Search

We further develop water computing as a variant of P systems. We propose an improved modular design, which duplicates the main water flows by associated control flows. We first solve the three open problems of the previous design by demonstrating: how functions can be stacked without a combinatorial explosion of valves; how termination of the system can be detected; and how to reset the system. We then prove that the system is Turing complete by modelling the construction of $$\mu$$ -recursive functions. The new system is based on directed acyclic graphs, where tanks are nodes and pipes are arcs; there are no loops anymore, waterfalls strictly in a ‘top down’ direction. Finally, we demonstrate how our water tank system can be viewed as a restricted version of cP systems. We conclude with a list of further challenging problems.

The topology of chemical reaction networks is commonly treated as a static structure. This might be sufficient if substrate concentrations and kinetic parameter values exclusively determine the behaviour of all considered reactions. In contrast, numerous phenomena observed in life sciences imply a different nature by dynamical composition of reaction schemes. Single reactions or functional groups of reactions (modules) become activated or deactivated by external signals such as light intensity while the system is in operation. In other scenarios, reactions emerge or disappear while modules can connect to each other or disconnect due to presence or absence of corresponding trigger signals. We capture dynamical reaction network structures by an extended version of deterministic P modules with evaluation of trigger signals which facilitates detailed in-silico simulation studies and hence an easier understanding and prediction of complex biological systems. A case study dedicated to photosynthesis in plants demonstrates its usefulness beyond pure employment of ordinary differential equations by consideration of events, non-differentiable external trigger signals, and thresholds which collaterally modify the underlying reaction scheme.

We introduce water tank systems as a new class of membrane systems inspired by a decentrally controlled circulation of water or other liquids throughout cells called tanks and capillaries called pipes. To our best knowledge, this is the first proposal addressing the behavioural principle of floating and stored water for modelling of information processing in terms of membrane computing. The volume of water within a tank stands for a non-negative rational value when acting in an analogue computation or it can be interpreted in a binary manner by distinction of “(nearly) full” or “(nearly) empty”. Water tanks might be interconnected by pipes for directed transport of water. Each pipe can be equipped with valves which in turn either fully open or fully close the hosting pipe according to permanent measurements whether the filling level in a dedicated water tank exceeds a certain threshold or not. We demonstrate dedicated water tank systems together with simulation case studies: a ring oscillator for generation of clock signals and for iteratively making available amounts of water in a cyclic scheme, analogue arithmetics by implementation of addition, non-negative subtraction, division, and multiplication complemented by systems in binary mode for implementation of selected logic gates.

Biological information processing is based on natural laws at a molecular level. Resulting principles make use of dedicated chemical reactions, mechanisms for transportation of biomolecules, and forces among molecules and their environment mainly induced by electric charges and by movement in local space. Since these interactions are typically located in a liquid surrounded by a membrane or a barrier, a dynamical molecular system emerges whose behavior follows the elementary rules of thermodynamics and molecular mechanics. For fine-grained modelling and simulation of processes and implementations of membrane computing, we strive for a corresponding software tool able to manage a multiple particle system in time and space at a medium level of abstraction and with no need of abstract parameter fitting. To this end, we provide the Java Environment for Nature-inspired Approaches (JENA) as a modular, configurable, and extendable platform toward a virtual laboratory and a virtual cell complementing more abstract and more idealized approaches in membrane computing. We give an introduction to JENA and its features and capabilities from the user’s perspective and from a technical point of view. Four illustrative case studies (chemical Lotka-Volterra oscillator, electrophoresis, centrifugation, neural signal transduction across synaptic cleft) demonstrate JENA’s practicability and descriptive capacity.

This chapter presents the Infobiotics Workbench (IBW), an integrated software suite developed for computational systems biology. The tool is built upon stochastic P systems, a probabilistic extension of P systems, as modelling framework. The platform utilizes computer-aided modelling and analysis of biological systems through simulation, verification, and optimization. IBW allows modelling and analyzing not only cell-level behavior but also multicompartmental population dynamics. This enables comparing between macroscopic and mesoscopic interpretations of molecular interaction networks and investigating temporospatial phenomena in multicellular systems. These capabilities make IBW a useful, coherent, and comprehensive in silico tool for systems biology research.

It is a long-cherished wish to implement numerical P systems (NPS) on a parallel architecture so that its large-scale parallelism can be exploited to speedup computation tremendously. Field-programmable gate array (FPGA) is a reconfigurable hardware in which operations are triggered so synchronized by edge or level of activating signals, making it an eligible platform to implement NPS and its variant, enzymatic numerical P system (ENPS). In this chapter, (E)NPS-based robot controllers and path planning algorithm are implemented in FPGA, achieving a speedup of 105 and 104 order of magnitude compared to software simulation. FPGA hardened (E)NPS in this research can be regarded as a heterogeneous multicore processor since membranes inside work as processing units which possess different functions.

This chapter presents several representative applications implemented with P systems software simulators, including automatic design of cell-like and spiking neural P systems with P-Lingua, modelling real ecosystems with MeCoSim, and robot motion planning. The automatic design of P systems uses evolutionary computing approaches to heuristically select the best models satisfying the predefined requirements from a population of candidate P systems, where each P system is evaluated on the P-Lingua platform according to an evaluation function. The models for investigating population dynamics of real ecosystems are realized with MeCoSim. Robot motion planning is implemented by simulating rapidly exploring random trees algorithm within the P system framework.

The performance of P system simulators is key when deploying these models in real applications, or when pushing the limits of efficiency. In this concern, HPC is the key technology to accelerate the simulation of P systems, and specially, GPU computing is a good alternative since they provide a highly parallel device with a shared memory system and a flexible programming. GPUs are currently the enabling technology for trending areas such as Deep Learning. So far, many GPU simulators have been developed using CUDA, the first general-purpose programming model for GPUs. In this chapter, we introduce the concepts behind GPU computing and a taxonomy of GPU-based simulators: generic, specific, and adaptive simulation.

Membrane computing is a broad and flexible paradigm, where many types of computing models have been conceived since the inception of the field—around 20 years ago. It was clear from its early days that a deep understanding of the dynamics of the computational devices of the discipline (so-called P systems) required the development of support simulation tools. Different approaches have been followed in this context, but a historical overview is out of scope here. This chapter mainly focuses on describing P-Lingua framework, probably the most widely used product for the specification and simulation of different types of P systems, consequently leading to a very prolific scientific production. A high-level tool for virtual experiments, MeCoSim, is also presented.

Belousov-Zhabotinsky (BZ) reactions exhibit spiking oscillations over time and in space. Treated as an in-vitro model system for chemical transmission of information via frequency encoding, they came into the focus of systems biology. Conducted in a Petri dish, monitoring of BZ reactions induces comprehensive video material and image sequences that need to be analysed in order to understand the reaction scheme and its parameters including side effects in detail. Aimed by the objective to relieve the biologist from this dreary task, we present a method to automate the identification and localisation of BZ oscillatory spots in a Petri dish by means of optic flow. As defined by Horn and Schunck, optic flow is the distribution of apparent movement velocities of brightness patterns in an image sequence. We extend the standard algorithm and adjust its parameters to make it applicable for automated analysis of BZ reactions in terms of membrane computing as molecular information processing units in space. Our approach introduces methods to cope with perturbations like different kinds of noise typically occurring by undesired alterations within brightness patterns caused by environmental influences and interference of different oscillatory spots. Current work in progress addresses estimation of propagation velocities for expanding concentric rings of each spot.

The capability of self-organisation belongs to the most fascinating features of many living organisms. It results in formation and continuous adjustment of dedicated spatial structures which in turn can sustain a high fitness and efficient use of resources even if environmental conditions or internal factors tend to vary. Spatial structures in this context might for instance incorporate topological arrangements of cellular compartments and filaments towards fast and effective signal transduction. Due to its discrete nature, the P systems approach represents an ideal candidate in order to capture emergence and evolution of topologies composed of membranes passable by molecular particles. We introduce grid-exploring P systems in which generalised membranes form the grid elements keeping the grid structure variable. Particles initially placed at different positions of the grid’s boundary individually run through the grid visiting a sequence of designated membranes in which they become successively processed. Using artificial evolution, the arrangement of membranes within the grid becomes optimised for shortening the total time duration necessary for complete passage and processing of all particles. Interestingly, the corresponding framework comprises numerous practical applications beyond modelling of biological self-organisation. When replacing membranes by queue-based treads, tools, or processing units and particles by customers, workpieces, or raw products, we obtain a multi-purpose optimisation strategy along with a simulation framework. Three case studies from cell signalling, retail industry, and manufacturing demonstrate various benefits from the concept.

Endogenous circadian rhythms allow living organisms to anticipate daily variations in their natural environment. Temperature regulation and entrainment mechanisms of circadian clocks are still poorly understood. To better understand the molecular basis of these processes, we built a mathematical model based on experimental data examining temperature regulation of the circadian RNA-binding protein CHLAMY1 from the unicellular green alga Chlamydomonas reinhardtii, simulating the effect of temperature on the rates by applying the Arrhenius equation. Using numerical simulations, we demonstrate that our model is temperature-compensated and can be entrained to temperature cycles of various length and amplitude. The range of periods that allow entrainment of the model depends on the shape of the temperature cycles and is larger for sinusoidal compared to rectangular temperature curves. We show that the response to temperature of protein (de)phosphorylation rates play a key role in facilitating temperature entrainment of the oscillator in Chlamydomonas reinhardtii. We systematically investigated the response of our model to single temperature pulses to explain experimentally observed phase response curves.

The construction of molecular-scale machines requires novel paradigms for their programming. Here, we assume a scenario of distributed devices that process in-formation by chemical reactions and that communicate by exchanging molecules. Programming such a distributed system requires specifing reaction rules as well as exchange rules. Here, we present an approach that helps to guide the manual construction of distributed chemical programs. We show how chemical organization theory can assist a programmer in predicting the behavior of the program. The basic idea is that a computation should be understood as a movement between chemical organizations, which are closed and self-maintaining sets of molecular species. When sticking to that design principle, fine-tuning of kinetic laws becomes less important. We demonstrate the approach by a novel chemical program that solves the maximal independent set problem on a distributed system without any central control—a typical situation in ad-hoc networks. We show that the computational result, which emerges from many local reaction events, can be explained in terms of chemical organizations, which assures robustness and low sensitivity to the choice of kinetic parameters. DOI: 10.4018/jnmc.2009120901 2 International Journal of Nanotechnology and Molecular Computation, 1(4), 1-19, October-December 2009 Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. ‘dry’ nanotechnology materials has been argued (Merkle, 2000) to be the drawback. Despite the rapid development of nanotechnology, wet-lab experiments are commonly exercised to operate the boolean logics (de Silva & Uchiyama, 2007). DNA computing demonstrated by Adleman (1994) already addressed that limitation of imperative computation paradigms by utilizing particular operation modes of DNA molecules. Assuming DNA as data carrier, up to 10 bytes can be saved and operated simultaneously within one liter of liquid providing a storage density of 1 3 bit nm / (Paun, Rozenberg, & Salomaa, 1998). One Joule allows up to10 molecular operations on DNA (Pisanti, 1998). This highly parallelized operation on DNA strands with high data density is the key characteristics of the DNA computation approach. The significance, we note here, is that the computation model is in concert with computation medium. In the nano-scale world, molecules are regarded to constitute a medium, and chemical reactions play an important role in biological information processing principles (Kuppers, 1990). Employing molecules and reaction rules as a metaphor, thus, novel computation paradigms have been explored (Banâtre & Metayer, 1986; Paun, 2002; Banâtre, Fradet, & Redenac, 2004; Tschudin, 2003; Berry & Boudo, 1992). Essentially, those chemical computing models refer the elementary units as molecules, and the operations are described in the form of reactions among those molecules. Given the inputs of the computation as the initial configuration of reaction vessels or reactors, the outputs emerge from local interactions in accordance with the given reaction rules (Banzhaf, Dittrich, & Rauhe, 1996). In these chemical computing models, programming corresponds to designing the reaction rules at the microscopic levels, and the desired computational result emerges at the macroscopic levels as a global systems’ state. The relation between those two levels is highly non-linear, and thus the question for effective programming techniques arises. It seems scarcely possible in this context to predict the macro behavior from the micro rules because of the parallel operations of the reaction rules that are possibly tangled in a complex manner. A common approach to this difficulty is to find a mapping from a known computation model like a Turing machine or a finite state automaton (Paun, 2002; Rothemund, 1996). Conrad (1989) argued that the conventional computers innately differ from natural molecular systems, such as brain or enzymes, and the natural molecular systems are (from today’s perspective) not highly programmable (Conrad, 1985). The conventional digital computers are designed to achieve high programmability by restricting the behaviors of computational entities, and the natural molecular systems operate to exploit the useful properties of the medium. In this article, we present a programming technique utilizing a notion of chemical organization (Dittrich & Speroni di Fenizio, 2007). Our programming approach is specialized for chemical computing paradigms, and thus the computational medium in the nano-scale world is highly respected. The maximal independent set (MIS) problem (Luby, 1986) is employed for our case story. Dynamical behaviors of the chemical programs are simulated for the validation purpose. Finally, we also discuss that changing the analysis coverage is fruitful. approaChes to ChemiCal proGramminG In general, there are several approaches to obtain a chemical program capable of solving a predefined computational problem. Here, we distinguish optimization versus construction. Optimization subsumes heuristically driven techniques: A more or less randomly chosen reaction network becomes successively improved e.g. by evolutionary methods (Ziegler & Banzhaf, 2001), learning inspired by neural networks, simulated annealing, or tabu search (Landweber, 2002). Within the optimization process, the reaction network topology as well as reaction parameters are fitted according to the desired behavior (Deckard & Sauro, 2004). 17 more pages are available in the full version of this document, which may be purchased using the "Add to Cart" button on the publisher's webpage: www.igi-global.com/article/organization-oriented-chemicalprogramming-distributed/40362

Biocomputing emerged as a promising paradigm capable of coping efficiently with challenges of programming decentralized but concerted reaction systems. The chemical programming metaphor subsumes different encoding techniques into molecular or spatial structures in conjunction with artificial reaction networks. Here, a variety of supplementary assumptions like predefined polymeric sequences or availability of inhibiting reactions is frequently used. Inspired by the idea to build chemical computers based on minimal requirements in chemistry from a theoretical perspective, we introduce a pure chemical register machine model operating on binary numbers. The register machine architecture is composed of reaction network motifs acting as fast switching logic gates, oscillators, and self-reproducible bit storage units. The dynamical machine behavior consistently employs mass-action kinetics. Two case studies, calculating the maximum of three natural numbers as well as numerical addition, illustrate the practicability of the design along with dynamical simulations.

Temperature represents an elementary environmental stimulus crucial for survival and fitness of organisms. Molecular membrane-based mechanisms for temperature sensing and behavioral response seem to be among the oldest principles of biological information processing. It is believed that some archaea – early microbes prior to bacteria and eukaryotes – developed thermoreceptors. In addition, they were able to maintain a circadian clock, a biochemical oscillatory system whose periodicity reflects a daily rhythm. Both features on their own, but especially their combination, gives raise for effective evolutionary advantage. Along with the notion of applied systems biology, we explore capabilities of resulting reaction models by exploitation of deterministic P modules and their dynamical coupling by means of simulation studies. Our findings indicate that a minimalistic circadian clock equipped with a chemical temperature sensor enables robust and practicable entrainment to an external daily temperature rhythm induced by the sun in contrast to a clock variant without thermoreceptor. Having a more adaptable circadian clock, archaea comprise better preconditions to populate larger oceanic regions from the equator towards the poles. From a modelling point of view, we incorporate the global quantity temperature and its effect on reaction velocity according to Arrhenius’ equation into the framework of deterministic P modules.

Blotting turns out to be a rather common and effective approach in molecular information processing. An initial pool of molecules considered as sets of individual data becomes spatially separated according to the presence or absence of specific attributes like weight index or chemical groups and labels. In this connection, molecules with similar properties form a spot or blot. Finally, each blot can be visualised or analysed revealing a corresponding score index or count from the number of accumulated molecules. The entire variety of blots which emerge over time provides crucial and condensed information about the molecular system under study. Inspired by the idea to obtain significant data reduction while keeping the essential characteristics of the molecular system as output, we introduce blotting membrane systems as a modelling framework open for numerous applications in data mining. By means of three dedicated case studies, we demonstrate its descriptive capability from an explorative point of view. Our case studies address particle-based numerical integration, which suggests a model for the synchronised 17-year life cycle of Magicicadas. Furthermore, we exemplify electrophoresis as a way to carry out a variant of bucket sort.

Im Zentrum der Betrachtungen zum DNA-Computing steht die Frage nach den Chancen und Grenzen dieses neuen Berechnungsmodells, nachdem in den letzten Jahren eine rasante Entwicklung auf das Thema aufmerksam machte. Neben beachtlichen theoretischen Untersuchungen zum “Rechnen im Reagenzglas” werden auch laborpraktische Implementierungen favorisiert. An der TU Dresden wurde in interdisziplinarer Arbeit ein Integer-Rucksackproblem mittels eines DNA-Algorithmus im Labor gelost und dabei eine Vielzahl molekularbiologischer Operationen analysiert. Mit Hilfe dieses Satzes von Operationen gelang eine universelle und labornahe Modellierung des DNA-Computing. Hierbei angewandte Techniken und Methoden werden vorgestellt und bewertet. Die Beschreibung des DNA-Algorithmus zeigt, wie sich Einzeloperationen vorteilhaft zu Operationsfolgen zusammensetzen lassen und gemeinsam mit einer geeigneten DNA-Kodierung der Eingangsdaten zur Losung des Problems im Labor fuhren. Erstmalig wurden hierbei naturliche Zahlen verarbeitet. Die Arbeitsgemeinschaft DNA-Computing Dresden konzentriert sich auf Aufgabenstellungen, die formale Modelle des DNA-Computing mit uberzeugenden Laborimplementierungen verbinden.

This chapter addresses three coordinated chronobiological studies demonstrating the convergence of experimental observations, fine-grained spatio-temporal modelling, and predictive simulation. Due to the discrete manner of molecular assembly in cell signalling and gene regulation, we define a framework of membrane systems equipped with discretised forms of reaction kinetics in concert with variable intramolecular structures. Our first study is dedicated to circadian clocks inducing daily biological rhythms. As an inspiring example, the KaiABC core oscillator reaches its functionality by cyclically varying protein structures. Within our second study, we present a meta-model of an entire circadian clockwork able to adapt its oscillation to an external stimulus in terms of a frequency control system acting in a phase-locked loop. Substrate concentration courses resulting from gene expression reflect its oscillatory behaviour utilised in a periodical trigger for subsequent processes. In this context, our third study cytometrically quantifies the dynamical behaviour of a bistable toggle switch resulting from mutual gene regulation.