Exothermic reactions:
In the reactions that I investigate the produced heat of reaction can be quite significant and therefore lead to increase in temperature. This of course affects the reaction rates again (assuming f.e. an Arrhenius like behavior). Is there any option to include the heat build up leading to an increase in temperature. Am I for instance able to track which reactions have been performed and update the kinetic constants over the course of the reaction?
Chain length dependent reactions:
For polymerization processes it is common to scale the reaction rates according to the length (or mol weight) of the chains reacting. F.e let’s assume we have only one species of Monomers “x-A-x” having two equal reactive sites”x”. So A can react with itself to form chains. I would like to take into account that
x-A-x + x-A-x -> x-A-y-A-x might have a different reaction constant than the corresponding reaction of two dimers: x-A-y-A-x + x-A-y-A-x -> x-A-y-A -y-A -y-A -x
--
You received this message because you are subscribed to the Google Groups "kappa-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to kappa-users...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/kappa-users/2114958f-986b-4338-872b-6f57e5614f0b%40googlegroups.com.
To unsubscribe from this group and stop receiving emails from it, send an email to kappa...@googlegroups.com.
To unsubscribe from this group and stop receiving emails from it, send an email to kappa-users...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/kappa-users/8dd3325c-32c4-448a-8669-3a20d0c26ce0%40googlegroups.com.
Experimental and theoretical studies have shown that the activ- ity of each signaling team can be represented in a coarse-grained manner using a Monod-Wyman-Changeux model12,33. In NFsim this is achieved by defining a local function, ‘pOn’, that depends on the methylation and ligand binding state of each receptor in a given signaling team (Fig. 5b). The argument ‘x’ refers to the particular complex over which pOn must be evaluated. The pOn function can then be used to represent how the rate of methyla- tion of one receptor depends on the activity of its signaling team. This is illustrated in the block of code in Figure 5b, where the rate of the methylation reaction is multiplied by pOn(x). The ‘%x’ prefixed to ‘Rec(m~?)’ tells NFsim to evaluate the function pOn(x) on the entire complex (signaling team) that is connected to the indicated receptor. The local-function syntax also allows the arrangement of receptors in a signaling team to be modified with- out changing rule definitions.
Local functions enable a single rule to specify many reactions with rates that depend on the specific properties of the reacting species
To view this discussion on the web visit https://groups.google.com/d/msgid/kappa-users/BD533FAA-7279-4CD1-AFEB-129C19EC41E7%40gmail.com.
--
You received this message because you are subscribed to the Google Groups "kappa-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to kappa-users...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/kappa-users/2114958f-986b-4338-872b-6f57e5614f0b%40googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/kappa-users/CAAs77FDZoJUO5mMg_%3DP0GfCAEKHrOyZKSMWM8LO%3Dko7ZLJWcMw%40mail.gmail.com.
Maybe a few words to my situation to clarify certain things. I work in the R&D Department of a chemical company. I deal with molecular modeling of polymeric systems on different length scales. However, one of my responsibilities is an in-house kmc code. The Problem with this code is, that it has been developed decades ago and it is not easy to access for modifications. Therefore I was just curious if there might be more state of the art Software packages that I could use. But of course I wanted to make sure that I wont get bottlenecked right from the start.
While I did not really find to many solutions in the chemical community the Bio-chemical community did offer much more :).
For the Chain length dependance we usually asume something like 1/n to 1/n^2 as a scaling factor where n is the number of monomers.
For the ring building I wanted to use md-simulations to predict for which length of a chain it is reasonable to form a ring.
Again, I thank you all for your Input. Please feel free to give me any other advice!
Have a nice Weekend.
Best regards from Germany,
Jan
Hi all,
Extremely late reply indeed, Sorry.
- Dealing with heat is indeed doable with tokens. Despite fighting with the syntax and having to be careful with counter intuitive behaviors of tokens (For example, nothing prevent them to become negative and a negative activity for a rule is a weird thing :-)), there should be no difficulty. Complains as soon as you don't find/understand something.
- Enumerating complexes up to length 100 (even 1000) is piece of
cake for the simulator but I'm not sure it is of any help if the
question is really to deal with arbitrary complexes.
Technical details for the insiders and the interested: Problems
with the sharing structure arise when you have 1 agent type with
17 phosphorylation sites and you enumerate all the possibilities
because by default we are enumerating the 17! possible paths to
discover all these possibilities (because sites can be visited in
any order...) To avoid that, you can indeed launch the simulator
with option `-sharing None` or wait that I finally take the time
to debug the real solution to the problem which is stuck at 85%
completion in the github branches `tons_of_rules` for 2 months...
The case of discovering a linear polymer is not problematic
because at each step of your exploration, you only have 2
possibilities : discovering a new agent because the only site you
haven't inspected yet is bound to someone or completing the
discovery of a complex because this site is free! So you end up
with only a few hundreds (thousands) elements.
- There is no way for a Kappa agent to introspect the size of the complex it is part of. I don't see any way to encode this knowledge in Kappa.
Technical details for the insiders and the interested: Yes, at the price of changing any bits of the simulator, we could add this knowledge with a computational complexity as bad as it is already when you deal with rule involving "molecular ambiguity". The real price to pay is in the increase in complexity of the engine. First, although we discuss that already for other use, there is no syntax to give a name to an agent in a rule yet. That means there are some work to do in order to refer in the rule rate to "(the size of the complex in which appears) the second agent of type EGF mentioned in the left hand side". Moreover, all instance of a rule has the same activity for now so the data-structure used to store and pick at random a instance is uniform. If now the rate depends on the instance, we must introduce the notion of weight in between occurrences. Again nothing is impossible here, but it reprensent a substantial surgery :-)
Best,
Pierre B.
To view this discussion on the web visit https://groups.google.com/d/msgid/kappa-users/BD533FAA-7279-4CD1-AFEB-129C19EC41E7%40gmail.com.