#49 Negative feedback based approach to simulation of complex biological systems
Authors: Irena Roterman, Leszek Konieczny, Marian Bubak and Jakub Wach
Simulation of complex biological system is of great importance due to potential applications in medicine for understanding many pathological phenomena occurring in cells of the human body [1].These systems are thermodynamically open and self-regulation is necessary to keep the system stable.. Negative feedback loop is considered as the simplest and fundamental way to maintain the homeostasis. The communication between cells and structuring of an organism is achieved by hormones, product-substrate relation or by an action of a nervous system [2].
In our approach, a negative feedback loop links both: a structure and a function. The negative feedback loop consists of an effector, which delivers a product and a controlling receptor. Consequently, a living system consists of a large number of such units, communicated using two ways: via effectors and via receptors.A set of such feedback units enables to investigate the time-dependence changes and stabilization of products concentration. The communication between cells and organism is mediated by receptors which receive signals from an organism (hormone) and adapt stabilization levels.
So far we have simulated a biological process of the blood glucose level stabilization and the pathological form - diabetes. They are defined in respect to real parameters as they occur in human body. Even though the systems in question are relatively simple, consisting of only a handful of components, the results are encouraging. Future work concentrates on a biological phenomenon ofthe circadian cycle as well as on improving the technical side of the simulation by implementing it over a distributed environment.
The current application is available as a standalone web service:http://crick.cm-uj.krakow.pl:8080/nfs/.
References
1. Karr, Jonathan R. et al., A Whole-Cell Computational Model Predicts Phenotype from Genotype, Cell , Volume 150 , Issue 2
2. Leszek Konieczny, Irena Roterman-Konieczna, Paweł Spólnik, Systems Biology: Functional Strategies of Living Organisms, ISBN 978-3-319-01336-7, Springer 2013
#57 Non-stationary individual and household income of poor, rich and middle classes in Mexico
Authors: Marcelo Del Castillo-Mussot, Pavel Soriano-Hernandez, Oscar Córdoba-Rodriguez and Ricardo Mansilla-Corona
Inequality in social systems has been a universal and robust phenomenon, not bound by either time or geography, but fortunately for scholars, it has a few statistical regularities, as in the case of income and wealth distributions over a wide range of societies and time periods. Despite Mexican peso crisis in 1994 followed by a severe economic recession, the individual and household income distributions in the period 1992-2008 always exhibit a two class-structure; a highly fluctuating high-income class adjusted to a Pareto power-law distribution, and a low- income class (including poor and middle classes) adjusted to either Log-normal or Gamma distributions, where poor agents are defined as those with income below the maximum of the uni-modal distribution. Then the effects of crisis on the income distributions of the three classes is briefly analyzed. Since data in Mexico is obtained from personal surveys, and not from information obtained indirectly from taxes or consumption, then it is possible that data may not reflect real incomes. Usually income data of poor and rich agents are not very reliable, particularly if obtained by personal interviews and in times of crisis. Thus, the observed resilience of the middle class income may be related to a kind of an averaging effect that compensates errors. The average population over the years of the three classes in Mexico are 25.89%, 68.01% and 6.09% for poor, middle and rich classes, respectively, in the decade of the 1990s, and in the 2000s are 28.65%, 65.55% and 5.80%.
#58 Adaptive Cities
Authors: Carlos Gershenson and Carlo Ratti
Cities are changing constantly. All urban systems face different conditions from day to day. Even when averaged regularities can be found, urban systems will be more efficient if they can adapt to changes at the same speeds at which these occur. Inspired by cybernetics, we propose a theory for adaptive cities. We identify three main components: information, algorithms, and agents, which we illustrate with current and future examples. The implications of adaptive cities are manifold, with direct impacts in mobility, sustainability, governance, and society. Still, the potential of adaptive cities will not depend so much on technology as on how we use it.
#63 Spreading Paths on Partially-Observed Temporal Social Networks
Author: Cory Cox
Understanding how and how far information, behaviors, and pathogens spread through social networks is clearly an important problem. Onnela and Christakis drew our attention to the problems of partial observability of both the topology of a (static) network as well as the dynamical processes occurring on a network . However, the networks underlying the spreading process are not static, which gives an additional dimension to the problem. I simulate spreading dynamics on a temporal communication network, which constrains spreading dynamics to time-respecting paths between the source and target vertices. Further, using various combinations of vertex and edge sampling, I compare the time-respecting path lengths, latency, and reachability for: (1) the fully-observed network; and (2) the partially-observed networks. Lastly, I compare the spreading curve for a susceptible-infected (SI) process in all cases of full and partial observation. Given the increasing availability of temporal data and the importance of dynamics *of* networks in constraining spreading dynamics *on* networks, my goal is to characterize systemic biases found in key network metrics resulting from partial observation of temporal networks.
#64 What is the effect of opinion change and switching cost on Technological Lock-Ins?
Author: Tamer Khraisha
Most of the existing literature on the problem of technological Lock-Ins assumes that once the system reaches a “tipping point” in terms of the number of users of a technology, successive agents will prefer to adopt the same technology, thus causing the system to lock-into that choice. However, the majority of these models are based on two strict assumptions, namely that agents are able to observe the choices of the entire population and, second, agents do not change their opinion. In this research I show that by locating agents on a network and introducing a dynamical component in the system, i.e. agents changing their opinion, the system exhibits less path dependency and, as a consequence, lock-in effects decrease. Initially, we have two technologies and two types of agents. At each time step, an agent is introduced whose type is chosen randomly and the decision of what technology to adopt will depend on two factors: First, his\her preference for one or the other technology and Second, information about the choices of the rest of the network. Information is local if it is received from neighbors, and global if it is received from next neighbors. To include local and global information, I model the information as a convex combination of neighbors’ choices with values of coefficients decreasing according to a scaling law as we move from immediate neighbors to neighbors of neighbors etc. After visiting each agent, I choose randomly a node, and make it change it's opinion, but paying a switching cost. The goal is to understand the effect of switching cost and position in the network on the lock in. To test the effect of network topology and local information on lock-ins and lock-outs, the model will be tested on different networks, including random, small-world and Scale-free networks.
Authors: Irena Roterman, Leszek Konieczny, Marian Bubak and Jakub Wach
Simulation of complex biological system is of great importance due to potential applications in medicine for understanding many pathological phenomena occurring in cells of the human body [1].These systems are thermodynamically open and self-regulation is necessary to keep the system stable.. Negative feedback loop is considered as the simplest and fundamental way to maintain the homeostasis. The communication between cells and structuring of an organism is achieved by hormones, product-substrate relation or by an action of a nervous system [2].
In our approach, a negative feedback loop links both: a structure and a function. The negative feedback loop consists of an effector, which delivers a product and a controlling receptor. Consequently, a living system consists of a large number of such units, communicated using two ways: via effectors and via receptors.A set of such feedback units enables to investigate the time-dependence changes and stabilization of products concentration. The communication between cells and organism is mediated by receptors which receive signals from an organism (hormone) and adapt stabilization levels.
So far we have simulated a biological process of the blood glucose level stabilization and the pathological form - diabetes. They are defined in respect to real parameters as they occur in human body. Even though the systems in question are relatively simple, consisting of only a handful of components, the results are encouraging. Future work concentrates on a biological phenomenon ofthe circadian cycle as well as on improving the technical side of the simulation by implementing it over a distributed environment.
The current application is available as a standalone web service:http://crick.cm-uj.krakow.pl:8080/nfs/.
References
1. Karr, Jonathan R. et al., A Whole-Cell Computational Model Predicts Phenotype from Genotype, Cell , Volume 150 , Issue 2
2. Leszek Konieczny, Irena Roterman-Konieczna, Paweł Spólnik, Systems Biology: Functional Strategies of Living Organisms, ISBN 978-3-319-01336-7, Springer 2013
#57 Non-stationary individual and household income of poor, rich and middle classes in Mexico
Authors: Marcelo Del Castillo-Mussot, Pavel Soriano-Hernandez, Oscar Córdoba-Rodriguez and Ricardo Mansilla-Corona
Inequality in social systems has been a universal and robust phenomenon, not bound by either time or geography, but fortunately for scholars, it has a few statistical regularities, as in the case of income and wealth distributions over a wide range of societies and time periods. Despite Mexican peso crisis in 1994 followed by a severe economic recession, the individual and household income distributions in the period 1992-2008 always exhibit a two class-structure; a highly fluctuating high-income class adjusted to a Pareto power-law distribution, and a low- income class (including poor and middle classes) adjusted to either Log-normal or Gamma distributions, where poor agents are defined as those with income below the maximum of the uni-modal distribution. Then the effects of crisis on the income distributions of the three classes is briefly analyzed. Since data in Mexico is obtained from personal surveys, and not from information obtained indirectly from taxes or consumption, then it is possible that data may not reflect real incomes. Usually income data of poor and rich agents are not very reliable, particularly if obtained by personal interviews and in times of crisis. Thus, the observed resilience of the middle class income may be related to a kind of an averaging effect that compensates errors. The average population over the years of the three classes in Mexico are 25.89%, 68.01% and 6.09% for poor, middle and rich classes, respectively, in the decade of the 1990s, and in the 2000s are 28.65%, 65.55% and 5.80%.
#58 Adaptive Cities
Authors: Carlos Gershenson and Carlo Ratti
Cities are changing constantly. All urban systems face different conditions from day to day. Even when averaged regularities can be found, urban systems will be more efficient if they can adapt to changes at the same speeds at which these occur. Inspired by cybernetics, we propose a theory for adaptive cities. We identify three main components: information, algorithms, and agents, which we illustrate with current and future examples. The implications of adaptive cities are manifold, with direct impacts in mobility, sustainability, governance, and society. Still, the potential of adaptive cities will not depend so much on technology as on how we use it.
#63 Spreading Paths on Partially-Observed Temporal Social Networks
Author: Cory Cox
Understanding how and how far information, behaviors, and pathogens spread through social networks is clearly an important problem. Onnela and Christakis drew our attention to the problems of partial observability of both the topology of a (static) network as well as the dynamical processes occurring on a network . However, the networks underlying the spreading process are not static, which gives an additional dimension to the problem. I simulate spreading dynamics on a temporal communication network, which constrains spreading dynamics to time-respecting paths between the source and target vertices. Further, using various combinations of vertex and edge sampling, I compare the time-respecting path lengths, latency, and reachability for: (1) the fully-observed network; and (2) the partially-observed networks. Lastly, I compare the spreading curve for a susceptible-infected (SI) process in all cases of full and partial observation. Given the increasing availability of temporal data and the importance of dynamics *of* networks in constraining spreading dynamics *on* networks, my goal is to characterize systemic biases found in key network metrics resulting from partial observation of temporal networks.
#64 What is the effect of opinion change and switching cost on Technological Lock-Ins?
Author: Tamer Khraisha
Most of the existing literature on the problem of technological Lock-Ins assumes that once the system reaches a “tipping point” in terms of the number of users of a technology, successive agents will prefer to adopt the same technology, thus causing the system to lock-into that choice. However, the majority of these models are based on two strict assumptions, namely that agents are able to observe the choices of the entire population and, second, agents do not change their opinion. In this research I show that by locating agents on a network and introducing a dynamical component in the system, i.e. agents changing their opinion, the system exhibits less path dependency and, as a consequence, lock-in effects decrease. Initially, we have two technologies and two types of agents. At each time step, an agent is introduced whose type is chosen randomly and the decision of what technology to adopt will depend on two factors: First, his\her preference for one or the other technology and Second, information about the choices of the rest of the network. Information is local if it is received from neighbors, and global if it is received from next neighbors. To include local and global information, I model the information as a convex combination of neighbors’ choices with values of coefficients decreasing according to a scaling law as we move from immediate neighbors to neighbors of neighbors etc. After visiting each agent, I choose randomly a node, and make it change it's opinion, but paying a switching cost. The goal is to understand the effect of switching cost and position in the network on the lock in. To test the effect of network topology and local information on lock-ins and lock-outs, the model will be tested on different networks, including random, small-world and Scale-free networks.
Home |
Calls |
Programme |
Registration |
Contact |
Copyright CCS © 2016
