beginner's questions on CellDesigner/SBGN/SBML

[Thomas Ligon (Hotmail)]

I have some beginner’s questions and put them in a pdf so that people using different systems can still see the graphics.  Basically, I am looking for

How to correctly define transcription (and translation) reactions in CellDesigner.

How to manipulate frozen anchor points in CellDesigner.

Maybe independent of CellDesigner, a recommendation for which software is today’s “standard” for creating diagrams (preferably SBGN) that can be exported as SBML.

 

Tom

(Dr. Thomas S. Ligon)

Thomas...@Hotmail.com

Frohnloher Str. 6a

81475 Muenchen

Germany

Tel. +49(89)74575075

 

[Felix Winter]

Hi Thomas, (CC to the SBGN discussion list as your questions are more related to
SBGN than to SBML)

I cannot answer all your questions but with respect to question 1 I would refer
you to the SBGN bricks site:

http://sbgnbricks.sourceforge.net/index.html

There you can find a list of "bricks" which show how to use SBGN to describe a
specific process. The bricks dictionary can be found here:

http://sbgnbricks.sourceforge.net/sbgnbricks_dictionary.html

In the transcription examples you can see that the protein is usually not
modelled as a substrate but as a modifier of the reaction which should make it
clear that it is not consumed. Another example can be found in figure 2.42 (The
creation of a messenger RNA X triggered by the gene X.) in the SBGN-PD user
manual which you can get here: http://www.sbgn.org/User_Page .

With respect to your second question I can only confirm that CellDesigner is
sometimes a little bit buggy. But in principle there should be no "frozen"
anchor points I guess.

Hope that helps,

Felix.



> On September 3, 2015 at 11:25 AM "Thomas Ligon (Hotmail)"

> <thomas...@hotmail.com> wrote:
>
>
> I have some beginner's questions and put them in a pdf so that people using
> different systems can still see the graphics. Basically, I am looking for
>
> How to correctly define transcription (and translation) reactions in
> CellDesigner.
>
> How to manipulate frozen anchor points in CellDesigner.
>
> Maybe independent of CellDesigner, a recommendation for which software is
> today's "standard" for creating diagrams (preferably SBGN) that can be
> exported as SBML.
>
>
>
> Tom
>
> (Dr. Thomas S. Ligon)
>

> <mailto:Thomas...@Hotmail.com> Thomas...@Hotmail.com

>
> Frohnloher Str. 6a
>
> 81475 Muenchen
>
> Germany
>
> Tel. +49(89)74575075
>
>
>

> --
> You received this message because you are subscribed to the Google Groups
> "sbml-discuss" group.

Felix Winter
Rostock, Germany

http://orcid.org/0000-0003-2987-6797

[Thomas Ligon (Hotmail)]

Hello Felix,

 

Many thanks for the detailed response.  I have replied to all because I value the discussion here, and hope that no one feels that I am spamming them. I would be happy to send the pdf and the model files to anyone who is interested, but will leave them out of the mails that go to everyone on the thread.

 

I will definitely take a closer look at SBGN bricks.  Do you recommend this as a replacement for CellDesigner?  The SBGN site lists so many software products that support it that I feel a bit overwhelmed and don't know which one to try. I was hoping to find a "standard" with the simple goal of producing usable graphics that are guaranteed to agree with the model.

 

When you say that the protein is usually modeled as a modifier instead of a substrate, I assume that you mean the gene, since there is no protein here. So, I have created a version where the empty set (“degraded” in CellDesigner) is transcribed into mRNA with help of the gene as a modifier. Yes, this gives me the right kinetics (in a "simple" model based on mass action). It looks like this:

Are we guessing that this is what the CellDesigner experts intended as usage of the transcription arrow?

 

PS: I will add this diagram to the pdf for the benefit of anyone who gets this email in plain text but still wants to see the graphics.

 

Tom

(Dr. Thomas S. Ligon)

Thomas...@Hotmail.com

Frohnloher Str. 6a

81475 Muenchen

Germany

Tel. +49(89)74575075

 

To unsubscribe from this group and stop receiving emails from it, send an email to sbml-discuss...@googlegroups.com.

To post to this group, send email to sbml-d...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/sbml-discuss/1058950413.5586094.1441373821742.JavaMail.open-xchange%40omgreatgod.store.

For more options, visit https://groups.google.com/d/optout.

[Robert Phair]

Hi Tom,

This is an interesting question. I agree with you and Felix that it does not make sense to connect gene directly to the transcribed mRNA. So your gene-as-modifier diagram is closer to the biology and makes more sense if your goal is a process map. One of the coolest things that SBGN has done is provide precision in the definition of three different kinds of maps: Process Descriptions, Activity Flows, and Entity Relationships. Most people creating diagrams, however, are not this precise and this leads to communication problems.

In our software, we only deal in process maps, so we face your question frequently and I've attached a screenshot showing a part of a bigger model that illustrates how we represent transcription and translation. The only thing this adds to the discussion is transcription factors. We would also emphasize, as you did, the importance of a complete and unambiguous mapping between the process map and the ODEs.

Because the gene is, as you and Felix emphasized, not consumed, we usually don't even include it as a species and instead focus on the gene's upstream promoters as the controllers of transcription. In the screenshot you will see promoters for the genes coding proteins named p21 and p27. These are both controlled by the transcription factor, NFkB. We represent the two promoter:NFkB complexes as species (the colon indicates a non-covalent complex), plus binding and unbinding reactions for the transcription factor. Each reaction is, as usual, represented as an arrow, and each arrow is labeled with the name of the process.

The NFkB:promoter complexes then are shown as activators of the transcription processes whose products are mRNAs. In turn, the mRNAs are used as activators of translation. You will see our glyphs are not as verbose as SBGN, but we think they capture all the required information and thus permit simulation of all the dynamics of genetic circuits. 

This discussion may be a little tangential to your question, but at least it offers another perspective.

There is a published model using this approach in a paper we did with the Melnick lab at USC, whhich I've attached as a link.

Good luck with your project.

Regards,

Robert

Robert D Phair, PhD  |  Chief Science Officer  |  Integrative Bioinformatics Inc

Mountain View, CA, USA

rph...@integrativebioinformatics.com

www.integrativebioinformatics.com/processdb.html

---

BMC Developmental Biology | Full text | Salivary gland branching ...

[Thomas Ligon (Hotmail)]

Thanks again to everyone who helped. Now I can create usable diagrams (in CellDesigner), export SBML, import the SBML into Copasi and create ODEs.  With that, it is guaranteed that cartoons, reactions, and ODEs all agree.  I like that!

 

Nevertheless, I still have a problem displaying SBGN in CellDesigner (error line 1 [Error(System)] : File unreadable.).

I also have difficulty formatting the reaction IDs in the diagram, but that may be a moot point.

 

In order to avoid sending too many files via email, I have deposited them here:

http://www.thomassligon.info/Pages/Research.aspx 

 

Question 4 – SBGN Error

When I select View/Convert to SBGN PD View, or File/Export SBGN ML and the Open/file just saved, I get the error

line 1 [Error(System)] : File unreadable.

Where can this come from?

 

I also tried View/Show SBGN Compliants (I assume this means SBGN Compliance), then nothing happens.

 

In summary, here is what I have learned from the recent discussion:

·         I have received recommendations to consider CellDesigner as the standard tool for creating graphics in systems biology, including SBGN diagrams.  Today, there are very many tools that support SBGN, and there is also a strong recommendation to use only those features that are available in SBGN.

·         For creating transcription and translation reactions in CellDesigner, it is better to avoid the (older) arrows for these reactions and simply use the state transition arrow.  This way, the result is compatible with SBGN and the older arrows don’t automatically create the correct kinetics, so the only value they add is a small visual hint.

·         For creating the correct kinetics for transcription and translation, it is possible to use “add product” or to use the gene (mRNA) as a modifier of the reaction.  The second is recommended, because the gene (mRNA) is not created or consumed in the reaction.

·         Things like anchor points that can’t be moved in CellDesigner are probably bugs, but fortunately they don’t occur often.

 

Tom

(Dr. Thomas S. Ligon)

Thomas...@Hotmail.com

Frohnloher Str. 6a
81475 Muenchen
Germany
Tel. +49(89)74575075

 

From: sbml-d...@googlegroups.com [mailto:sbml-d...@googlegroups.com] On Behalf Of Robert Phair
Sent: Thursday, September 10, 2015 6:26 AM
To: sbml-discuss <sbml-d...@googlegroups.com>
Cc: thomas...@hotmail.com
Subject: [sbml-discuss] Re: beginner's questions on CellDesigner/SBGN/SBML

 

<<Transcription translation diagram in ProcessDB.png>>

--

You received this message because you are subscribed to the Google Groups "sbml-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sbml-discuss...@googlegroups.com.
To post to this group, send email to sbml-d...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/sbml-discuss/ee28b46f-2e48-4756-946c-b963e255d8a2%40googlegroups.com.

[cmoul1]

Hi,

Sorry if my question isn't right.
I have a similar problem (I create something with CellDesigner, I save as .xml and I cannot open it with CellDesigner... and I have the error "line 1 [Error(System)] : File unreadable.")

Does this problem have a solution ?

Thank you and sorry for the mistakes.

Cecile Moulin