Andre's post crossed with mine, but he is right, this should be on CellML Discussion, so I am forwarding my message there too...
-------- Original Message -------- Subject: Re: [team-cellml] Agenda items Date: Tue, 17 Jul 2007 17:53:59 +1200 From: Andrew Miller <[EMAIL PROTECTED]> Reply-To: Bioengineering CellML Team <[EMAIL PROTECTED]> To: Bioengineering CellML Team <[EMAIL PROTECTED]> References: <[EMAIL PROTECTED]> <[EMAIL PROTECTED]> <[EMAIL PROTECTED]> <[EMAIL PROTECTED]@dp ag.ox.ac.uk> <[EMAIL PROTECTED]> <[EMAIL PROTECTED] c.nz> <[EMAIL PROTECTED]@dpag.ox.ac.uk> <46955DE1.70006 [EMAIL PROTECTED]><[EMAIL PROTECTED]@dpag.ox.ac.uk> <[EMAIL PROTECTED]> <[EMAIL PROTECTED]> <[EMAIL PROTECTED]> <[EMAIL PROTECTED]> <[EMAIL PROTECTED]@dpag.ox.ac.uk> Alan Garny wrote: >> One other little point related to stimulii: When you run the Noble 98 >> > model > > (http://www.cellml.org/models/noble_varghese_kohl_noble_1998_version06_varia > >> nt03) with default parameters it works fine but when it is run for 50s it >> fails to stimulate after the 10th beat simply because the stimulus >> > protocol > >> is set in the default params to last for 10s only (see below). Two >> > possible > >> solutions to this are either (i) to always set the stimulus protocol for >> much longer (presumably doesn't matter), or (ii) always display the >> > stimulus > >> current. I think we will need better ways of providing user control over >> > the > >> input stimulus and I think we will need better ways of analysing the >> > output > >> waveforms generally. In fact we have a very powerful signal analysis >> environment already in our UnEmap software (written by David Bullivant). I >> suggest that we look at moving this to PCEnv - I'll discuss this with >> Randall as it is something we could ask Alan Wu to do when he has finished >> updating the old cmgui Xwindows GUI to wxwidgets. >> > > CellML models that require a stimulus protocol to generate an action > potential currently implement one in the CellML model itself (Noble 98 is > one such example indeed). Still, I don't think such a protocol should be > part of the model. The only reason Penny Noble did it that way in the first > place is that we had no other choice in COR (she was waiting for me to > implement my concept of COR project). > > Still, what happens if I want to use that model in a multi-cellular > environment? Ideally I would, in that environment and at some point, specify > the different cellular models that I want to use by selecting different > CellML files. Now, if those CellML files contain a hard-coded stimulus > protocol in them, this will obviously cause problems. > > We may therefore want to think of another way of incorporating a stimulus > protocol, not to mention that one could come up with a different protocol > and still get an action potential (granted, the underlying physiology will > likely be different, but that's another debate). What about, therefore, > making the stimulus protocol(s?) available in some metadata? From there, the > modelling software could extract that protocol and use it, if and only if > the user wants it. > > I would imagine you guys have already discussed this before, in which case I > would appreciate any conclusion that has been drawn about this. > Hi Alan, I don't think that we have discussed this on-list before, but I know Andre in particular has put some thought into this. With CellML 1.1, you can of course put the stimulus protocols in a different file, and build a library of stimulus protocols. I think that this is a good approach, and is far cleaner than trying to build something on an ad hoc basis into the integrator program. Andre has developed some good example models which define stimulus protocols in a separate file. There probably is a case for making stimulus protocols a little more explicit in domain specific interfaces. The proper way to handle this without compromising the generality of CellML would be to add metadata describing which variables are, for example, stimulus voltages, and which components (which will generally be imported components) are stimulus currents. A domain specific command could then perform tasks such as, for example, replacing the stimulus protocol connected to the designated variables with a 'standard' stimulus of a user defined nature (which could be an arbitrary mathematical function, or even a FieldML field). A curve-based graphical editor for some small subset of MathML (which lets you drag points around on the screen to edit MathML parameters) would also make it easier to input and edit stimuli. Best regards, Andrew _______________________________________________ team-cellml mailing list [EMAIL PROTECTED] http://list.bioeng.auckland.ac.nz/mailman/listinfo/team-cellml _______________________________________________ cellml-discussion mailing list [email protected] http://www.cellml.org/mailman/listinfo/cellml-discussion
