This article is 'Highly accessed' (total accesses to this article since publication: 1197) relative to age in BioMed Central: Increasing risk behaviour can outweigh the benefits of antiretroviral drug treatment on the HIV incidence among men-having-sex-with-men in Amsterdam Shan Mei, Rick Quax, David VAN de Vijver, Yifan Zhu and P.m.a. Sloot BMC Infectious Diseases, 11:118 (11 May 2011)
As of August 27, the most downloaded/accessed paper in BMC Systems Biology is: D. van Dijk et al.: Identifying potential survival strategies of HIV-1 through virus-host protein interaction networks, BMC Systems Biology, vol. 4, nr 1 pp. 96+17. 2010.
This paper received the ICCS best paper award 2010:
N. Zarrabi et al.: Modeling HIV-1 intracellular replication, Procedia Computer Science vol. 1, nr 1 pp. 555-564. Elsevier B.V., Amsterdam, May 2010.
Work Packages Description
In order to achieve the objectives of the project, we have structured the main thrusts of efforts into six work packages (WPs).
The first is a separate WP for management. The three main WPs correspond to Areas A1, A2 and A3 of the efforts of the S/T methodology (WP2 corresponds to Area A3, WP3 to Area A1, WP4 to Area A2, more on this at the individual WP descriptions below). Finally, two supporting WPs are dedicated to the integration of the results and to the handling of various risks and benefits, in particular the demonstration of the value added element of the project, and of the concept's generality and applicability. Below we briefly summarize the planned WPs, with the WP leaders in parenthesis, as well as the estimated consortium level effort expressed in person months (PMs).
In the following part we detail the goals and the methods relevant to the above defined Work Packages. (In parenthesis we give the effort in terms of Person Months together with the name of the coordinating partner.)
Overall project management, setup and maintenance of project cooperation infrastructure, S&T Coordination, quality management. The activities in this work package will include:
- Coordinate the overall financial, administrative and scientific activities in the project;
- Monitor the progress towards the overall objectives of the project within set time limits
- Monitor the overall adherence to the financial budgets;
- Monitor inter-work package alignment (tuning of activities in different work packages);
- Periodic monitoring of the progress towards the project's objectives and taking corrective measures as and when appropriate
- Prepare progress reports (Financial and Activity reports), ensure timely delivery to EC;
- Act as liaison to the EC on behalf of the consortium in all verbal and written communication;
- Act as liaison between the Project Steering Group and the Work package leaders;
- Regular progress reporting to Project Steering group.
- Act as the consortiums formal contact to external stakeholder organizations, the press and the public at large;
- Processing all technical, financial and administrative information into project periodic and final reports compliant with Commission criteria;
- Initiate, prepare and preside consortium progress or extraordinary meetings (the latter if deemed necessary) and the spreading of information to all partners pertaining to these meetings;
- Manage the joint plan for the use and dissemination of knowledge, including IPR.
- Coordinate and Liaise with the EURESIST project and other cluster projects from the IST call 4. Adequate resources have been budgeted in case clustering activities with other relevant projects are organized, participation to meetings, contribution to common plans and actions as well as participation to events organized by the Commission".
- Develop a Plan for Quality Assurance
Network centric data mining, data acquisition, data driven inference, qualitative and quantitative analysis, the identifying of the structural dynamics for the network, problems of viral resistance transmission. The activities in this work package will include:
- Identification of datasets (Task 2.1)
In order to provide empirically validated models for controlling and forecasting, it is crucial to gain access to relevant data. We will focus on available public, clinical and epidemiological datasets. Public repositories that will be used are the Stanford HIV db (http://hivdb.stanford.edu) and the Los Alamos HIV database (www.hiv.lanl.gov). The Stanford database can be used to relate particular antiretroviral drugs to mutational patterns. The Los Alamos database is a useful source for HIV-SIV sequences from across the World. Local clinical databases from partners 3 (UCSC) and 5 (EMC) will be included. These hospitals have data sources which, contrary to the public databases, include longitudinal data of relevant parameters–e.g. clinical, demographic and treatment history.
The partners that are going to provide HIV data are also actively involved in European studies on the epidemiology of drug resistant HIV. Partner 5 is involved in the coordination of Europe HIV-Resistance which included demographic, clinical and HIV-sequence data from thousands of patients newly diagnosed with HIV across Europe. Partners 1,3 and 5 actively collaborate in the EU-supported project ViroLab (www.virolab.org) in which longitudinal data is available from patients who are in clinical care and for whom a HIV-1 sequence is available. Also, included will be the longitudinal databases from the ARCA databases which collects information from HIV-infected individuals in Italy.
Additional data that will become available during the course of the project will be explored and integrated.
- Tools for data enhancement, and dealing with incomplete/inconsistent data (Task 2.2)
The databases available to DYNANETS will not always be complete. For instance, complete sexual networks cannot always be identified and data will sometimes be missing. DYNANETS will use existing and develop new methods for data enhancement. The existing methods include the recent progress in the theory of complex networks to analyze these empirical datasets and extract meaningful information in the presence of inconsistent or incomplete data (e.g. the advances in the traceroute sampling of the internet, e.g. Dall'Asta et al. Phys Rev 2005).
Additional approaches will be used to analyze and characterize the network datasets that describe the elements of the system (such as demographic data and HIV-sequences). Among them: Bayesian inference; methods borrowed from bioinformatics that take full advantage of the correlation between the topological pattern and the nodes associated features to predict missing features (Vazquez et al. Nat Biotechnol 2003); inference techniques that take advantage of the hierarchical structure of the known pattern of contacts to make predictions of missing links (Clauset et al. Nature 2008); thresholding techniques to filter out the dominant flows of information in a weighted network (Barrat et al. PNAS 2004); classification methods to uncover the role of nodes with respect to the rest of the system (Ramasco et al. Phys Rev 2008).
Finally, dealing with longitudinal datasets, DynaNets will develop new tools in order to analyze the networked system and extract useful and meaningful information. Only recently research attention has moved to dynamical networks, since longitudinal empirical data started to become available. However, no well defined framework to analyze these datasets and uncover their crucial properties is available. Effects of different time windows adopted to create successive instances of networks will need to be explored, analysis of the changes in time of the topological properties will be studied, different methods to deal with data incompleteness or overabundance will be defined.
- Tools for estimation of relevant parameters (Task 2.3)
A series of analysis will be performed on the available datasets in order to uncover relevant and non-trivial correlations between the structure and dynamics of the network on the one hand (e.g. the sexual network) and the nodes features on the other (e.g. demographic data). Although many of these features are not considered in current modeling, the identification and estimation of relevant parameters might be crucial in providing important predictors of the course of an outbreak, thus improving the modeling developed in WP3 and WP4. Comparisons between different datasets for the estimation of the relevant parameters will uncover possible biases related to the sampling techniques, and will be necessary to validate the conclusions.
Phylogenetic analysis will be used to provide proxy networks of sexual contacts, based on the correlations within clusters of genetically related HIV-1 sequences. Methods borrowed from information theory will be used to define an overlap between the empirical network of sexual contacts and the theoretically built one starting from the micro-level analysis. Approaches from weighted network analysis used for the study of transportation networks will be used to further our understanding of the optimal paths and dominant flows of information/correlation/spread among nodes (Colizza et al, 2007).
- Complex networks of HIV epidemiology and changes at the molecular level (Task 2.4)
Drug resistance associated mutations found among patients failing treatment are more complex as compared to mutations found in transmitted resistance. Several processes can explain this discrepancy: patients who know that they are infected reduce their risk behavior, compensatory fixation and/or reversion of transmitted resistance, the role of specific drug pressure as well as the genetic barrier to drug resistance.
We will use a complex network to study transmitted resistance. The complex network will help to understand the relative contribution of various processes involved in transmitted resistance. This result will be important to predict future trends in transmitted resistance. Pharmaceutical industry could benefit from the result for future development of drugs and estimation of potential market shares.
Tools for the dynamics on and of networks; generation of networks conforming to data,, e.g. by Kronecker multiplication and density-based methods. The activities in this work package will include:
WP3 will design and realize the software framework enabling the representation of Dynanets (including the generation of synthesized emblematic examples of such dynanets) together with the associated simulation engine. First, the modeling software must enable the modeling of a variety of network structures (extracted from the data) together with hierarchies of such network components thus providing means for modeling static complex networks.
WP3 will provide means for describing and modeling the dynamics (both of and on) such static complex networks. Eventually, given a full description of a complex network together with its dynamics, WP3 will provide means for conducing simulations of such a dynanet.
The exact representational needs involved in the static structural part of the concrete models considered by the project will emanate from the work of WP2. Although we cannot foresee what the exact implied graph categories might be, we can already expect that they will encompass the classical graph structures of virus interactions, social networks and transportation or geographical networks. For any such particular type of graph (possibly parametrized) WP3 will provide not only means of description (in a portable language) but also generators for providing a wide spectrum of synthesized emblematic examples (e.g. social like networks (scale free, small world) generated with methods for example inspired from techniques described by Clauset/Moore/Newman in "Hierachical structure and the prediction of missing links in networks" (Nature, 2008)).
If we now consider any such characteristic type of graph as an atomic building block, WP3 will also provide means of description of compositions (assembling or blendings) of atomic blocks and naturally compositions of such compositions. Furthermore WP3 will provide means of description of hierarchies of graphs i.e. a modeler should be able to consider a full graph as a node of a higher level of description.
Let us now assume that the description of a complex network is given (obtained with the help of the above described work). On top of such description WP3 will provide additional means for separately describing the dynamics of this network (dynamical addition/deletion of a node/arc/edge, dynamical modification of a modeling attribute of arc/edge e.g. changing the weight associated to an arc/edge) and the dynamics on this network (what are the sequence of action that each node should realize in order to update its state). The definition of the dynamics should also include the description of the scheduling of those node updates i.e. the precise relative ordering of the updating processes: should the update take place in parallel order, in sequential order (and precisely which one among all possible combinations) or in any arbitrary sequential/parallel tree representing the updating order. Indeed, this feature is a key element for defining coupled network dynamics and more importantly for describing different time scales.
Further than the description or the generation of models (both structure and dynamics) WP3 will provide means not only for describing simulations of such models (initial conditions, how many times should the dynamics be applied, what are the observable categories and when should the measure be executed, where should resulting measurements be collected...) but also means for realizing those simulations. The components of this execution framework should comply with the basic requirements of modern software engineering (correctness, stability, portability, scalability).
The task splitting of this work package is mainly due to its articulation with the respective works of WP2 and WP4. But this splitting is also an organizational mean for stressing the clear formal distinction between a complex network and the possible dynamics of and on such a complex network.
-Static structural modeling (Task 3.1)
For each graph category (handled over by WP2) usually parametrized by a set of statistical parameters, generate a family of associated graphs. Provide constructive means for composing/blending/combining two or many such concrete instantiation of graphs. Provide means for the description and the handling of hierarchies of graphs by integrating them at various levels.
-Dynamics (temporal) modeling (Task 3.2)
For both dynamics OF the graphs (handled over by WP4) and dynamics ON the graph (handled over by WP4 interacting with WP2) realize the associated implementation together with the associated execution framework.
Foundations of information processing on and of dynamical complex networks; Individual based models for the dynamics of network nodes and link dynamics as well as information transmission on dynamically changing networks with a focus on infectious diseases. The activities in this work package will include:
Goal: Understand how nature processes information on all scales. Natural processes seem to self-organize into (hierarchical) complex network structures. The goal of this workpackage is to explore the foundations of this phenomenon and to develop methods to apply this knowledge to a prototypical multiscale model of HIV epidemics.
State of the art: From structure to function. Networks have in recent years emerged as an invaluable tool for describing and quantifying complex systems in many branches of science. Recent studies suggest that networks often exhibit hierarchical organization, in which vertices divide into groups that further subdivide into groups of groups, and so forth over multiple scales (A. Clauset et al., Nature, Vol 453, May 2008| doi:10.1038/nature06830). So far the focus of research in this area was on describing the static structure of those networks (see e.g. Newman, SIAM Review, Vol. 45, No . 2, pp. 167–256, and references therein). In this work package we focus on the dynamics of those networks and how that dynamics gives rise to function. We believe that the central ‘driving force' behind this emergence of function from structure comes from the way Information is processed in these networks.
Approach: Metrics of Information. Quantifying information processing in dynamically changing Complex Networks (CN) is far from trivial and to a very large extent a terra incognita. In this curiousity driven research we aim at exploring metrics that are uniquely sensitive to the spatio-temporal information content.
-Theory of Information processing on hierarchical networks (Task 4.1)
Recently it was conjectured that, in general, systems with evolving spatio-temporal structures are open to symbolic encoding. In this respect we can study how microscopic stochastic update rules are amenable to the modeling of macroscopic processes and the explicit simulation of CN dynamics.
A possible approach comes from work on epsilon-machines revealing the group and semi-group symmetries possessed by the spatial patterns and indicating the minimum amount of memory required to reproduce the configuration ensemble, a quantity known as the statistical complexity. The notion of excess entropy, a form of mutual information, can be used as an information theoretic measure of apparent spatial memory required to describe complex networks. However, there is no unique indicator of complexity in the same way as e.g. entropy characterizes disorder, nor are there any successful mean field approaches or renormalization theories known to capture the hierarchy of information processing observed. We will investigate if the evolution of space-time processes in dynamical complex networks can be studied through for instance (epsilon-tau)-entropy models or alternative measures such as Fisher Information, where the global characteristics from Shannon Information are differentiated in space.
A good understanding of these information metrics will facilitate the design of rules and algorithms for the scale-separation of the node dynamics and the network-structure dynamics. For this we will take as in input the network structures generated in WP3 and the node characteristics (like disease parameters) from WP2, then with the help of the above developed theory and methods to create the models that will be integrated and executed in WP5.
-Individual node dynamics and coupling - micro, meso, macro (Task 4.2)
Individual node dynamics affects at the same time the processes and the network. For example, the dynamical activities of individual agents provide a microscopic and empirically informed basis to construct networks whose meso and macro topology emerges from the interplay of diffusion, proximity and concurrency of interactions. At the same time the emergent topology at the meso and macro scale affects the dynamical activities of the agent at the individual micro scale creating a loop that eventually set in stationary state where dynamics and coupling at the various scales are equally relevant.
The understanding of the feedback mechanisms and the construction of synthetic models that are able to consider these mechanisms in an algorithmic formulation will provide a novel approach to the simulation of multiscale processes of technological and social systems.
In this area we intend to pragmatically work on HIV modeling that face us with all the facets of this problem. DYNANETS will develop a modeling framework which, based on the behaviors and properties observed in the real data analyzed in WP2, will define dynamic models able to take into account the intrinsically dynamic aspect of the network of contacts between people: In particular, models will account for: the infection risk changing in time during a partnership; the heterogeneity of the infection risk per partnership; the partnership formation and dissolution; the change in behavior due to risk perception; the change in behaviour due to the intra-host evolution of the infection. Microscopic rules will be defined to reproduce realistic models for HIV transmission dynamics. The influence of concurrent sexual partnerships will be analyzed with respect to the persistence of diseases in the population. Moreover, adaptive response of individuals due to disease spread will be studied to characterize its feedback-type of impact to the disease dynamics itself. The role of fluctuations and the impact of heterogeneities at multiple levels (e.g. topological fluctuations, individual heterogeneities, temporal fluctuations) on the dynamical aspects – both of the network and of the disease dynamics – will be assessed. DYNANETS will study how these models can be used to examine the effectiveness of various control strategies, in particular screening programs and contact tracing.
In addition, the small scale problem reverberates at the very large scale of entire populations because of the long time scale of the disease. We plan to study the problem in different scale settings. At the micro scale by using agent based models entangling the network generation and the process evolution. At the meso scale of coupled subpopulations we will study the issue of describing multi-scale nested networks and introducing those in the mathematical definition of large scale computational models. This will require interfacing the population level description with the microscopic agent models coupled at different scales or via appropriate coarse grained PDEs. Finally we will tackle the large scale spreading of HIV across populations by integrating local population dynanets in global epidemic modeling schemes across different world region. While large scale models have proved to be effective in the case of diseases acting on small time scale (i.e. influenza), HIV works on the scale of years. This makes it extremely important to account for the detailed dynamics of individuals, even at such large time scales The mobility dynamics has to take into account traveling and population movement both at the short time scale (the infecting interactions) and the long time scale (the long term mobility across populations and communities as migration processes responsible for the appearance of new strains in given regions).
-Validation of individual node dynamics (Task 4.3)
To understand the propagation of epidemics at the world scale, it is essential to include long range travels, i.e. airline connections between airports. A large amount of work in this direction has been carried out previously (see for example : The role of the airline transportation network in the prediction and predictability of global epidemics, V. Colizza, A. Barrat, M. Barthélemy, A. Vespignani, Proc. Natl. Acad. Sci. USA 103 (2006) 2015; Predictability and epidemic pathways in global outbreaks of infectious diseases: the SARS case study, V. Colizza, A. Barrat, M. Barthélemy, A. Vespignani, BMC Medicine 5 (2007) 34). These first studies have adopted, for simplicity, mean-field assumptions. For example, they assume that individuals that travel from town A to town B are drawn randomly from town A population, mix homogeneously within town B, and may be selected, again randomly, to travel from town B to any other destination. In short, they have focused on the town-town network, taking each town in a mean field way.
We will study the dynamics of disease propagation in more realistic network models, which take into account the population correlations. In the context of epidemic propagation, one can think of the world population as a hierarchical network composed of densely connected subpopulations (within towns) and more loosely connected individuals (in different towns). Also, it is well-known that the probability to travel by airplane is not homogeneously distributed in the population (for an example on the French case, see http://ethel.ish-lyon.cnrs.fr/Documents/ETHEL-Rapport_R3_v1.pdf). Therefore, considering the propagation of a disease in a hierarchical model of the world population, which takes into account the correlations sketched above, may lead to a better understanding of its dynamics and final states. Our study could also be generalized to other dynamics of propagation (rumors, trust...) in hierarchical networks.
Unified representation, interactive visualization, front-ends, web interface; model validation with statistical and qualitative tools, evaluation of value-added elements, comparison with other models and approaches. The activities in this work package will include:
The platform must provide the complete spectrum of instrumental abilities to conduct studies through the whole life-cycle of an application model: analysis of the data with methods ranging from interactive data analysis to using statistical analysis tools in order to extract some characteristic structural/dynamical traits of the observed phenomenon, authoring of an associated dynanet initial model, execution of low level simulation tests, with the help of visualizations displaying both the static structural dimension and also the dynamics, definition of one or many scenarios, with tools for describing carefully crafted sets of simulations, execution of such scenarios, analysis of the scenarios results (in order to validate/invalidate the current dynanet model) with tools ranging from specialised visualizations to advanced statistical analysis, cycling back to any of the previous stages by feeding back previously acquired knowledge (be it positive in case of validation or negative in case of failure to comply with field data) into the current model and proceeding with its refinement, eventually (once some confidence of the validity of model is gained or established) use further scenarios (with above mentioned tools) for e.g. predictive studies.
Some of the above mentioned tools will be developed for another WP (e.g. dedicated statistical data analysis tools will be required by WP2) whereas the simulation engine is a result of WP3. For the others instruments (e.g. visualisation of the structure of a hierarchical graph, visualisation of some dynamics or scenario authoring tools) specific implementations will be required and will be designed and realised within WP5.
No programming knowledge or computational expertise will be required to use the platform. Datasets and modelling approaches will be associated with corresponding documentation files, which will report detailed information on the algorithms, implementation details, notations and conceptualization adopted, references, hints for the user, instructions on how to use it, applications, suggested actions, and examples. This approach will be extremely helpful for the interaction of a non-expert user with the platform, allowing the use of complex computational tools by people who do not necessarily have this kind of expertise.
-Specialized software components and platform integration (Task 5.1)
The task will, in accordance to the above, be to twofold. First to develop the required additional instruments (each such implementation being organized as a sub-task) and second to smoothly integrate all resulting software tools within a single unified front ending tool.
The decomposition of T5.1 will be as follows: T5.1.1 Model integration (using result of WP3), T5.1.2 Scenarios description and execution tools, T5.1.3 Interactive data analysis and visualization tools (static and dynamic), T5.1.4 Unified representation and front ending.
We must here give further precisions on the visualization effort within task T5.1.3 by stating the four crucial needs for Dynanets (together with a corresponding state of the art):
- The visualization of large (static) network. Many available graph visualization algorithms [Herman, Melançon, Marshall, "Graph visualization and navigation in information visualization: A survey", IEEE Transactions on Visualization and Computer Graphics, 2000.] are capable of visualizing graphs that contain tens to a couple of hundreds of nodes. Very few can handle more. The key challenges in graph visualization are, therefore, directly related to the size of a graph: larger networks pose more constraints on the graph visualization algorithms. The number of elements (nodes and/or edges) in a graph has a direct effect on viewing performance in two ways. First: it takes longer to compute a layout for larger graphs and responsiveness decreases. Second: even if the computational and graphical resources are available to create a responsive visualization, the issue of viewability and usability arises as it may be impossible to discern individual nodes and edges.
- The ability to interact with visualization tools entices exploratory research. A sufficiently flexible and responsive interactive visualization environment enables the researcher to take control of the graphical representation and use his cognitive abilities, experience and expertise to detect structure and patterns that would be difficult to obtain otherwise. Several interactive graph visualization algorithms and toolkits have been developed that are now available to the public in the form of standalone programs and open source toolkits [Fröhlich, Werner, Demonstration of the interactive graph visualization system daVinci, Proceedings of DIMACS Workshop on Graph Drawing, 1994], [Munzner, Interactive Visualization of Large Graphs and Networks, Ph.D. Dissertation, Stanford University, 2000]., [Heer, Card, Landay, prefuse: a toolkit for interactive information visualization, Proceedings of the SIGCHI conference on Human factors in computing systems, 2005].
- Visualization of the dynamics: Many of the existing algorithms for displaying dynamics of the network are designed for specific application areas and lack the abstraction that would be expected from a general framework [Heer, Agrawala, Software Design Patterns for Information Visualization, IEEE Transactions on Visualisation and Computer Graphics, 2006]. The visualization of the dynamics on the network remains alas (and to our knowledge) an unformalized domain where ad-hoc choices dominate.
- Web integration has the advantage that visualization methods can be made omnipresent and globally available to anyone with access to a web browser which in turn opens possibilities for collaborative research, including annotation facilities to highlight structures of interest [Heer, Viégas, Wattenberg, Voyagers and Voyeurs: Supporting Asynchronous Collaborative Information Visualization, in Proceedings of CHI 2007]. However, the computational and graphical resources in such environments are often limited.
Work to be realized: for each of the four above described needs, WP5 will select the most appropriate (to the data) of the available tools or algorithms (and we cannot expect the technique for displaying large graphs to be the same one as the technique we will use for interactive visualization). This effort will not be limited to a mere modular intregration (or customization/integration) process and some of the components will require active research within WP5. WP5 will provide adaptive multiple scales visualization that change in visual appearance depending on the spatial or temporal range that is represented. WP5 will explore the capabilities of optimized programs that exploit high performance interactive graphics devices for the visualization and intuitive interaction with large graphs. Eventually WP5 will incorporate a distributed framework that facilitates web-based interactive visualization using visualization services that are able to exploit distributed high performance computational and graphical resources.
-Platform support and software exploitation (Task T5.2)
One of the central efforts (Task T6.2) within WP6 is dedicated to the applications and assumes the availability of the simulation platform. At month 24, a first operational integrated simulation environment will be delivered enabling its exploitation in WP6. The second task of WP5, T5.2 will then concentrate on application support while consolidating the software quality, usability and re-usability features.
Execution of scenario analyses, exploration of future uses in forecasting and prevention; possible application domains e.g. in food webs, metabolic nets, buying behavior; assessment of the feasibility for applications; dissemination, publication, publicity.
The main aim of this WP is to pro-actively seek and develop further application areas for the novel ICT paradigm (i.e., concepts, models, methodology and tools) developed in the other WPs. This activity is augmented by the dissemination the results of the project, the maintenance of the project web-site and by participation in concertation activities, respectively.
The development of the Dynamically Changing Complex Networks concept, its methodologies and tools will be driven by our motivating master example (the HIV epidemic) in the other project. However, the paradigm proposed in the current proposal is believed to be generally applicable in a variety of areas, therefore the project dedicates an entire WP to elaborate this idea in general, and with demonstrative examples from a few concrete further domains. While this activity is the main thrust of the WP, the temporal structure dictates the following ordering of its tasks.
-Applications, Dissemination and Exploitation (T6.1)
This is the out-reaching part of the WP, consisting of several activities, organized in three subtasks. One activity develops the public face of the project, which is achieved by the development and up-to-date maintenance of the project's web page. Another activity is the active dissemination of the foreseeable results, including the identification of potentially relevant domains and interested parties and the development and execution of a plan for their inclusion in the uptake of the paradigm.
User Club. During the first year, WP6 reaches out and involves interested partners outside of the consortium in a User Club. The User Club is a loosely organized advisory group of interested parties and potentially relevant application developers, convened and organized by the WP6, that interacts via e-mail and teleconferencing facilities and that meets in person during a few workshops focusing on specific sub-domains. The output of the work performed in the User Club, like comments and statements of interests, provides inputs for the valorization activities of Task 6.2.
In the second half of the project, Task 6.1 shifts its activities towards dissemination and exploitation to disseminate, communicate and utilize the results and experiences of the project. This includes the careful implementation of the dissemination and exploitation plan developed earlier, as well as the active seeking out of the possible opportunities and venues to increase the project's impact. Concertation, that is, seeking collaboration and mutual exploitation with professional groups as well as other EU funded (and national level) projects is another important part of Task 6.1's activities.
The "Web page and content management" (T6.1.1) sub-task is dedicated to the setting up of the project web-site and its active maintenance. Since the results of out-reaching activities are important for the successful completion of Task6.2, a special emphasis will be dedicated to up-to-date content management.
The "Application, dissemination and exploitation plan" (T6.1.2) sub-task is responsible for the identification and elaboration of the potential further application areas of the Dynamically Changing Complex Network paradigm, as well as for the detailed dissemination and execution plan. The founding and convening of the User Club is also part of this sub-task.
The Dissemination (T6.1.3) sub-task is responsible for the careful execution of the dissemination and exploitation plan and for the maintenance of the User Club as a forum for increasing the project's impact.
-Applications and Valorization (T6.2)
This task is focused on the establishment and validation of the applicability of the novel ICT paradigm (the Dynamically Changing Complex Network concept, its methods and tools).
On the one hand, this means the execution and validation of the scenarios developed in WP5 for the HIV example, in close collaboration with the aforementioned work package. This activity is twofold: it provides a usability test and feedback for the integration platform developers, as well as to the core concept developers in WP2-4. In addition, this activity also demonstrates the power of the new paradigm in the specific application domain of the HIV example.
On the other hand, Task 6.2 also aims at the deepening and elaborating the general applicability of the novel ICT paradigm developed during the project. This is implemented by the identification of a few further application domains, or proto-Use Cases (partially based on the expected input from the User Club convened and operated by Task 6.1).
Application Areas. Potential application areas, as foreseen at the time of writing and to be explored and elaborated in WP6 include, e.g.,
• second generation P2P applications,
• food web and ecosystem models,
• economic networks, etc.
The identified application domains will be documented and screened, and a subset of them will be singled for preliminary development for demonstration.
The "Scenario execution and validation of scenario platform" (T6.2.1) sub-task is responsible for the execution and validation of the scenarios and tools developed for the HIV example in WP5. The responsibilities also include providing feedback to WP5, with respect to the applicability and usability of the integrated platform.
The "Development of further application domains" (T6.2.2) sub-task is responsible for the implementation of the deepening and elaboration of the general applicability of the novel ICT paradigm. This includes the identification of the proto-Use Cases, their documentation and screening, and the preliminary development of the subset of them singled out for demonstration.