BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:On the deduction of chemical reaction rate constants from measurem
 ents of time series of concentration - Paola Lecca\, CoSBi
DTSTART:20090324T140000Z
DTEND:20090324T150000Z
UID:TALK14747@talks.cam.ac.uk
CONTACT:Dr Fabien Petitcolas
DESCRIPTION:*Abstract*: The estimation of parameter values is the bottlene
 ck of the computational analysis of biological systems. Modeling approache
 s are central in systems biology\, as they provide a rational framework to
  guide systematic strategies for key issues in medicine as well as the pha
 rmaceutical and biotechnological industries. Inter- and intra-cellular pro
 cesses require dynamic models\, that contain the rate constants of the bio
 chemical reactions. These kinetic parameters are often not accessible dire
 ctly through experiments. Therefore methods that estimate rate constants w
 ith the maximum precision and accuracy are needed. We present here a new m
 ethod for estimating rate coefficients from noisy observations of concentr
 ation levels at discrete time points. This is traditionally done by comput
 ing an error or cost function\, that measures the distance between the beh
 avior of the experimental data and the behavior of the model. However\, es
 timation of the error function generally requires solving the reaction rat
 e equations\, which can become computationally unfeasible. We propose an a
 lternative approach based on a probabilistic model of the variations in re
 actant concentration. Our method returns the rate coefficients\, the level
  of noise and an error range on the estimates of rate constants. Its proba
 bilistic formulation is key to a principled handling of the noise inherent
  in biological data\, and it allows for a number of further extensions. We
  developed KInfer (Knowledge Inference)\, a freeware software tool that im
 plements the new model of parameter inference (downloadable for free at ht
 tp://www.cosbi.eu).\n\n*Biography*: Paola Lecca received her B. S. in Theo
 retical Physics at the University of Trento (Italy) in 1997 and a PhD in C
 omputer Science in 2006 at the International Doctorate School in Informati
 on and Communication Technologies of University of Trento. From 1997 to 20
 00 she was a research assistant by the Interactive Sensory System Division
  of ITC-Centre for Scientific and Technical Research of Trento. She worked
  by the group of Predictive Models for Biological and Environmental Data A
 nalysis\, where she dealt with the development of predictive models of arc
 haeological sites in Trentino on the basis of historical and environmental
  information integrated and processes by GIS-Geographic Resources Analysis
  Support System. From 2001 to 2002 she obtained a scholarship at the Depar
 tment of Physics of University of Trento in the area of data processing an
 d modelling in Diamine and Explodet projects of the Italian National Insti
 tute of Nuclear Physics. She dealt with the development of new predictive 
 processing methods to be applied to data obtained by neutrons radiation of
  the soil\, for the detection of hidden explosives. In December 2006 Paola
  Lecca joined the Microsoft Research – University of Trento Centre for C
 omputational and System Biology. Her current research interests in the are
 as of systems biology and computational cell biology include issues relate
 d to conceptual frameworks of stochasticity in modelling and simulating bi
 ochemical networks dynamics\, model's structure and model's parameter infe
 rence for optimal experimental design\, and the exploration of the express
 iveness capabilities of process algebras formalisms for describing biologi
 cal systems.
LOCATION:Small public lecture room\, Microsoft Research Ltd\, 7 J J Thomso
 n Avenue (Off Madingley Road)\, Cambridge
END:VEVENT
END:VCALENDAR