Computing Science academics support Covid-19 research
Dr Ke Yuan, Lecturer in Machine Learning and Computational Biology in our Information, Data & Analysis Section, is involved in a Wellcome Trust Covid grant to support some African countries fight with COVID-19:
Within our Formal Analysis, Theory and Algorithms Section, Drs Jess Enright, William Petterson, Kitty Meeks, and Blair Archibald are assisting with model development within the Scottish COVID Response Consortium (SCRC), endorsed by the RAMP (Rapid Assistance in Modelling the Pandemic). Enright is currently the modelling lead on one of the six COVID-19 models currently under development, and Petterson, Meeks, and Archibald will be assisting in development of these models as volunteer programmers and data managers. Each modelling team consists of a modelling lead, a consulting epidemiologist, a research software engineer, and a number of volunteer developers sourced from organisations that have agreed to release their employees to assist with RAMP. Enright is also an expert reviewer for the RAMP’s rapid review team providing 24 to 48-hour turnaround review of modelling outputs to support SPI-M.
How do you celebrate securing a grant about multilayer algorithmics? With a multilayer cake, of course!
Congratulations to Kitty Meeks (PI) and Jess Enright (co-I) on the funding of their EPSRC grant proposal entitled “MultilayerALGS (Multilayer Algorithmics to Leverage Graph Structure)”, total value £766k. Kitty writes:
Multi-layer networks – see the cake for an illustration – can be used to represent qualitatively different types of connections between objects, and are very useful for modelling a wide range of real-world systems. For example, when analysing a social network, we might want to consider face-to-face contact and online interaction separately. These two types of links are different in the processes that produce them, and in their applicability to different questions (If we wish to compute the spread of an amusing online video then the online linkages may be more important, but if we are interested in the spread of a cold or flu then we only need consider the in-person links). Also, perhaps most importantly for our purposes, we see different mathematical structure in the network “layers” formed by each type of connection - for example, online friendships are not geographically constrained, and you may have online friends that you have never met in person.
The goal of the project is to develop efficient computational methods to extract information from and to optimise multilayer systems of this kind. Specifically, our aim is to understand and exploit the processes that generate the different layers and use the resulting structural properties to tackle important computational problems that are intractable in general. In addition to general theoretical research, the project will involve a variety of case studies based on real-world multilayer systems arising in online social networks, medical statistics, and ecological and epidemiological disease transmission data. The project will start in early 2020, led by Kitty Meeks and Jess Enright, with co-Investigators Mark Wong (Urban Studies), Duncan Lee (Statistics) and Heng Guo (Informatics, University of Edinburgh); two postdoctoral positions associated with the project will be advertised soon.