TY - JOUR TI - Computations of DSGRN DO - https://doi.org/doi:10.7282/t3-xc52-yt51 PY - 2020 AB - The study of this thesis is mainly motivated by the computational problems that arised from paper [7] where authors designed a new dynamic system model called Dynamic Signatures Generated by Regulatory Networks (DSGRN) to simulate gene regulatory networks. An essential property of DSGRN model is that the phase transition graph is invariant over a family of subvariety which is defined by a set of simple polynomials over parameter space. One of the work in DSGRN is to study the family of subvariety over parameter space and two basic problem arises: realizability and topology. As to reailizability, we are concerned with constructing the whole family of subvariety. For topology, we want to compute the homology group of a set of subvariety for a given phenotype. This thesis has two main contributions. First, we develop a new algorithm that greatly extends the computational ability of DSGRN to wider class of regulatory network. Second, we design a computational framework for the homology computation of subvariety over the parameter space. KW - Mathematics LA - English ER -