In its initial presentation, the P system formalism describes the topology of the membranes as a set of nested regions. This description is too rough and presents several shortcommings: only the nesting of membranes is taken into account, not their adjacency and there is an artificial distinction between a membrane and its enclosed region. To answer

Since processor performance scalability will now mostly be achieved through thread-level parallelism, there is a strong incentive to parallelize a broad range of applications, including those with complex control flow and data structures. And writing parallel programs is a notoriously difficult task. Beyond processor performance, the architect can help by facilitating the task of the programmer, especially by simplifying the model exposed to the programmer. In this article, among the many issues associated with writing parallel programs, we focus on finding the appropriate parallelism granularity, and efficiently mapping tasks with complex control and data flow to threads. We propose to relieve the user and compiler of both tasks by delegating the parallelization decision to the architecture at run-time, through a combination of hardware and

Abstract. In this paper, we propose a topological metaphor for computations: computing consists in moving through a path in a data space and making some elementary computations along this path. This idea underlies an experimental declarative programming language called mgs. mgs introduces the notion of topological collection: a set of values organized by a neighborhood relationship. The basic computation step in mgs relies on the notion of path: a path C is substituted for a path B in a topological collection A. This step is called a transformation and several features are proposed to control the transformation applications. By changing the topological structure of the collection, the underlying computational model is changed. Thus, mgs enables a unified view on several computational mechanisms. Some of them are initially inspired by biological or chemical processes (Gamma and the CHAM, Lindenmayer systems, Paun systems and cellular automata).

All information processing systems found in living organisms are based on chemical processes. Harnessing the power of chemistry for computing might lead to a new unifying paradigm coping with the rapidly increasing complexity and autonomy of computational systems. Chemical computing refers to computing with real molecules as well as to programming electronic devices using principles taken from chemistry. The paper focuses on the latter, called artificial chemical computing, and discusses several aspects of how the metaphor of chemistry can be employed to build technical information processing systems. In these systems, computation emerges out of an interplay of many decentralized relatively simple components analogized to molecules. Chemical programming encompassed then the definition of molecules, reaction rules, and the topology and dynamics of the reaction space. Due to the selforganizing nature of chemical dynamics, new programming methods are required. Potential approaches for chemical programming are discussed and a road map for developing chemical computing into a unifying and well grounded approach is sketched.

Abstract. Two classes of tissue P systems based on evolution communication rules are introduced, some results are proved, but many more are listed as further research problems. A framework to develop population P systems is defined and a number of variants formulated with a strong biological motivation. 1

In this report, we present a new framework for the definition of various data-structures (including trees and arrays) together with a generic language of filters enabling a rule-based programming style of functions. This framework is implemented in an experimental language called MGS. The underlying notions funding our framework have a topological nature and make possible to extend the case-based definition of functions found in modern functional languages beyond algebraic data-structures. Keywords group-based data fields, group indexed data structure, path pattern, combinatorial matching, array pattern matching, Cayley graphs, rule based array transformation. The authors of this research report can be contacted at: La.M.I., CNRS UMR 8042 GENOPOLE – Université d ’ Évry Val d’Essonne

During a discussion taking place at WMC'01, G. Paun put the question of what could be computed only by moving symbols between membranes. In this paper we provide some elements of the answer, in a setting similar to tissue P systems, where the set of membranes is organized into a finite graph or into a Cayley graph, and using a very simple propagation process characterizing accretive growth. Our main result is to characterize the final configuration as a least fixed point and to establish two series of approximations that converge to it. All the notions introduced (Cayley graph of membranes, accretive rule and iteration) have been implemented in the MGS programming language and the two approximation series can be e#ectively computed in Pressburger arithmetics using the omega calculator in the case of Abelian Cayley graphs.

Machine (CHAM) extends these ideas with a focus on the expression of semantic of non deterministic processes [5]. The CHAM introduces a mechanism to isolate some parts of the chemical solution. This idea has been seriously taken into account in the notion of P systems. P systems [44, 45] are a recent distributed parallel computing model based on the notion of a membrane structure. A membrane structure is a nesting of cells represented, e.g, by a Venn diagram without intersection and with a unique superset: the skin. Objects are placed in the regions defined by the membranes and evolve following various transformations: an object can evolve into another object, can pass trough a membrane or dissolve its enclosing membrane. As for Gamma, the computation is finished when no object can further evolve. By using nested multisets, MGS is able to emulate more or less the notion of P systems. In addition, patterns like the iteration + go beyond what is possible to specify in the l.h.s. of a Gamma rule.

Pattern-matching programming is an example of a rule-based programming style developed in functional languages. This programming style is intensively used in dialects of ML but is restricted to algebraic data-types. This restriction limits the field of application. However, as shown by [9] at RULE’02, case-based function definitions can be extended to more general data structures called topological collections. We show in this paper that this extension retains the benefits of the typed discipline of the functional languages. More precisely, we show that topological collections and the rule-based definition of functions associated with them fit in a polytypic extension of mini-ML where type inference is still possible.

Computational techniques for modelling and simulating biological systems are surveyed and their semantic and representational capabilities evaluated. To illustrate some of these capabilties, their main applications in this area are also described. A broad range of perspectives is included, represented by three important computational approaches: rewriting systems, cellular automata and agent-based modeling. A computational model can be decomposed into various elements and interaction relationships, which have correspondents in the biological. In this survey, computational techniques are evaluated according to their ability to satisfy these different semantic requirements. We conclude by identifying some of the important shortfalls in current approaches and suggesting areas where further work is required.