kirschner_1998
Original Model Status:
Model is missing values for variables s1 and s2.
s1 is a treatment term that represents an external source of effector cells such as LAK or TIL cells s2 is a treatment term that represents an input of IL2 into the system
If values of 1.0 are put into the model, it will run in PCEnv and produce output, but not the oscillating spikes shown in the paper.
From looking at the paper, it appears that the model is more complex than the few ODEs that have been encoded. Some of the paper uses matrices for bifurcation analysis and it is unclear whether this should be incorporated into the core set of ODEs defining the system. Obviously something is amiss.
Work done on model:
03/08/07: I have looked at the paper, and come to the above conclusion. I have emailed the author (see kirschner_1998_panetta-email) and asked for the original .ode file used for this model. The paper says XPPAUT was used for the bifurcation analysis, although it is not quite clear whether XPPAUT was used for the whole model.
06/08/07: On suggestion from the author, recoded model to use non-dimensionalised/scaled set of equations. Model now produces a train of spikes. There are still some parameters that are undefined in the paper which could help the model produce more outputs than just the train of spikes. This is a definite improvement over the previous version, in any case. Non-dimensionalized version uploaded as version 03.
Curation status of model:
03/08/07: Functional if arbitrary values for s1 and s2 are put into the model. The units for s1 and s2 are 'flux' which does not appear to be correct, given the role of these variables in the biology. I have uploaded a running version of this model into the repository, using values of 1.0 for s1 and 0.1 for s2. This model can be found at http://www.cellml.org/models/kirschner_panetta_1998_version02/pmr_view_model
06/08/07: Model is now reproducing at least some of the output shown in the figures in the paper. Also meets unit tests because everything is non-dimensionalized -i.e. dimensionless (I have not checked the integrity of this process in the paper...) - therefore model given 2 stars for standard, tentatively COR (because there are few connections and none are duplicated, although testing has not actually been done in COR,) and PCEnv. Uploaded version 3 at http://www.cellml.org/models/kirschner_panetta_1998_version03/pmr_view_model
Further work required:
Have emailed the author re: the values of certain parameters. See J_Carl_Panetta_correspondence for transcripts of correspondence with J. Carl Panetta, the co-author of this paper.
Hopefully Dr Panetta can help me fine tune the output of this model.
Return to PMR Curation Wiki