THE EUROPEAN MOLECULAR COMPUTING CONSORTIUM (EMCC)

AIMS AND VISIONS of the EMCC:

Molecular computing is a novel and exciting development at the interface of Computer Science and Molecular Biology. Computation using DNA or proteins, for example, has the potential for massive parallelism, allowing trillions of operations per second. Such a parallel molecular computer will have huge implications, both for theoreticians and practitioners. Traditional definitions of computation are being re-defined in the light of recent theoretical and experimental developments. Although thriving, the field of molecular computing is still at an early stage in its development, and a huge and concerted effort is required to assess and exploit its real potential.

The European Molecular Computing Consortium (EMCC) has been created in order to co-ordinate, foster and expand research in this exciting new field, especially in Europe. The EMCC is the result of discussions between different research groups in nine different European countries. A key function of the consortium is to foster co-operation between scientific, technological and industrial partners. A particular effort will be made to create genuinely multi-disciplinary co-operation between Computer Science, Molecular Biology, and other relevant scientific areas. The EMCC will organize various research-enhancing activities such as conferences, workshops, schools, and mutual visits that will provide forums for the exchange of results, and for establishing or strengthening existing co-operations in the field of molecular computing. The EMCC will also actively seek to promote the field, both within the scientific community and to the public at large, via scientific publications, seminars, workshops and public lectures. All participating sites will make the utmost effort to develop their theoretical and laboratory resources. It is hoped that all of these combined efforts will allow the field of molecular computing to thrive in Europe.

The EMCC will also strive to establish and maintain fruitful cooperation with researchers in the area of molecular computing from outside Europe, and in particular with the project "Consortium for Biomolecular Computing" in the US and with the Japanese "Molecular Computer Project".

RESEARCH TOPICS/AREAS

DNA-based logic

Micro- & Nano-tech. interfaces to DNA computing

Plasmid computing

Splicing systems

Twin shuffle models of DNA computing

Coding for DNA computing

RNA computing

Evolvable molecular hardware

Molecular expert systems

Gene assembly in ciliates

Dataflow molecular computers

Membrane computing

Models of self-assembly

Forbidding-Enforcing paradigm for DNA computing

Software tools for DNA computing

Error-resistant tools and techniques

Group summaries

Austria

The Vienna Molecular Computing Group develops theoretical models of computation based on biological reactions that can be found in nature, e.g. cutting (by enzymes) and recombination (by ligases). These operations are not only applied to linear and circular structures, but also to more-dimensional objects. Research focuses on different topologies of test tube systems with static and dynamically developing sets of operations in the tubes as well as of the filters between the tubes. We also investigate the theoretical computational power of different variants of P systems (i.e., of computing with membranes).

Finland (Group web page)

At the moment we are interested only in the theoretical foundations of DNA computing. It is possible that this situation will change in the near future when alterations occur in the personnel. General topics of current interest include the following three. (1) Computationally universal models of DNA computing, both automata and grammars. Overall conditions, based on Watson-Crick complementarity, for universality. (2) Models of DNA computing taking properly into account the massive parallelism of DNA strands. This has so far not been accomplished on the theoretical level. (3) Complementarity viewed in the operational sense: the "bad" strings produced by a generative process are replaced by their complementary ones. This approach has turned out to be a powerful tool in many traditional areas of language theory. (4) Models of DNA computing that properly consider various naturally occuring DNA structures, particularly circular and other non-linear structures.

France (Group web page)

Our group is a new one, starting its work on DNA computing from a theoretical approach, mainly on the computational power of DNA. Up to now, we studied two models of DNA computations. The first one is the well known model of test tubes, an extension of Head systems proposed by Gheorgh Paun, Gregorz Rozenberg and Arto Salomaa. The second one is a more recent model devised by Gheorghe Paun: time-varying distributed H-systems. It was shown that a finite set of tubes, respectively a finite set of components, are enough for, starting from a finite set of axioms, generating any recursively enumerable language. Later people were interested in knowing the smallest number of basic sets necessary for such a result in each model.

Germany

Germany has particular strengths in the areas of evolution (molecular evolution, evolutionary biotechnology, evolvable hardware and evolutionary algorithms) and in self-organizing molecular systems. There is also significant expertise in Germany in microreaction technology and in the single molecule fluorescence detection of DNA. From the theoretical side, the German group also has special competence in the areas of functional programming, complexity theory and self-reproducing automata. This expertise has been focussed in a nationally funded pilot study on DNA Computing conducted by the GMD (The National Research Center on Information Technology in Bonn) with partners in Dortmund, Jena and Leiden. Special facilities of the German group include (1) massively parallel evolvable hardware for electronic A simulation of DNA processing (2) microflow reactors for dataflow architectures in DNA Computing (3) in vitro molecular ecologies involving self-replicating DNA and RNA (4) The above mentioned theoretical competencies.

Italy

Our main focus has been the study of theoretical models, especially covering splicing systems and distributed test tube systems. For the future, we plan to concentrate our efforts on (1) Characterising classes of languages generated by different types of splicing systems (2) Study of encoding problems, and development of encoding techniques and of a theory of codes taking into account complementarity between symbols (3) Merging molecular computing and evolutionary computing techniques (4) Laboratory experiments using plasmids and RNA.

Moldova

Theoretical foundations of molecular computing. Finding the computational power of various models of molecular computing. Development of programming languages for biocomputers and their implementation.

Netherlands (Group web page)

Molecular computing is one of the research programs within Leiden Center for Natural Computing (LCNC). This program is a joint program of computer science and molecular biology. The research covers both theoretical models of molecular computing, and laboratory experiments validating the main principles behind these models. The currently investigated models are forbidding-enforcing, molecular landscapes, plasmid computing, and evolutionary DNA computing. The evolutionary models are developed in cooperation with the evolutionary computing group of LCNC.

Poland (Group web page)

We concentrate mainly on laboratory-based work validating the main concepts of our DNA computing models. We cooperate with University of Warsaw, Faculty of Biology, as well as Institute of Biotechnology and Antibiotics in Warsaw, and Warsaw Agricultural University, Department of Plant Genetics. The currently investigated topics are the following: the inference based on DNA strands, as well as circular DNA molecules, molecular expert systems, implementation of logic gates (binary, fuzzy and multiple-valued), data-flow processors, application of chemically modifed DNA strands, magnetic beads, electrophoretic distributors, use of fluorescent microscopy, and DNA chips.

Romania

At the moment the group is informally constituted and consists of 8 (theoretical) computer scientists, one biologist, one biochemist and one engineer (electronics) from Bucharest (University, Technical University, Romanian Academy, National Institute for Microtechnology) and Iassy (University). The main research direction concerns DNA computing "in info": looking for computability models imagined by the way nature "computes" at the genetic and cellular level. Tools from automata theory, language theory, process algebra and distributed systems are widely used. The previous work was done on H systems, sticker systems and Watson-Crick automata, insertion-deletion algorithms, crossing-over, PA-match and other DNA operations, and the possibility of implementing DNA-like computations in silicon. The researchers will continue in these areas, as well as in a new, growing direction: computing with membranes (P systems). The main goal is to find universal (hence programmable) models of computation and to compare/ integrate these new models with "classic" models, such as Turing machines, rewriting systems, lambda and pi-calculus, etc. We have very good contacts with the groups in the Netherlands, Finland, Austria (significant joint research already done), and also with the groups in Italy and Spain.

Spain (Group web page)

The Natural and Neural Computing Group is located at the Technical University of Madrid. Among different research areas that the group develops are: Theory and Applications of Artificial Neural Networks, Human Brain Modelling, Evolutionary Computation and Molecular Computing. The Group has stablished relationship with the National Center of Biotecnology (CNB - Centro Nacional de Biotecnologia)to cooperate with it. This Center is the host place for the Spanish node of the European Molecular Biology Network (EMBnet).

United Kingdom (Group web page)

The Liverpool-Warwick molecular computing group is distributed between Liverpool (Computer Science) and Warwick (Biological Sciences). The work at Liverpool is concerned with the development of scalable and efficient abstract models of DNA computation. Specifically, we are working on a framework for the automatic translation of P-RAM algorithms down to the level of molecular operations, by first translating them into Boolean circuits. The work at Warwick concentrates on the biological realisation of such models. The Warwick team is currently testing the scalability and error-resistance of strand removal techniques that are central to the implementation of the Boolean circuit simulation mentioned above. The work in molecular computing is one aspect of the wider activities of the Liverpool Biocomputation Group (LBG), which is currently engaged in research on evolutionary computing, neural computing, DNA computing, computational ecologies, cell and tissue information processing and bioinformatics.