Periodic Reporting for period 1 - i-CONN (Interdisciplinary connectivity: Understanding and managing complex systems using connectivity science)
Período documentado: 2019-10-01 hasta 2021-09-30
Connectivity, simply defined as the degree of connectedness of elements of a system, is a concept that has had a profound impact on diverse disciplines. i-CONN brings together many of these diverse disciplines with two aims. First, we seek to identify common theoretical frameworks and methods that can be applied across these diverse disciplines. Secondly, we aim to train a group of researchers who, no matter what their disciplinary background, will emerge as connectivity scientists able to apply the concepts and methods of connectivity to many research problems.
Fundamentally, all disciplines study phenomena, and those phenomena are typically part of a system of such phenomena, be they galaxies or social groups. Connectivity science focuses on the interactions of the elements of a system. Connectivity science has gained further relevance because of its use in studying complex systems, the characteristics of which cannot be easily determined, and which display such features as emergence and non-linear behaviour. Increasingly, society seeks not only to understand, but also to manage such complex systems, and it is in such management that connectivity science can play a pivotal role. For example, by studying the pathways of pollutants in the water-supply chain and across the landscape, we hope to be able to determine the most effective interventions to stop pollutants getting into, or traveling through, watercourses. The involvement of non-academic organisations within i-CONN emphasises the development of our research for societal benefits.
The overall objectives of i-CONN are to:
1) develop the theoretical underpinning of connectivity science for applications in complex systems;
2) develop a unified framework of methods and approaches that can be applied across disciplines;
3) explore applications of connectivity science to understand, adapt to, and manage, complex systems.
Fundamentally, all disciplines study phenomena, and those phenomena are typically part of a system of such phenomena, be they galaxies or social groups. Connectivity science focuses on the interactions of the elements of a system. Connectivity science has gained further relevance because of its use in studying complex systems, the characteristics of which cannot be easily determined, and which display such features as emergence and non-linear behaviour. Increasingly, society seeks not only to understand, but also to manage such complex systems, and it is in such management that connectivity science can play a pivotal role. For example, by studying the pathways of pollutants in the water-supply chain and across the landscape, we hope to be able to determine the most effective interventions to stop pollutants getting into, or traveling through, watercourses. The involvement of non-academic organisations within i-CONN emphasises the development of our research for societal benefits.
The overall objectives of i-CONN are to:
1) develop the theoretical underpinning of connectivity science for applications in complex systems;
2) develop a unified framework of methods and approaches that can be applied across disciplines;
3) explore applications of connectivity science to understand, adapt to, and manage, complex systems.
The ESRs have completed the first year of their contracts within the reporting period. So far, we have reproduced known correlations between structural connectivity (SC) and functional connectivity (FC) – the quantitative comparison of system architecture (SC) with a (network) representation of the dynamics (FC) – in minimal models. We have shown that it is possible to distinguish two classes of functional connectivity – one based on simultaneous activity (co-activity) of nodes within a network, the other based on sequential activity of nodes – and that these two types of SC/FC correlations are of relevance to a range of applications. This work has been published in Royal Society Interface (Voutsa et al., 2021) and involved ten ESRs and fourteen investigators from nine i-CONN organisations, including three of our partner organisations.
We are now in the process of formulating minimal dynamical models in a range of application areas with the aim of understanding the identified SC/FC relationships. This research involves collaborations between ESRs during secondments and includes applications to sediments, the study of corruption and neuroscience.
We have produced a handbook of the existing connectivity methods used across the i-CONN network, and across the range of disciplines involved. For each method presented we provided: an introduction; specific related background; the problem that it addresses; important algorithms/framework addressing the problem; discussion of the advantages and disadvantages of the existing methods, and available implementations. Additionally, and most importantly within the context of i-CONN, we discussed whether the connectivity methods described are applicable to other disciplines.
Having identified the existing connectivity methods used across the various disciplines represented within i-CONN, we are now in a phase of relating the methods to the potential range of suitable applications. This phase involves testing the applicability, compatibility, and enhancement of consistent methods. A successful event, the Datathon, was organised which allowed the ESRs to study each others’ datasets and apply them to new disciplinary areas. Furthermore, this process of exploring existing connectivity methods and testing their application across different disciplines in the Datathon event enables us to determine if the existing methods fall short of the needs of certain application domains.
We have implemented a substantial training programme for the ESRs covering a wide range of research tools, learning about the disciplinary areas represented in i-CONN and important transferable skills such as scientific writing and collaboration skills. The impact of the pandemic meant that most of the planned range of activities had to move online. We established a successful virtual seminar series and so far, nine experts have delivered seminars that were widely advertised and attended by people from outside the network.
We are now in the process of formulating minimal dynamical models in a range of application areas with the aim of understanding the identified SC/FC relationships. This research involves collaborations between ESRs during secondments and includes applications to sediments, the study of corruption and neuroscience.
We have produced a handbook of the existing connectivity methods used across the i-CONN network, and across the range of disciplines involved. For each method presented we provided: an introduction; specific related background; the problem that it addresses; important algorithms/framework addressing the problem; discussion of the advantages and disadvantages of the existing methods, and available implementations. Additionally, and most importantly within the context of i-CONN, we discussed whether the connectivity methods described are applicable to other disciplines.
Having identified the existing connectivity methods used across the various disciplines represented within i-CONN, we are now in a phase of relating the methods to the potential range of suitable applications. This phase involves testing the applicability, compatibility, and enhancement of consistent methods. A successful event, the Datathon, was organised which allowed the ESRs to study each others’ datasets and apply them to new disciplinary areas. Furthermore, this process of exploring existing connectivity methods and testing their application across different disciplines in the Datathon event enables us to determine if the existing methods fall short of the needs of certain application domains.
We have implemented a substantial training programme for the ESRs covering a wide range of research tools, learning about the disciplinary areas represented in i-CONN and important transferable skills such as scientific writing and collaboration skills. The impact of the pandemic meant that most of the planned range of activities had to move online. We established a successful virtual seminar series and so far, nine experts have delivered seminars that were widely advertised and attended by people from outside the network.
i-CONN is the first attempt to draw together the disparate strands of connectivity research that have hitherto been undertaken independently within different disciplines. Our novel perspective has so far allowed us to (i) distinguish two types of functional connectivity, and the relevance of each to different discipline applications; (ii) take a new approach to solving the problem of missing information in real-life networks, in which we study the behaviour of the network as a means of inferring the missing elements of a network; (iii) use the unique transdisciplinary nature of i-CONN to make a formal study of our network and its evolution; and (iv) use the Datathon to foster transdisciplinary working. The ESRs are continuing to work on the projects that emerged at that event.
The close involvement of our non-academic partners, together with the planned dissemination activities will promote the rapid and widespread uptake of the research being undertaken within i-CONN.
The close involvement of our non-academic partners, together with the planned dissemination activities will promote the rapid and widespread uptake of the research being undertaken within i-CONN.