# Introduction n spite of all the recent emphasis and advancements in systems biology, synthetic biology, and network science about modelling of gene networks, protein networks, metabolic and signaling networks, etc. some of the most important computational properties of membrane cells have not been grappled and "abstracted" et: scalability, tissular differentiation, and morphogenesis -i.e., the capability to informationally transcend the cellular level and organize higher level information processes by means of heterogeneous populations of membrane cells organized as "computational tissues and organs". Synthetic biology has become extraordinarily active in the manufacture of very simple and robust models and simulations tailored to the realization problems of circuits and modules in vivo, mostly addressed to prokaryotic systems. In the first wave of these studies, very basic elements such as promoters, transcription factors, and repressors were combined to form small modules with specified behaviors. Currently modules include switches, cascades, pulse generators, oscillators, spatial patterns, and logic formulas (Purnick & Weiss, 2009). The second wave of synthetic biology is integrating basic parts and modules to create systemslevel circuitry. genomes and synthetic life organisms are envisioned, and application-oriented systems are contemplated. Different computational tools and programming abstractions are actively developed (the Registry of Standard Biological Parts; the Growing Point Language GLP; the Origami Shape Language OSL, the PROTO bio programming language, etc. See details at the Open Wetware site). Evolving cell models of prokaryotes have also been addressed (Cao et al., 2010). (Bashor et al., 2010). As some have put, "systems broaden the scope of synthetic biology designing synthetic circuits to operate in reliably in the context of differentiating and morphologically complex membrane cells present unique challenges and opportunities for progress in the field" (Haynes & Silver, 2009). However, very few synthetic biology researchers do contemplate using systems. In systems biology, a plethora of modelling developments have been built around signaling pathways, cell cycle control, topologies of protein networks, transcriptional networks, etc. There is a relatively well consolidated thinking, in part due to traditional physiology and to systems science and control theory which were at the origins of this new field, of going "from genes to membrane cells to the whole organ" as D. Noble has done for heart models (Noble, 2002). The integration of proteins to organs has also been promoted by bioinformatic-related projects such as the "Physiome Project" (Hunter et al., 2002). Important works have been done in the vicinity of "network science" in order to make sense of gene networks, protein networks, transcription networks, complexes formation, etc. For instance, about how is dynamically or2anized modularity in the yeast proteinprotein interaction network (Han et al., 2004), it was uncovered that two types of "hub" contribute to the organized modularity of the proteome: "party" hubs which interact with their partners simultaneously, and "date" hubs, which bind their different partners at different times and locations (we will see later on the importance of the discussion on "modularity" in the evodevo field). Predictive models of mammalian membrane cells have been described using graph theory, assembling networks and integrative procedures (Ma'yan et al., 2005). Important systems biology compilations and far-reaching cellular models have been made by Balazsi et al. (2005), Kitano (see in Oda et al., 2004), Luscombe et al. (2004), Huh et al., (2010) ...It has to be emphasized that concerning the views advocated in this proposal, most of systems biology works depart from the goal of "abstracting computational power out from systems" and focus instead on "applying computational power to analyze the organization of systems." Notwithstanding the foregoing, studies such as A. Dan chin (2009) on bacteria as computers making computers, and by Ray et al. (2010) on the operating system of bacteria could be considered as forerunners in the former direction. In the science of development (the "evo-devo" discipline) most of the emphasis has been on modularity. What it exactly means in developmental terms is still a matter of controversy (Schlosser & Wagner, 2004;Carroll, 2005;Sprinzak, 2010); but undoubtedly modularity refers to the capability of cellular networks to dissociate networked processes at a lower level and to recombine or redeploy them at the higher level of the multicellular organism. Thanks to the cellular signaling system, the genetic switches, the cytoskeleton, and some other topobiological mechanisms (Edelman, 1988;Szathmary, 2001), the unitary network of cellular processes integrated into the cell-cycle may be broken down into coherent modules and be performed separately in different membrane cells within differently specialized tissues (Palmer, 2004). This implies a flexible organization for the deployment of biomolecular processing modules, which actually are "cut" differently in each tissue along the developmental process, due also to chromatin remodelling during development (Ho & Crabtree, 2010). Interestingly, not only differentiation but also morphology becomes an instance of the scalable "modular" processing, throughout the "tensegrity" emergent property and the ontogenetic arrangement of symmetry breakings in a force field. The emergence of cellular bauplans where signaling, force fields, and cytoskeletal mechanical modes conspire together to create but a few basic morphologies for membrane cells, depending also on the populations present, seems to be another important consequence (Mojica et al., 2009). Interestingly, complex morphologies obtained out from Turing diffusion model have been cogently discussed as a result of cell-to-cell developmental interactions (Kondo & Miura, 2010) .Currently, the evo-devo field accumulates a considerable mass of biomolecular-or2anizationfacts, poorly conceptualized yet, to be computationally "abstracted" in the perspective of MCA advancement. equations used up to now. Proteins and other biomolecules become molecular "automata" and the aggregate behavior that emerges out from these models is the combinatorial expression of all those automata doing their specific micro-functions (Blow, 2009). This approach shows promise for "evolvable" advancement of network models endowed with the flexible modularity property. It is somehow close to the already mentioned predictive models of mammalian membrane cells that are using graph theory, assembling networks and integrative procedures (Mayan et al., 2005). New generations of cellular models (of "automata") have been developed too, with powerful data content and with potential for modelling multi-cellular systems in a general way, supporting userfriendly in silicon experimentation and discovery of emergent properties (Amir- Kroll et al., 2008). Under the approach of Artificial Embryology, a developmental system has been obtained by means of cellular automata systems capable of following "rewriting rules" procedures, emulating elementary morphologies and multicellular distributions (Federici & Downing, 2006). As for the developments in molecular Biocomputing, the idea that bio-molecules (DNA, RNA, proteins) might be used for computing already emerged in the fifties and was reconsidered periodically with more and more arguments which made it more viable. But the definitive confirmation came in 1994 (Adleman, 1994) when L. Adleman successfully accomplished the first experimental close connection between molecular biology and computer science. He described how a small instance of a computationally intractable problem might be solved via a massively parallel random search using molecular biology methods. An important part of this project is focusing on bio-inspired models of computation abstracted from the very complex networks in living systems. Its goal is to investigate several aspects of these models particularly focused on connections between theoretical models and natural (biological) networks. The main topics are: Computational aspects (computational power, structural and description complexity). Application aspects (simulation, physical implementation, experimental results, training issues). This part is intended to be a contribution to both Global Computing (which includes neural networks, cellular automata, etc.) and Bio-inspired Computing (as a part of Natural Computing) a new and interdisciplinary field which lies at the crossroads of mathematics, computer science, molecular biology and linguistics. There are research groups working in similar or connected topics in Europe (Germany, France, Spain, Holland, Hungary, Romania, Moldavia, Finland, Poland, Austria, Italy), USA, Japan, India, China. In the fields closer to computer science and Biocomputing, it has been important the introduction of the agent based approach (as pioneered by W. Fontana and others), which uses sets of rules to define relationships between cellular components substituting for the simple Boolean networks and differential Several new directions of research have been initiated in the last decade: computing devices inspired from the genome evolution Dassow et al., 2002), membrane systems (Nun, 2002) with an explosive development, evolutionary systems based on the behavior of cell populations (Ardelean et al.,2004) computing models simulating the process of gene assembly in ciliates (Ehrenfeucht et al., 2003), (Freund et al., 2002), (Istrail et al., 2007), networks of evolutionary processors (Manea et al., 2010), etc. The joint efforts of biologists and computer scientists led to a new concept, namely the template-guided recombination which seems to offer a "bioware" implementation of the process of gene assembly (Angeleska et al., 2007), (Presscot et al., 2003). Swarm computation is mainly based on the same idea: a swarm is a group of mobile biological organisms wherein each individual communicates with others by acting on its local environment (Engelbrecht, 2005). A computational model based on multiset rewriting is used to simulate the emergence of autocatalytic cycles which are often found in living systems is proposed in (Suzuki&Tanaka, 1997). The use of X-machines, a variant of finite state machines with much more computational power, is used to model immunological pathways (Holcombe&Be11,1998). Moreover, (Istrail et al., 2007) proposes a new paradigm, "genomic computer", where the entire genomic regulatory system is viewed as a computational system and not only the immune system as it was considered in (Dasgupta,1998). Many works were devoted to the study of a wide range of operations on biological sequences in vivo and in vitro (bio-operations): PA-matching, annealing, Watson-Crick superposition, transposition, inversion, duplication, translocation, etc. (Karp,2002) gives an overview of the most important and attractive problems for mathematicians coming from genomics and molecular biology. Last but not least, the molecular computing contributed to the understanding of selfassembly which is one of the key concepts in nanoscience (Reif&LaBean,2007). The new sub-area of Computation Theory called Bio-Inspired Computing is very dynamic. After approximately 12 years the bibliography about Bio-Inspired Computing counts nearly 1000 papers and several books and grows rapidly each year. These papers were published in either computer science forums or biological ones. Many prestigious international journals hosted special issues but new journals were also created: a permanent column in the # II. # Membrane Computing A Transition P System of degree n 1 > n is a construct ( ) 0 1 1 1 i ) , R ) , . . ( R ( , , . . , , , V n , n , n ? ? ? ? µ = ? Where: V is an alphabet; its elements are called objects; ? is a membrane structure of degree n, with the membranes and the regions labeled in a one-to-one manner with elements in a given set; in this section we always use the labels 1, 2, n; n i i ? ? ? 1? ? ? ? 2 2 1 1 2 2 2 2 2 2 1 1 2 1 1 1 2 2 1 1 1 1 P-systems evolve, which makes it change upon time; therefore, it is a dynamic system. Every time that there is a change on the p-system we will say that the psystem is in a new transition. The step from one transition to another one will be referred to as an evolutionary step, and the set of all evolutionary steps will be named computation. Processes within the psystem will be acting in a massively parallel and nondeterministic manner. (Similar to the way the living cells process and combine information). We will say that the computation has been successful if: III. # The Upgrade The proposal is a new computational paradigm based on Membrane cells, scalable ones which are capable to produce "computational tissues and organs". The organization of such computational tissues and organs is inspired by the emerging informational properties of biomolecular networks and will be based on scalable "membrane cells" guided by functional rules similar to the biological ones (molecular recognition, self-assembly and topo biology-theory rules). The direct inspiration from the membrane cells is precisely the breakthrough of the MCA project. By building computational tissues our proposal makes an evolutionary jump with respect of today research in this field, mainly focused on aggregates of unicellular organisms (e.g. bacteria). Far from modelling and simulating the cellular processes, our computational paradigm will be a clear abstraction of the basic mechanisms and computational capabilities of the membrane cells and tissues, in order to solve complex problems in a new (bioinspired) way. Real tissues display far more complex properties (emergent properties) than the sum of the properties of the individual membrane cells they are made from. In the same way, the emergent properties and functions of our membrane cells and computational tissues will be used for the resolution of real problems, impossible to be appropriately solved by conventional methods: not only biological morphogenesis, but also evolution of economic systems and prediction of crisis, optimization of "industrial ecologies", analysis of the dynamics of social interactions and conflicts, ecosystem disturbances, etc., that are more complex than combinatorial optimization, as well as other classical NP-Complete ones. Our "membrane cells" will be a species of "proto-membrane cells" and a far objective of the project is also the ex-novosynthesis of " membrane cells" and tissues performing as living computational biomolecular networks. The lon2-term vision that motivates this breakthrough is to build new information processing devices with evolving capabilities, which will adapt themselves to the complexity of the problems. In particular, we foresee a synthetic approach to build computational membrane cells and tissues, and to create computational bio-inspired devices of higher complexity (tissues-organs). A far future objective of the project goes beyond the mathematical, software and hardware tools. It is to obtain in lab synthesized "living" information processing systems based on artificial "membrane cells" and hybrid systems combining living components (our "synthesized membrane cells") and non-living elements (e.g. silicon-based). MCA approach is the most appropriate to deal with extremely complex problems that will be crucial in the future. It shows potential to go beyond classical Biocomputing strategies such as self-reproducing machines, cellular automata, perceptron's & neural networks, genetic algorithms, adaptive computing, bacteria-based computation, artificial membrane cells, etc. Specifically, a new generation of natural computing could be built, based upon the scalable " membrane cells" with problem solving capacity in very different realms: biomaterials and bioengineering, non-linear parallel processing, design of bioinspired systems, modelling of economic, industrial and financial systems, optimization strategies in social settings, etc. For the achievement of our long-term objectives we need to: analyze the wide amount of existing knowledge regarding one of the deepest sources of biocomputational power, the topological and flexible networking properties of biomolecular scalable modules in membrane cells, realize an abstraction of the basic mechanisms and computational capabilities of the membrane cells both at sub cellular and networking level, and develop formal models to be used in new information processing technologies, basically based on combinatory processes of protein domains and genetic switches, together with cytoskeleton dynamics and topobiology-theory, use the above proposed models to create scalable "/proto membrane cells" and abstract-formal "evolvable" cellular networks and computational tissues & organs endowed with these flexible modularity properties. For our far final objective we need to obtain in lab proof that synthesis of new forms of living" membrane cells" in an inverse process: "membrane cells and tissues" => "theoretical abstract/formal models" => "artificial membrane cells and tissues" => "in lab synthesized living membrane cells" is possible. MCA breakthrou2h is an essential step towards the achievement of our lon2-term vision because it will set the theoretical basis and develop the experimental tools for the creation of the scalable membrane cells, computational tissues and organs (both abstract and living ones). # IV. # MCA System A MCA is a set ?= {? 0 , ? 1 , ? 2 , ? , ? ???1 , ? ?? ) and a set ? of aggregation rules among membranes. The set of aggregation rules are not fully integrated with the evolution rules of a given p-System but establishes the correlation between 2 given membrane models by deciding the way 2 or mere P-systems are being aggregated. The rules can be defined as a Matrix relation ( ) ? ? ? ? ? ? ? ? ? ? ? ? ? ? = ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? n m m n n n m m m u . . . u u k . . . k k u . . . . . . u . . . u . . . u . . . u u u . . . u u , k , . . , k , k 2 1 2 1 2 1 2 2 2 1 2 1 2 1 1 1 2 1 1 Where ?1(k) is the aggregation relation and is defined by the association of n P-systems, k determines the aggregation rules of each component in every psystem Iand Uare the component (objects). Evolution rule application phase. This phase is the one that has been implemented following different techniques. In every region within a p-system, the evolution rules application phase is described as follows: Rules application to a multiset of object in a region is a transforming process of information which has input, output and conditions for making the transformation. Given a region within a p-system, let U= } n i | a { i ? ? 1 be the alphabet of objects, m a multiset of objects over U and R(U,T) a multiset of evolution rules with antecedents in U and targets in T. The input in the region is the initial multiset m. The output is a maximal multiset m'. The transformations have been made based on the application of the evolution rules over m until m' is obtained. Application of evolution rules in each region of P systems involves subtracting objects from the initial multiset by using rules antecedents. Rules used are chosen in a non-deterministic manner. This phase ends when no rule is applicable anymore. The transformation only needs rules antecedents as the consequents are part of the communication phase. # Observation Let N k i ? be the number of times that the rule i r is applied. Therefore, the number of symbols j a which have been consumed after applying the evolution rules a specific number of times will be: ? ? ? ? 2 2 1 1 2 2 2 2 2 2 1 1 2 1 1 1 2 2 1 1 1 1 Maximal multiset is that one that complies with: ? ? m l n i l i m j j i j i u ) u k ( - u 1 1 1 = = = ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? [1] [Arteta,2010] V. # Correction The correction of the system fully relies in the correction of the internal P-system of the MCA. In order to prove the aggregation system is distributed then 2 processes need to be proven. 1. Correction of the formal definition of Transition P-System (Paun , 1998) 2. Correction of the aggregation rules applying to 2 given P-systems. The correction of the second point gets reduced to a deductive demonstration where the aggregation of 2 given P-systems is base case and the generic case of n-P-systems can be seen as the aggregation of n-1 Psystems (inductive case) with a correct aggregation to the last one. Thus, the key is to prove that aggregation of 2 given P-system is a correct process and indeed reinforce the idea of full inherent parallelism and nondeterministic modelling that membrane models are after. Aggregation rule. Let us use a short definition of a given P-System ( ) The result is the Unionof both. Correctness for this operation is also obvious. 0 1 1 1 i ) , R ) , . . ( ? The aggregation of the 2set of the set of the evolution rules ?? 12 is obvious. The result is the Union of both. Correctness for this operation is also obvious. There are 2 factors in the aggregation that are not obvious which are the aggregated Set of regions ?? 12 . This set of regions is constructed in our proposal as supervised and directed by the factor ? that defines the capabilities previously mentioned. This ? is defined dynamically by the nature of problem the MCA is about to fix. i.e.in a problem of sum of squares is not necessary aggregation as 2 independent P-system could calculate their squares [Paun,2001] and send those outputs to a third (obvious) one that calculates the sum of both results. However, for didactic purposes and aggregated solution could be provided in where a MCA is created with 2 Input P-systems. The aggregated would assign equal ? (priority) to both of them, and then either of them could contain the other one. The container P-system process the output of the contained P-system by adding it to an another square number. ? The aggregation of the regions of 2 P-systems would be determined by a priority or hierarchy described by ?. This is a dynamic factor that must be configured right before the problem is dealt with. ? The aggregated P-system will have to work the communication phase after every evolutionary step. This communication phase also fully relies on the hierarchy establish by ? and will operate as normal when the aggregation is complete and the MCA is finished. # a) Inductive case Given a successful aggregation (MCA) of n P-systems MCA (n), is it correct to aggregate n+1 P-systems? The inductive case is a direct consequence of the aggregated property. MCA (n) system becomes a complex P-System with an aggregation of regions according to the ? factor .MCA (n)= let's call the aggregated P-system as ?? ?? ={? 1 , ??, ð??"ð??", ??1}. Once the aggregation is seen as a P-system, aggregating it with another ?? 1 is obvious by applying the base case. # b) Simulations and results We have been performing some simulations in simple problem solving in same traditional computing paradigm for small problems clearly aggregation is not necessary, although the advantage of this proposal shows up, when the complexity of the problem increases. Theoretically a fully and corrected aggregated Solution (A whole MCS) would overweight Other problems, especially those that requires sub solutions that are part of optimization techniques would be required to establish a clear hierarchy in the aggregation of MCA. Thus: the cost of the calculation of ? and he redesign of the membrane system that can always occur during compiling time anyways. The analysis is very direct. The simulations are running in the same platform and just focuses in performance time based. All problems are considered simple problems due to the limitations of processing a complex problem with a complex set of aggregation rules which will jeopardize the accuracy of the analysis. Nevertheless, it is indicative to see that there is a variation in the performance when the level of complexity slightly increases which suggest that aggregation can be a good approach when the level of complexity increases. # VI. # Conclusions Membrane computing has been growing since George Paun defined it in 1998. Since then new variations have been suggested to try to fit this model to new realities. The main goal for this unconventional paradigm is to improve the performance of the traditional algorithms due to the inherent limitation of the model. Simulations are still a big part of membrane computing and they are useful to extract right conclusions about the new model. In particular, this model is a great candidate to be applied to complex models that require an aggregated solution that is part of other sub solution whole super solutions as long as the defined rules in the MCA are followed. The aggregation factor that is linked to the minimal membrane cells is the component that complement the use membrane computing as a whole and as unite aggregated model. As the creation of this factor generates difficulties because it depends on the nature of the problem, it does not damage the performance during the execution as the factor is calculated in compiling time. New techniques to atomize the generation of ? as this could create a complete dynamic model that fully adjust to the problem and create the right MCA. The necessity of opening the line of research is out of question. The field is growing and new experiments are required. MCA systems are provided as a natural solution to upgrade the nature of membrane computing by not only taking advantage of the properties of the membrane cells but by the way these cells are aggregated. The future work will be involving complex problems in complex aggregated structures, so the analysis can be more relevant. Nevertheless, the evidence points out that aggregation is a natural solution to deal with complex problems that nowadays are being processed by conventional approaches such as backtracking or dynamic programming. ![Bulletin of the European Association for Theoretical Computer Science, Natural Computing, Journal of Unconventional Computing, Theoretical Computer Science-Track C, Theory of Natural Computing, etc. Each year several conferences are devoted mainly to this area: DNA Based Computers (15 editions so far), Unconventional Models of Computation (8 editions so far), Workshop in Membrane Computing (11 editions so far), International Work-Conference on Artificial and Natural Networks (9 editions so far), International Work-Conference on the Interplay between Natural and Artificial Computation (5 editions so far), Pacific Symposium on Biocomputing (first edition in 1995). Regarding applicative models there are many attempts to update Cells computing paradigm in Arteta (2009) Arteta (2010) Arteta (2011) Arteta (2012) Arteta (2013) Arteta (2014), Frutos (2009) and Frutos (2013) among others.](image-2.png "") 11![Fig. 1: P-system structure Definition Multiset of objectsLet U be a finite and not empty set of objects and N the set of natural numbers. A multiset of objects is defined as a mapping:](image-3.png ", 1 , 1 ,a") ![Note: Initial Multiset is the multiset existing within a given region in where no application of evolution rules has occurred yet.Definition Evolution rule with objects in U and targets in T Evolution rule with objects in U and targets in on 'c' will be referred a s the consequent of the evolution rule 'r'.Note: The set of evolution rules with objects in Uand targets in T is represented by R (U, T). Definition Multiplicity of an object in a multiset of objects M(U) Let U a i ? be an object and let ) U ( M m ? be a multiset of objects. The multiplicity of an object is defined over a multiset of objects such as: Definition Multiplicity of an object in an evolution rule r Let U a i ? be an object and let ) an object is defined over an evolution rules such as: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .](image-4.png "") 1![Given a region R and alphabet of objects U, and R (U, T) set of evolution rules over U and targets in T.](image-5.png "1 Definition") ?=V,µ,?R , . . , ( , n ?, ?n, ?nBase case. Given 2Transition P-system?? 1 © 2019 Global Journals * Molecular computation of solutions to combinatorial problems ; ) LMAdelman HAmir-Kroll ASadot IRCohen DHarel Theoretical Computer Science 226 1994. 1994. 2008 Science * ) an isomorphism based algorithm to solve complex problems), Alberto Arteta ;Arteta FLuis 2018 * JuanMingo CastellanosWseas Transactions on Information Science and Applications * New Approach to Optimize Membrane Systems Arteta Journal of bioinformatics 2014. Jan 2014 * Iss * Technical P-systems: Operating in the stock markets with Transition P-systems Arteta 2013 * AlbertoArteta AngelLuisCastellanos Nuria Gómez Blas "Information Research and Applications 2013 * Membrane systems working with the P-factor: best strategy to solve complex problems Arteta factor: 1.253 JCR Indexed in ISI Web of Science AdvancedScience Letters Alberto Arteta Luis Fernando Mingo NuriaGomez SciVerse 2012. 2010 * 10.1166/asl.2013.4453 19 * 2 Meia Systems: membrane encrypted information application systems Arteta Natural Information Technologies Madrid International Journal "Information Theories and Applications 2012. 2012 19 pages cirrus Google Scholar * 3 Solving Diophantine equations with a parallel membrane computing model AlbertoArteta NuriaArteta Gomez Rafael Gonzalo International Journal "Information Models and Analyses 1 2012 printed) 6 páges * Engineering of Application of Artificial Intelligence, Alberto Arteta Luis Fernando Mingo Nuria Gomez Arteta JCR Indexed 0952-1976 24 6 2011. September 2011. 2010 Solving complex problems with a Bioinspired model. Impact factor 1.44 Q1 * Intelligent P-systems AlbertoArteta AngelLuisArteta JoseLuisCastellanos Sanchez I TECH 2010, International Conference ITHEA Journal 07/2010 " New Trends in Information Technologies Sofia; Bulgaria 2010 * P-systems: Study of randomness when applying evolution rules AlbertoArteta LuisArteta Fernández International Book Series Information Science and Computing 2009. 2009 9 * Analysis of P-systems under multiagents perspective AlbertoArteta AngelArteta JuanGoñi Castellanos2009 Information Science and Computing 1313- 0455 12 2 2009 Pages * RNA-guided DNA assembly AAngeleska J. Theoretical Biology 248 2007 * A computational model for cell differentiation IArdelean Bio Systems 76 2004 * Topological units of environmental signal processing in the transcriptional regulatory network of Escherichia coli GBalazsi A.-LBarabasi ZNOltvai PNAS 102 22 2005 * Rewiring Cells: Synthetic Biology as a Tool to Interrogate the Organizational Principles of Living Systems CJBashor AAHorwitz SJPeisajovich WALim Annual Review of Biophysics 39 2010 * Systems biology: Untangling the protein web NBlow Nature 460 2009 * Wetware: A Computer in Every Living Cell: The Computer in Every Living Cell DBray 2009 Yale University Press New Haven * The Relationship Between Matter and Life RABrooks Nature 409 2001 * Evolving cell models for systems and synthetic biology HCao FJRomero-Campero SHeeb MCamara NKrasnogor Systems and Synthetic Biology 4 1 2010 * Endless Forms Most Beautiful SBCarroll 2005 Norton, Chicago * Bacteria as computers making computers ADanchin FEMS Microbiol Rev 33 1 2009 * Artificial Immune Systems and Their Applications DDasgupta Springer 1998 * On some operations suggested by genome evolution JDassow VMitrana Proc. Second Pacific Symposium on Biocomputing RBAltman Second Pacific Symposium on Biocomputing 1997 * Context-free evolutionary grammars and the structural language of nucleic acids JDassow VMitrana ASalomaa BioSystems 43 1997 * Operations and language generating devices suggested by the genome evolution JDassow VMitrana ASalomaa Theoretical Computer Science 270 2002 * Topobiology: An introduction to molecular embriology GMEdelman 1988 Basic Books New York * AEhrenfeucht THarju IPetre DMPrescott GRozenberg Computation in Living Cells: Gene Assembly in Ciliate Springer 2003 * Fundamentals of Computational Swarm Intelligence AEngelbrecht 2005 Wiley and Sons * Evolution and development of a multicellular organism: scalability, resilience, and neutral complexification DFederici KDowning Artif Life 12 3 2006 * On some operations suggested by gene assembly in ciliates RFreund CMartin-Vide VMitrana New Generation Computing 20 2002 * Alberto Arteta Information Technologies and Knowledge LuisFrutos ; Juan Alberto De Frutos CarmenFernández Luengo Madrid) 09/2010 Membrane Computing:Non deterministic technique to calculate extinguished multisets" Information Technologies and KnowledgeISSN 1313-048X 2010. 2010. 2010 10 Improving active rules performance in new P system communication architectures * FernandoFrutos ; Juan Alberto De Frutos Arroyo 10.1007/978-3-540-95885-7_13SJR0.29Q2 Usefulness States in New P System Communication Architecture Lecture Notes in Computer Science 2009. 2009. 2009 5391 * Evidence for dynamically organized modularity in the yeast protein-protein interaction network JHan NBertin THao DSGoldberg GFBeniz LVZhang DDupuy AJ MWalhout MECusick FPRoth MVidal Nature 430 2004 * Membrane Computing Aggregation (MCA): An Upgraded Framework for Transition P-Systems * Eukaryotic systems broaden the scope of synthetic biology KAHaynes PASilver Nature Ho L. & Crabtree G.R. 187 5 2009. 2010 JCB * Computational models of immunological pathways MHolcombe ABell Information Processing in Cells and Tissues MHolcombe RPaton New York and London Plenum Press 1998 * Reconstituting Organ-Level Lung Functions on a Chip DHuh BDMatthews AMammoto MMontoya-Zavala HYHsin DEIngbe Science 25 5986 2010 * The IUPS human physiome project PHunter PRobbins DNoble Eur J Physiol 445 2002 * The regulatory genome and the computer SIstrail BTDe-Leon EDavidson Developmental Biology 310 2007 * Reactiondiffusion model as a framework for understanding biological pattern formation LKari LLandweber RMKarp Notices of the AMS Kondo S. & Miura T. 54 5999 2000. 2002. 2010 American Mathematical Society Science * Genomic analysis of regulatory network dynamics reveals large topological changes NMLuscombe MMBabu HYu MSnyder SATeichmann MGerstein Nature 16 431 2004 * Toward predictive models of mammalian cells AMa' Ayan RDBlitzer RIyengar Annual Review of Biophysics and Biomolecular Structure 34 2005 * Modelling the Heart--from Genes to Cells to the Whole Organ FManea CMartin-Vide VMitrana ; Oda KKimura TMatsuoka YFunahashi AMuramatsu MKitano H Accepting Networks of Evolutionary Word and Picture Processors: ANoble D 2002. 2004 1 Molecular Interaction Map of a Macrophage * Symmetry breaking and the evolution of development RAPalmer Membrane Computing: An Introduction Berlin Springer Verlag 2004. 2002 306 * Where are the hard knapsack problems? DPisinger Comput. Oper. Res 32 9 2003 * Template guided recombination for IES elimination and unscrambling of genes in stichotrichous ciliates DPrescott AEhrenfeucht GRozenberg J. Theoretical Biology 222 2003 * The second wave of synthetic biology: from modules to systems PE MPurnick RWeiss Nature Reviews Molecular Cell Biology 10 2009 * Networks Evolving Bacteria and Computers LBRay Sci. Signal 3 123 2010 * Autonomous programmable biomolecular devices using self-assembled DNA nanostr tures JReif TLabean Commun. ACM 50 2007 * Cis-interactions between Notch and Delta generate mutually exclusive signalling states GSchlosser GPWagner Chicago DSprinzak ALakhanpal LLebon LASantat MEFontes GAAnderson JGarcia-Ojalvo MBElowitz Nature 465 2004. 2010 Chicago University Press Modularity in development and evolution * YSuzuki HTanaka Symbolic chemical system based on abstract rewriting system and its behavior pattern 1997 1 * Evolution: Developmental circuits rewired ESzathmary Nature 411 2001