A Parameter Estimation Technique for Bio-Pathway Models
Prof. P.S. Thiagarajan
National University of Singapore.
The construction of dynamic models of bio-signaling pathways has drawn much attention recently. It is now recognized that a serious stumbling block to making progress in this area is the phenomenon of unknown parameters. Briefly, a signaling pathway can be viewed as a network of bio-chemical reactions. In practice, the values of many of the rate parameters governing these reactions will be unknown. Hence a major challenge is to develop algorithmic techniques for estimating the values of the unknown parameters of large intra-cellular bio-chemical networks.
In the present talk, after setting the background, we will present a decompositional approach that exploits the structure of a large pathway model to break it into smaller components. The parameter estimation problem can then be solved independently for the smaller components. This leads to significant improvements in computational efficiency.
We have tested our technique on a detailed model of the Akt and MAPK pathways with two known and one hypothesized crosstalk mechanisms. The entire model contains 84 unknown parameters. Our simulation results exhibit good correlation with experimental data. Further, they yield positive evidence in support of the hypothesized crosstalk between the two pathways.