBrZ849Ohks5smYvEgaJpZM4PL2AS> .

--

___________________________________ Raik Grünberg http://www.raiks.de/contact.html

___________________________________

[MadraghRua]

Clone constructs being used over time.docx

Hi Folks
Please let me know if this approaches what you need.
kevin

[graik]

Sorry for having let this slip. I had to deal with a transcontinental move and other issues.
In the meantime, Chris and Bryan are promoting their "design-build-test" SEP 19 which is, in contrast to the original plan, now written to replace this one but is, in my subjective view, doing a very poor job at it.

@MadraghRua , @jakebeal please read SEP 19 and speak up on the mailing list. I think their complete focus on generic providence transactions is overly complicated and missing the point of why we need to be able to represent biological replicates.

[cjmyers]

Raik, you are completely mischaracterizing the situation. Implementation class proposed was developed by Nic and Bryan in consultation with Jake. The Design-Build-Test was developed by Bryan in consultation with Anil and James at Newcastle. I’ve simply been providing comments and suggestions for improvement. Indeed, my last message actually stated that I agreed with you that the Implementation class needed a link to the the design/build, but the current two links should be merged into one with a different, more meaningful name. I am though supportive of the fact that Bryan has been urging which is that the use of prov-o to capture the entire design-build-test workflow should be very strongly encouraged. Is key to getting to our long-term goal of completely capturing everything done in a synthetic biology experiment to promote reproducibly.

…

On Nov 5, 2017, at 1:44 AM, Raik Grünberg ***@***.***> wrote: Sorry for having let this slip. I had to deal with a transcontinental move and other issues. In the meantime, Chris and Bryan are promoting their "design-build-test" SEP 19 which is, in contrast to the original plan, now written to replace this one but is, in my subjective view, doing a very poor job at it. @MadraghRua <https://github.com/madraghrua> , @jakebeal <https://github.com/jakebeal> please read SEP 19 and speak up on the mailing list. I think their complete focus on generic providence transactions is overly complicated and missing the point of why we need to be able to represent biological replicates. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#35 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ADWD99eunGZFZJxaKpc_28qnihvY-yipks5szXVygaJpZM4PL2AS>.