Campus Event Calendar
Event Entry
New for: D3
What and Who
Uniformization of Non-Homogeneous Markov Chains
Aleksandr Andreychenko
IMPRS-CS
IMPRS Research Seminar
IMPRS-CS Master student
AG 1, AG 3, AG 4, AG 5, SWS, RG1, MMCI
Public Audience
English
Note: We use this to send email in the morning.
Date, Time and Location
Thursday, 7 January 2010
17:00
30 Minutes
E1 4
024
Saarbrücken
Abstract
Computational biology uses wide range of techniques to analyze behaviour of complex systems. There exist many approaches to solve this task and
choice of proper one mainly depends on the scale of the system. We stick to stochastic models as they can describe effects arising due to
random microscopic events in a cell. Such models can provide better analysis of reaction networks as compared to deterministic methods.
Chemical Master Equation is basic tool to derive properties of reaction networks given by stochastic model. Essentially, it is a system of coupled
ordinary differential equations that is hard to solve in straightforward manner. One good way to do this is to use uniformization of Markov chain.
This method gives good approximation quality with small computational effort for homogeneous models.
More challenging task is to apply these ideas to another type of systems,where probability of reaction occurrence depends on time. Is it important to check how this approach would work for such non-homogeneous reaction networks and compare results with those given by other methods.
choice of proper one mainly depends on the scale of the system. We stick to stochastic models as they can describe effects arising due to
random microscopic events in a cell. Such models can provide better analysis of reaction networks as compared to deterministic methods.
Chemical Master Equation is basic tool to derive properties of reaction networks given by stochastic model. Essentially, it is a system of coupled
ordinary differential equations that is hard to solve in straightforward manner. One good way to do this is to use uniformization of Markov chain.
This method gives good approximation quality with small computational effort for homogeneous models.
More challenging task is to apply these ideas to another type of systems,where probability of reaction occurrence depends on time. Is it important to check how this approach would work for such non-homogeneous reaction networks and compare results with those given by other methods.
Contact
IMPRS Office
0681 9325 225
--email hidden
passcode not visible
logged in users only
Stephanie Jörg, 01/06/2010 11:13 -- Created document.