I am modelling a pathway using the simbiology toolbox. I would like to know wat all are the data required to start with the modelling process. I would also like to know approximately as to how long does it take to model and simulate a 4 step pathway ?

"Padmaja " <padmaja.v30@live.com> wrote in message <hjefvu$l5v$1@fred.mathworks.com>...
> I am modelling a pathway using the simbiology toolbox. I would like to know wat all are the data required to start with the modelling process. I would also like to know approximately as to how long does it take to model and simulate a 4 step pathway ?

Hi Padmaja,

It's very difficult to answer your question without more specific information. But let me try to give you some general answers.

If you've never used SimBiology before, I would recommend setting aside at least one day to learn how to use the software. Depending on how you learn best, I recommend reading the Getting Started Guide in the documentation and working through a few examples or demos.

After that, you should be ready to start building a model. Running a single simulation generally only takes a few seconds. So the majority of your time will probably be spend building and parameterizing the model. How much time that will take really depends on what you want to do with the model and what is already known about your system. I have been able to build a model based on a paper in an hour or two, when the paper has all the details. The paper made the work easy because it very clearly described all the reactions, the reaction rates, and the initial concentrations of all the species.

If you are not starting from someone else's model, the problem is significantly more difficult. You will need to determine what the reactions, reaction rates, and initial concentrations are. If you're only trying to understand some qualitative phenomena, the exact parameters could be less important. But if you're trying to reproduce specific experimental results, you will need to invest a lot more time in checking the structure of the model and determining whether you have the appropriate parameters. That's getting into the territory of "research" and is a very open-ended problem! That's the sort of thing that is never truly finished...

I hope this helps you get started.

Good luck!
-Arthur

"Arthur Goldsipe" <REMOVE.Arthur.Goldsipe@REMOVE.mathworks.com> wrote in message <hjfh58$9r1$1@fred.mathworks.com>...
> "Padmaja " <padmaja.v30@live.com> wrote in message <hjefvu$l5v$1@fred.mathworks.com>...
> > I am modelling a pathway using the simbiology toolbox. I would like to know wat all are the data required to start with the modelling process. I would also like to know approximately as to how long does it take to model and simulate a 4 step pathway ?
> 
> Hi Padmaja,
> 
> It's very difficult to answer your question without more specific information. But let me try to give you some general answers.
> 
> If you've never used SimBiology before, I would recommend setting aside at least one day to learn how to use the software. Depending on how you learn best, I recommend reading the Getting Started Guide in the documentation and working through a few examples or demos.
> 
> After that, you should be ready to start building a model. Running a single simulation generally only takes a few seconds. So the majority of your time will probably be spend building and parameterizing the model. How much time that will take really depends on what you want to do with the model and what is already known about your system. I have been able to build a model based on a paper in an hour or two, when the paper has all the details. The paper made the work easy because it very clearly described all the reactions, the reaction rates, and the initial concentrations of all the species.
> 
> If you are not starting from someone else's model, the problem is significantly more difficult. You will need to determine what the reactions, reaction rates, and initial concentrations are. If you're only trying to understand some qualitative phenomena, the exact parameters could be less important. But if you're trying to reproduce specific experimental results, you will need to invest a lot more time in checking the structure of the model and determining whether you have the appropriate parameters. That's getting into the territory of "research" and is a very open-ended problem! That's the sort of thing that is never truly finished...
> 
> I hope this helps you get started.
> 
> Good luck!
> -Arthur

Hello sir,

Thank you for the basic information. I am trying to model a portion of Alzheimer disease's pathway. I am starting to work from scratch and till now have information regarding my species and reaction and a little about the concentrations. Do I have to know the rate before i start to model or can i compute it using the toolbox ? I would also like to know what more do i require to start with the modelling properly ? 

Regards,
Padmaja, S

"Padmaja " <padmaja.v30@live.com> wrote in message <hjov8p$qgl$1@fred.mathworks.com>...
> Thank you for the basic information. I am trying to model a portion of Alzheimer disease's pathway. I am starting to work from scratch and till now have information regarding my species and reaction and a little about the concentrations. Do I have to know the rate before i start to model or can i compute it using the toolbox ? I would also like to know what more do i require to start with the modelling properly ? 

Hi Padmaja,

SimBiology requires you to specify a model for the reaction rate. You can either specify the explicit functional form of the reaction rate or choose among common kinetic laws (such as mass action, Michaelis-Menten, or Hill kinetics). In addition, you must specify the values of any parameters appearing in the reaction rate or kinetic law. In other words, you need to have a detailed mechanistic model in mind to use SimBiology. I strongly encourage you to look at the documentation and other models in the literature to see what is required.

Regards,
Arthur

"Arthur Goldsipe" <REMOVE.Arthur.Goldsipe@REMOVE.mathworks.com> wrote in message <hjprh5$ntq$1@fred.mathworks.com>...
> "Padmaja " <padmaja.v30@live.com> wrote in message <hjov8p$qgl$1@fred.mathworks.com>...
> > Thank you for the basic information. I am trying to model a portion of Alzheimer disease's pathway. I am starting to work from scratch and till now have information regarding my species and reaction and a little about the concentrations. Do I have to know the rate before i start to model or can i compute it using the toolbox ? I would also like to know what more do i require to start with the modelling properly ? 
> 
> Hi Padmaja,
> 
> SimBiology requires you to specify a model for the reaction rate. You can either specify the explicit functional form of the reaction rate or choose among common kinetic laws (such as mass action, Michaelis-Menten, or Hill kinetics). In addition, you must specify the values of any parameters appearing in the reaction rate or kinetic law. In other words, you need to have a detailed mechanistic model in mind to use SimBiology. I strongly encourage you to look at the documentation and other models in the literature to see what is required.
> 
> Regards,
> Arthur

Hello sir,
Thank you. I think this information is enough for now to start away with my modelling. I'd get back if i have more queries.
Regards,
Padmaja

"Padmaja " <padmaja.v30@live.com> wrote in message <hjr5ot$j1i$1@fred.mathworks.com>...
> "Arthur Goldsipe" <REMOVE.Arthur.Goldsipe@REMOVE.mathworks.com> wrote in message <hjprh5$ntq$1@fred.mathworks.com>...
> > "Padmaja " <padmaja.v30@live.com> wrote in message <hjov8p$qgl$1@fred.mathworks.com>...
> > > Thank you for the basic information. I am trying to model a portion of Alzheimer disease's pathway. I am starting to work from scratch and till now have information regarding my species and reaction and a little about the concentrations. Do I have to know the rate before i start to model or can i compute it using the toolbox ? I would also like to know what more do i require to start with the modelling properly ? 
> > 
> > Hi Padmaja,
> > 
> > SimBiology requires you to specify a model for the reaction rate. You can either specify the explicit functional form of the reaction rate or choose among common kinetic laws (such as mass action, Michaelis-Menten, or Hill kinetics). In addition, you must specify the values of any parameters appearing in the reaction rate or kinetic law. In other words, you need to have a detailed mechanistic model in mind to use SimBiology. I strongly encourage you to look at the documentation and other models in the literature to see what is required.
> > 
> > Regards,
> > Arthur
> 
> Hello sir,
> Thank you. I think this information is enough for now to start away with my modelling. I'd get back if i have more queries.
> Regards,
> Padmaja

Hello sir,
I simulated a pathway using matlab. But there was a small discrepancy. Wen i simulated the pathway using the names of the reactant and the product molecules as it is in the pathway, i get a linear graph but when i do the simulation of the same pathway using names such as A, B, C, etc.,. in place of the reactant's and products' original names, i get a different curved graph inspite of having the parameters and concentrations unaltered. What is the reason behind this variation and is there any inbuilt library of molecules that could possibly cause this ?

Thanking you in advance for your kind help,
Regards,
Padmaja

> I simulated a pathway using matlab. But there was a small discrepancy. Wen i simulated the pathway using the names of the reactant and the product molecules as it is in the pathway, i get a linear graph but when i do the simulation of the same pathway using names such as A, B, C, etc.,. in place of the reactant's and products' original names, i get a different curved graph inspite of having the parameters and concentrations unaltered. What is the reason behind this variation and is there any inbuilt library of molecules that could possibly cause this ?

Hi Padmaja,

I'm happy to hear that you've been able to get started using SimBiology. You should indeed get the same numerical results regardless of what you name the species in your model. There are no built-in species in SimBiology, but if write a reaction rate using a name that is not defined in the model, then SimBiology will try to use a MATLAB function by that name. For example, if you defined a rate as "pi*X" without defining a species "pi", then SimBiology will evaluate the rate as "3.14159*X" (because there is a function called pi that returns this value).

So, it sounds to me like you have a mistake in one or both of your models. Please double-check that you replaced every instance of the species names correctly and that nothing else in your model has changed. One way you might be able to isolate the problem is by temporarily "turning off" all reactions except one and simulate each model. Probably the easiest way to do that is to set the "Active" property of a reaction to false to temporarily remove it from a simulation. Of course, don't forget to turn set Active to true when you're ready to simulate the entire model!

I hope that's enough information to help you locate the problem. If not, then you will probably need to share with us the exact steps necessary to reproduce your problem.

Good luck!
-Arthur

"Arthur Goldsipe" <REMOVE.Arthur.Goldsipe@REMOVE.mathworks.com> wrote in message <hkerio$i9i$1@fred.mathworks.com>...
> > I simulated a pathway using matlab. But there was a small discrepancy. Wen i simulated the pathway using the names of the reactant and the product molecules as it is in the pathway, i get a linear graph but when i do the simulation of the same pathway using names such as A, B, C, etc.,. in place of the reactant's and products' original names, i get a different curved graph inspite of having the parameters and concentrations unaltered. What is the reason behind this variation and is there any inbuilt library of molecules that could possibly cause this ?
> 
> Hi Padmaja,
> 
> I'm happy to hear that you've been able to get started using SimBiology. You should indeed get the same numerical results regardless of what you name the species in your model. There are no built-in species in SimBiology, but if write a reaction rate using a name that is not defined in the model, then SimBiology will try to use a MATLAB function by that name. For example, if you defined a rate as "pi*X" without defining a species "pi", then SimBiology will evaluate the rate as "3.14159*X" (because there is a function called pi that returns this value).
> 
> So, it sounds to me like you have a mistake in one or both of your models. Please double-check that you replaced every instance of the species names correctly and that nothing else in your model has changed. One way you might be able to isolate the problem is by temporarily "turning off" all reactions except one and simulate each model. Probably the easiest way to do that is to set the "Active" property of a reaction to false to temporarily remove it from a simulation. Of course, don't forget to turn set Active to true when you're ready to simulate the entire model!
> 
> I hope that's enough information to help you locate the problem. If not, then you will probably need to share with us the exact steps necessary to reproduce your problem.
> 
> Good luck!
> -Arthur

Hello sir,
I am glad to tell you that i have started working with simbiology, The tips you suggested were useful. Thank you sir. Can u also help me understand the meaning of states in simbiology ? Does it mean concentration of reactants ? The graphs show the y axis as states. Wat can be inferred from it and wat does a sensitivity analysis plot give and how do we infer from the plot ?

Thank you,
regards,
Padmaja,S

> I am glad to tell you that i have started working with simbiology, The tips you suggested were useful. Thank you sir. Can u also help me understand the meaning of states in simbiology ? Does it mean concentration of reactants ? The graphs show the y axis as states. Wat can be inferred from it and wat does a sensitivity analysis plot give and how do we infer from the plot ?

Hi Padmaja,

I am very glad that you're continuing to make progress using SimBiology. I'm also glad you're asking the questions on this newsgroup, so that others can learn from what you're doing.

"State" refers to the value of *something* in the model over the course of a simulation. The most common kind of "state" is the concentration (or amount) of a species. But "state" can also refer to the the volume of a compartment or the value of a parameter. So, a plot of state versus time allows you to see how something changes over the course of a simulation.

With regard to sensitivity analysis, I'm not sure what specific plot you are referring to. You will find some specific examples and discussion by searching the MATLAB help for "Sensitivity Analysis", but let me go ahead and give a short summary here. SimBiology calculates time-dependent sensitivities. Sensitivity plots allow you to make statements like the following: "If I increase rate constant k by 1%, then the concentration of glucose will be 5% lower at 30 seconds and 10% lower at 60 seconds."

It is also very common to use a single number (independent of simulation time) to quantify the sensitivity. One approach is to integrate the sensitivity over time. Another approach is to take the maximum sensitivity. By default, we show the integral of the sensitivity when doing sensitivity analysis in the SimBiology desktop.

Hope that helps,
-Arthur