UCL Community-based Initiative Scheme 2021

Data and its analysis can enable cross-disciplinary research between silos, create new disciplines and generate impact on society. This scheme provided opportunties for early career researchers to develop their leadership skills delivering activities which enable researchers to develop their skills using research software and data.

Hacking the climate: projecting and dealing with extreme events over the UK

Chris Brierley (Geography); Clair Barnes (Statistical Sciences)

Rare events such as floods and extreme wind storms have a substantial impact on the lives of those affected by them, and can cause millions of pounds worth of damage. Understanding how climate change will affect the risk of such catastrophic events in the future is critical for planning and mitigation. The insurance/reinsurance industry, with which UCL has strong links, has a particular interest in understanding the likely changes in the probabilities of such events occurring, under the various climate change scenarios simulated in UKCP and other regional climate models. As part of UCL’s Academic Partnership with the Met Office, the departments of Geography and Statistical Science will host a virtual hackathon considering extreme weather events in the Met Office’s UKCP18 climate projections and other climate datasets. UCL students and staff from all departments will be invited to collaborate on a hackathon encompassing a number of projects considering the likely impacts of projected climate change on hazard events such as floods, heatwaves and wind storms over the UK and globally. Projects will take the form of real-world challenges set by experts in natural hazards and industry partners in the insurance/reinsurance and catastrophe modelling industries, although participants are positively encouraged to submit their own project proposals. Dr Mark Harrison (Head of UK Applied Science, Met Office) shares his thoughts on the UCL-MOAP four-day hackathon in June 2020.

Pharmaco-epidemiology Big Data Collaborative

Cini Bhanu (Institute of Epidemiology & Health Care); Annie Jeffery (IEHC and School of Pharmacy)

We will establish a UCL researcher network focused on pharmaco-epidemiology (pharmaco-epi) methods with big data. The use of big data sets, such as large UK primary care databases like CPRD and THIN, is increasing among health researchers, as they provide a unique opportunity to work with large sample sizes (CPRD has over 50 million patients). Prescriptions are often one of the best coded areas in electronic health care records, and therefore they are growing in use amongst researchers, to answer key health research questions. Pharmaco-epi analysis in big datasets is often complex, with challenges unique to this area of data analysis, this includes: challenges with analysing dosing and duration, different preparations of medicines, understanding prescription use and compliance, establishing definitions for “long term” use of prescriptions. There is not a standard approach for preparing prescription data from electronic health records – this alone can often take months of trial and error. We will optimise learning on the topic with specialist talks from experts in the area, arrange journal club themed discussion on new pharmaco-epi research, dissemination of information on wider courses and conferences in the subject area and establish networks with pharmaco-epi researchers beyond UCL, to share & advance our knowledge of updates in the field.

Find out about their latest activities

Curating Equality: applying systems thinking to achieve gender balance in the visual arts

Nina Pearlman (UCL Culture); Julianne Nyhan (UCL Digital Humanities)

The project addresses gender inequality in the visual arts nationally and internationally. The low percentage of work by women artists reported in museum acquisitions, exhibitions and auction sales, nationally and internationally, and the impact of these persistent trends on the careers of living female artists is widely documented. Nonetheless, attempts to map absences, whether identified in the gaps found between celebrations of ‘male genius’ or in the gaps of history, in exhibitions and publications have failed to give rise to viable solutions that change the status quo. To do so requires an interdisciplinary and systems-based approach that can revolutionise the curatorial research, documentation, collecting and philanthropic practices that underpin the art world, to the benefit of all marginalised and under-represented groups. The starting point is to identify existing solutions from across the computational and data sciences used in different disciplines that have the potential to provide solutions for capturing the latent spaces and networks that underpin the works in museum collections. In tandem, we will identify researchers at UCL whose work is relevant to the conversation and hold a virtual knowledge event to develop a collaborative research proposal.

UCL Code Clubs

Albert Henry (Institute of Health Informatics); Tereza Masonou (Institute of Child Health)

The UCL Code Clubs are collaborative initiatives run by early career researchers which to teach and promote good coding practice within research communities. Initiated at the UCL Institute of Health Informatics (IHI) in June 2019 and followed by the UCL Institute of Child Health (ICH) in January 2020, the UCL Code Clubs are now joining forces to further expand our activities and share our resources with the wider research communities. With the funding from this​ call​, we plan to upscale our established workshop formats by inviting professionals from both academia and industry, and introduce coding challenges which will now be shared between both Code Clubs. With experience of successfully running community-driven events to date, ourteam is committed to provide various learning opportunities to the wider UCL researchcommunities, and promote both inter- and intra-departmental collaboration. Our aims are to increase the number of regular events held, to encourage more contributions toour open-access learning resources, as well as to widen our membership across UCL and beyond.

Join their online monthly meetings via MS Teams, details are available on streams, Github and @UCL_CodeClub.

UCL R Showcase 2021

Ellen Webborn (UCL Energy Institute); Scott Orr (Institute of Sustainable Heritage)

The UCL R Showcase 2021 will build on the first year of the recently-formed bseeR user group and establish it as the UCL-wide R user group. The bseeR user group is a friendly community of postgraduate students and staff who use R within their research and studies within the built environment, or are interested in how they might do so. We hold a monthly meeting to exchange experiences, meet fellow R users, ask questions and find new inspiration for our R projects. This half-day virtual event in June 2021 will showcase the work of Master’s, PhD students and early career researchers through short talks on projects using R, and they will have a chance to win a £50 prize. A keynote speaker from industry or government will demonstrate the role of R in their work and inspire the UCL community to further develop their R skills for their own work. The winners of a separate student competition to attend the (virtual) useR!2021 conference will also be announced. This is currently the only R-specific programming group within UCL (to the best of our knowledge). It is our ambition to bring people together from different disciplines and create a stronger support network for those using R within UCL.

The UCL R Showcase was part of the UCL Festival of Code and the book of abstracts is available here. Find out more about the R User group (website, twitter). Some talk materials are in their Github repository.