Efficient simulation algorithms for models of stochastic processes in computational biology and synthesis of the signaling pathways models.

Grant NCN: Opus UMO-2016/23/B/ST6/03455
Total budget [PLN]: 895 400
Start date: 24/08/2017; End date: 21/08/2021
Status: 6 month/s left

The term signaling pathways refers to cascades of biochemical processes resulting in change of gene expression in response to changes in intra- or extracellular environment conditions. Discovery of processes that form these cascades is a key to understanding intracellular regulatory mechanisms and, consequently, is the first step toward our controlling these processes. Such control may take form of a therapies targeted at particular diseases. It should be stressed that it is not just the knowledge which molecular players are involved in signaling pathways that is important, but also the dynamics of the processes under consideration.In recent years signaling pathways have been the subject of extensive research, both experimental and theoretical. These studies led to discovery of new interactions between proteins, protein complexes, mRNAs and oter molecules as well as development of original mathematical models that describe them and help I analysis of dynamical properties of processes involved in signaling pathways and predict cellular responses to external stimuli. This, in turn, yielded new ways of influencing cells behawior – in terms of therapeutical actions. Mathematical methods facilitated faster progress in the field, since at least some of time- and resource-consuming experiments could be replaced by computer simulations and formal mathwamatical analysis of systems described by mathematical equations.It should be noted, however, that usually researchers focus only on a small fragment of an extremely large number of intracellular processes. Even in these cases, mathematical models are very complex  and their simulation requires large computational power. If one aims at investigating interactions between two such fragments, the computational power requirements increase even farther. Therefore it is crucial to develop new methods that facilitate effective merging of existing models and algorithms  for simulations of their dynamics. These are te two main topics of the proposed project.The project consists of four intertwinned tasks: (1) Development of original methods for simulations of the mathematical models of stochastic processes into the signaling pathways that take into account heterogeneous cellular responses to external stimuli; (2) Development of original methods for merging chosen signaling pathways models; (3) Experimental identification of parameters characterizing activation of the considered signaling pathways and (4) Identification of key interactions between NFκB and HSF1 pathways at the protein and transcription regulation levels. First two tasks belong to the fields of computational biology and system engineering and constitute the main part of the project. However, theoretical work in these fields should be related to practical applications. Therefore, two other tasks have been planned, of experimental and bioinformatical nature. The third of aforementioned tasks will provide data for estimation of parameters of the mathematical models, whereas the last one will allow to determine specific structure of the chosen biological systems.The methods developed in the project will be tested on analysis of signaling pathways, in which p53 and NFkB proteins are key elements. These pathways are involved in regulation of inflammatory and immune responses, apoptosis, carcinogenesis, control of cell cycle progression, angiogenesis and metastasis. Additionally, heat-shock HSF1-dependent pathways will be merged with them, as they potentially may be utilized to sensitize cancer cells to radio- or chemotherapy.New methods and computational tools developed in the project will facilitate expansion of knowledge on regulatory mechanisms of biological processes and, due to the nature of the pathways under consideration, may ultimately lead to potential clinical applications. New computational methods developed in the project will be general enough to support research on crosstalk between other pathways as well. Development of new techniques of analysis of mathematical models should also result in increasing of costs effectiveness of biological experiments due to new model-based experimental protocols. Moreover, results obtained in the project will elucidate important issues concerning the mechanisms of the interactions between pathways critical for cellular response to stress. As a result, acquired knowledge will pave the way for modulating cell resistance to cytotoxic factors used in anticancer therapies and, consequently, provide hints on possible changes in currently used therapy protocols.
The main goal of the project is to develop original methods of merging existing models of signaling pathways and their computational analysis. To verify applicability of these methods, functional interactions between NFκB and HSF1 signaling pathways will be analyzed.To reach that goal two intermediate objectives will be pursued:1. development of original mathematical models and methods of their analysis that would support experimental work in investigation of pathways crosstalk.2. identification and characterization of the crosstalk between NFkB and HSF1 pathways at two levels: (i) regulation of transcription of target NFκB- and HSF1-dependent genes and (ii) protein-protein interactions;
The project consist of four tasks that will be run simultaneously but with clear dependencies among them. These tasks are:·      Task 1. Development of original methods for simulation of the mathematical models of the stochastic processes into the signaling pathways that take into account heterogeneous cellular responses to external stimuliDevelopment of original methods for simulation of stochastic models of signaling pathways;bifurcation and sensitivity analysis of the models developed in Task 2;development of stochastic versions of the models developed in Task 2;comparison of simpler models, including only NFkB and HSF1 pathways, with the extended ones including p53 pathway to determine minimum level of complexity that allows to predict cell fate on the basis of intracellular protein levels.  Task 2: Development of original methods of merging models of chosen signaling pathways:development of deterministic models incorporating known and hypothetical crosstalk mechanisms between NFkB, p53 and HSF1 pathways;estimation of model parameters basing on experimental results from Task 3;sensitivity analysis of the developed models, aimed at finding the most crucial processes regulating the crosstalk;computational analysis of the model, aimed at finding optimal time window between heat shock and TNF stimulation, in the sense of maximum level of NFkB pathway inhibition.   Task 3: Experimental identification of parameters characterizing activation of the considered signaling pathwaysexperimental estimation of kinetic parameters of processes involved in the crosstalk, including knowledge resulting from Task 4;analysis of heterogeneity in cellular responses to stimuli.Task 4: Identification of key interactions between NFκB and HSF1 pathways at the protein and transcription regulation levelsidentification of target genes whose transcription is regulated by NFkB and/or HSF1;identification of genes whose regulatory regions contain binding sites for both HSF1 and NFκB;identification and analysis of protein complexes containing transcription factors of interest.

Project manager

Krzysztof Puszyński


Magdalena Ochab, Krzysztof Puszynski, Jarosław Smieja, Anna Naumowicz