ecell-devel Mailing List for E-Cell System

Hi all,

I have built rpms of libsbml and ecell3 for Mandrakelinux cooker (the
development version, I hope the packages will be made available for the
next, soon-to-be-released stable version 10.1). Libsbml uses xerces-c++
(already available for Mandrakelinux).

Those packages will be available in the "contributions" tree of official
Mandrakelinux FTP mirrors, but will make them available also on my
personal site for those interested:
http://www.ebi.ac.uk/~efernand/index.html

I shall maintain them regularly now.

Eric

Hi,

good point. It doesnt contain Xerces-c, but I have
created a xerces-c-2.3 rpm a few months ago. I have
uploaded that one as well as the spec file.
Gabor


 --- Tatsuya Ishida <ee...@sf...> wrote: 
> Hi,
> 
> It's great!
> I thought that the libsbml rpm for ecell3 was my
> work.
> 
> I have a question.
> The libsbml requires the Xerces-C++ XML library.
> Does the libsbml you have created contain the
> Xerces?
> Or, do I have to install it for using the libsbml? 
> 
> I'll check it and report to you.
> 
> --
> Tatsuya Ishida
> 
> 
> > -----Original Message-----
> > From: ece...@li... 
> > [mailto:ece...@li...]
> On Behalf Of 
> > Gabor Bereczki
> > Sent: Sunday, August 29, 2004 11:48 PM
> > To: ece...@li...
> > Subject: [Ecell-devel] libsbml rpm
> > 
> > 
> > Hi,
> > 
> > I have created libsbml-2.1.0 rpms with python
> bindings
> > for python 2.3. It should run without problems on
> FC2.
> > I put it out on cs0:/home/gabor/libsbml.
> > If anybody interested, please check it out and
> test
> > it.
> > If it is working maybe should be put out to ecell
> > webpage. I also included a working
> > gnome-python2-canvas-2.0.2 for FC2.
> > Is there going to be a windows release soon?
> > Gabor
> > 
> > 
> > 	
> > 	
> > 		
> >
>
___________________________________________________________ALL
> > -NEW Yahoo! Messenger - all new features - even
> more fun!  
> > http://uk.messenger.yahoo.com
> > 
> > 
> > 
> >
>
-------------------------------------------------------
> > This SF.Net email is sponsored by BEA Weblogic
> Workshop
> > FREE Java Enterprise J2EE developer tools!
> > Get your free copy of BEA WebLogic Workshop 8.1
> today. 
> >
>
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
> > 
> > _______________________________________________
> > Ecell-devel mailing list
> > Ece...@li... 
> >
>
https://lists.sourceforge.net/lists/listinfo/ecell-devel
> > 
> 
> 
> 
> 
>
-------------------------------------------------------
> This SF.Net email is sponsored by BEA Weblogic
> Workshop
> FREE Java Enterprise J2EE developer tools!
> Get your free copy of BEA WebLogic Workshop 8.1
> today.
>
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
> _______________________________________________
> Ecell-devel mailing list
> Ece...@li...
>
https://lists.sourceforge.net/lists/listinfo/ecell-devel
>  


	
	
		
___________________________________________________________ALL-NEW Yahoo! Messenger - all new features - even more fun!  http://uk.messenger.yahoo.com

Hi,

It's great!
I thought that the libsbml rpm for ecell3 was my work.

I have a question.
The libsbml requires the Xerces-C++ XML library.
Does the libsbml you have created contain the Xerces?
Or, do I have to install it for using the libsbml? 

I'll check it and report to you.

--
Tatsuya Ishida


> -----Original Message-----
> From: ece...@li... 
> [mailto:ece...@li...] On Behalf Of 
> Gabor Bereczki
> Sent: Sunday, August 29, 2004 11:48 PM
> To: ece...@li...
> Subject: [Ecell-devel] libsbml rpm
> 
> 
> Hi,
> 
> I have created libsbml-2.1.0 rpms with python bindings
> for python 2.3. It should run without problems on FC2.
> I put it out on cs0:/home/gabor/libsbml.
> If anybody interested, please check it out and test
> it.
> If it is working maybe should be put out to ecell
> webpage. I also included a working
> gnome-python2-canvas-2.0.2 for FC2.
> Is there going to be a windows release soon?
> Gabor
> 
> 
> 	
> 	
> 		
> ___________________________________________________________ALL
> -NEW Yahoo! Messenger - all new features - even more fun!  
> http://uk.messenger.yahoo.com
> 
> 
> 
> -------------------------------------------------------
> This SF.Net email is sponsored by BEA Weblogic Workshop
> FREE Java Enterprise J2EE developer tools!
> Get your free copy of BEA WebLogic Workshop 8.1 today. 
> http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
> 
> _______________________________________________
> Ecell-devel mailing list
> Ece...@li... 
> https://lists.sourceforge.net/lists/listinfo/ecell-devel
> 

Hi,

I have created libsbml-2.1.0 rpms with python bindings
for python 2.3. It should run without problems on FC2.
I put it out on cs0:/home/gabor/libsbml.
If anybody interested, please check it out and test
it.
If it is working maybe should be put out to ecell
webpage. I also included a working
gnome-python2-canvas-2.0.2 for FC2.
Is there going to be a windows release soon?
Gabor


	
	
		
___________________________________________________________ALL-NEW Yahoo! Messenger - all new features - even more fun!  http://uk.messenger.yahoo.com

Hi,

could anyone give me some hints about how
ExpressionFlux process' Expression property can be
coded?
How can I refer to variables there?
Is there some sample code?
Gabor


	
	
		
___________________________________________________________ALL-NEW Yahoo! Messenger - all new features - even more fun!  http://uk.messenger.yahoo.com

> Hi
> >
> > In general, using specialized processes for each reaction mechanisms
> > yields  best simulation speed, but considering the potential number of
> > reaction mechanisms and equations (well over 100, I guess), I would
> > recommend considering use of ExpressionFluxProcess (described in the
> > manual). With that Process class you can put any equation by yourself.
> >
> >> Also, I shall try to complete the wiki, such info is useful for users.
> >
> > great!  Thanks for your contributions.

Tatsuya Ishida and I have rewritten ExpressionFluxProcess last month,
using the state-of-the-art techniques for its Stack Virtual Machine
(SVM), including stack operations folding.   It is now about 3x-4x
faster than before (just 1.5x-3x slower than native C++).   It will be
included in the next version.  Stay tuned.      

Tatsuya is currently working on a JIT (Just-In-Time compiler) version
that generates assembly instructions on the fly for further performance.


-sha

Hi
>
> In general, using specialized processes for each reaction mechanisms
> yields  best simulation speed, but considering the potential number of
> reaction mechanisms and equations (well over 100, I guess), I would
> recommend considering use of ExpressionFluxProcess (described in the
> manual). With that Process class you can put any equation by yourself.
>
>> Also, I shall try to complete the wiki, such info is useful for users.
>
> great!  Thanks for your contributions.
>
>
> -sha

Thanks a lot for these explanations. I shall try this.
Eric

Hi eric,


> > If you have Segel's Enzyme Kinetics, open page 29.  Everything is there.

> Thank you very much, Sha.
> I was wondering where I could find documentation about the available
> processes in ecell and their parameters. 

The users manual should describe all the Process classes, but it is
still incomplete.   I'm sorry, but I need 36 hours a day.

Emily Wang and Takeshi Sakurada are now editing E-Cell Tutorial, which I
believe has a fairly user-friendly and complete coverage of that topic.
I think it will become ready for download soon.

> Is this book useful for this ?

IMO, this is a classic, and a must-have when doing biochemical
simulations :)

> For example, I try to add in my model (among other michaelis processes) a
> simple inhibition like this :
> I + E <--> IE
> which is not giving any product (it is just an association to an enzyme
> which prevents it to work) and was wondering which process is best for
> this.

It depends on if you can assume steady state of the enzyme-substrate
complex.     When you use a Michaelis-Menten equation it is effectively
converting the meaning of [E] into [E]_total = [E] + [ES] (plus any
other possible complex forms).   In this case you cannot simply add a
separate reaction that uses [E] (such as I + E <-> IE).   You need to
include everything into a single long equation.  This is why there are a
huge number of variations of Michaelis Menten.

In general, using specialized processes for each reaction mechanisms
yields  best simulation speed, but considering the potential number of
reaction mechanisms and equations (well over 100, I guess), I would
recommend considering use of ExpressionFluxProcess (described in the
manual). With that Process class you can put any equation by yourself. 

> Also, I shall try to complete the wiki, such info is useful for users.

great!  Thanks for your contributions.


-sha



> Cheers
> Eric
> 
> 
> -------------------------------------------------------
> SF.Net email is sponsored by Shop4tech.com-Lowest price on Blank Media
> 100pk Sonic DVD-R 4x for only $29 -100pk Sonic DVD+R for only $33
> Save 50% off Retail on Ink & Toner - Free Shipping and Free Gift.
> http://www.shop4tech.com/z/Inkjet_Cartridges/9_108_r285
> _______________________________________________
> Ecell-devel mailing list
> Ece...@li...
> https://lists.sourceforge.net/lists/listinfo/ecell-devel
-- 

>
> If you have Segel's Enzyme Kinetics, open page 29.  Everything is there.
>
>
> regards,
> -sha

Thank you very much, Sha.
I was wondering where I could find documentation about the available
processes in ecell and their parameters. Is this book useful for this ?

For example, I try to add in my model (among other michaelis processes) a
simple inhibition like this :
I + E <--> IE
which is not giving any product (it is just an association to an enzyme
which prevents it to work) and was wondering which process is best for
this.

Also, I shall try to complete the wiki, such info is useful for users.

Cheers
Eric

Hi Eric,

Ok, there can be two representation of the uniuni reaction mechanism.

One central complex representation:

         k1       k2
E + S <---> ES <---> E + P                   ...(1)
      k-1      k-2

and the one which has two central complexes

         k1       k2       k3
E + S <---> ES <---> EP <---> E + P          ...(2)
      k-1      k-2      k-3


Interestingly, for both representations, exactly the same equation 
can be used: (optimized in the implementation):

       KcF ( S / KmS ) - KcR ( P / KmP )
v =  ------------------------------------                ... (3)
        1 + ( S / KmS ) + ( P / KmP )


For the first one-complex representation (1), steady-state of ES complex
is assumed.  Definitions of the parameters in this case are:

KmS, KmP:  Michaelis constants for S and P, assuming steady state of
          the [ES] complex.
          (KmS = (k2 + k-1) / k1, KmP = (k2 + k-1) / k-2)

KcF, KcR:  Catalytic constants for forward and reverse reactions.
          (KcF = k2, KcR = k-1)


If we use two-central complex representation as in (2), and take rapid-
equilibrium assumption, then the net velocity is derived as:

                 k2 ES - k-2 EP 
net velocity = ------------------                 ...(4)
                  E + ES + EP

Eliminating ES and EP, of which concentrations are not explicitly given,
and by using Ks, Kp (dissociation constants for ES and EP,
respectively), then exactly the same equation as (3) can be used
here.   In this case, KmS = Ks, KmP = Kp, KcF = k2, KcR = k-2.
(Note that KmS and KmP are no longer 'Michaelis constants' 
(concentrations of a substrate species when v = Vmax/2, with absence of
the opposite substrate) here.)

Which way to go is just up to you, probably mainly depending on what
data is available. (Michaelis or dissociation constants)

If you have Segel's Enzyme Kinetics, open page 29.  Everything is there.


regards,
-sha



> Sorry to ask a user question on the developer list. I did not find a
> Ecell-user mailing list.
> 
> I have a question about the MichaelisUniUniFluxProcess. What exactly are
> the KmS, KmP, KcF and KcR parameters ? Is this right :
> 
> 
>       k1       KcF     k3
>       ---\    ---\    ---\
> S + E /--- SE /--- SP /--- S + P
>       k-1              k-3
> 
> and that :
> KmS = (k-1 + KcF)/k1
> KmP = (k-3 + KcR)/k3
> 
> or is this false ?
> 
> Thanks a lot
> Eric
> 
> 
> -------------------------------------------------------
> SF.Net email is sponsored by Shop4tech.com-Lowest price on Blank Media
> 100pk Sonic DVD-R 4x for only $29 -100pk Sonic DVD+R for only $33
> Save 50% off Retail on Ink & Toner - Free Shipping and Free Gift.
> http://www.shop4tech.com/z/Inkjet_Cartridges/9_108_r285
> _______________________________________________
> Ecell-devel mailing list
> Ece...@li...
> https://lists.sourceforge.net/lists/listinfo/ecell-devel
-- 

Hi everyone

Sorry to ask a user question on the developer list. I did not find a
Ecell-user mailing list.

I have a question about the MichaelisUniUniFluxProcess. What exactly are
the KmS, KmP, KcF and KcR parameters ? Is this right :


      k1       KcF     k3
      ---\    ---\    ---\
S + E /--- SE /--- SP /--- S + P
      k-1              k-3

and that :
KmS = (k-1 + KcF)/k1
KmP = (k-3 + KcR)/k3

or is this false ?

Thanks a lot
Eric

Hi,

I have noticed that in the new CVS there are some
modules deleted ( such as CashKarp stepper ) and some
new modules added ( such as TauLeap ). Could the
authors be so kind to delete/create .desc  files
accordingly in order to enable users to use modules
from ModelEditor?
Thanks,

Gabor


	
	
		
___________________________________________________________ALL-NEW Yahoo! Messenger - all new features - even more fun!  http://uk.messenger.yahoo.com