Introduction
Our vision is to shape ‘intelligent dementia research' in collaboration with key stakeholders whilst building a more diverse community of researchers. Dementia is a public health priority that costs the UK economy more than heart disease and cancer combined. Progress has been slow as underfunded scientists have worked in silos whilst generating small-scale data.
However, there have been several important changes as public awareness and research funding have increased, and the quality of data available has been enhanced. We, therefore, have an opportunity to fundamentally change the way in which dementia research is being conducted.
Explaining the science
The Alan Turing Institute was founded on the premise that data science and artificial intelligence will change the world. We agree. But will it change the world for the better, and will it help us to address dementia, one of the biggest challenges to our society? We believe that we are ideally positioned to act now to ensure that we directly tackle this global public health emergency.
The increasingly widespread availability of dementia-related data in combination with the development of powerful computers and machine learning techniques means that we are now in an ideal position to accelerate the pace of discovery. By training computers to identify hidden patterns in high-dimensional data it may be possible to exceed the performance achieved by human intelligence. From understanding disease mechanisms to precision drug discovery, from enhanced diagnostics to targeted prevention, the potential impact is enormous.
Key challenges include the translational gap which prevents experimental medicine from delivering disease-modifying treatments, the reproducibility crisis which hampers progress, the need for better analytic techniques for high dimensional data, and ultimately the need for greater diversity.
Activities
Aims
We will play a crucial role in shaping the research that is prioritised in the field of data science and AI applied to dementia research and the promotion of brain health. We will enhance the sharing of knowledge and in particular the skills and abilities of early career researchers who represent the majority of the workforce actively combating dementia.
We will champion the development of technologies, tools and practical models that have the potential to transform healthcare in the near future.
We will proactively develop practical guidelines for the effective conduct and reporting of original research studies that have the potential to be transformative. We will also provide an open forum for innovation without boundaries that champions the best ideas regardless of traditional labels or boundaries.
In December 2023, the Demon network symposium in collaboration with the Turing Precision Dementia Medicine Interest Group held a research symposium on ‘Mendelian randomisation for causal inference in precision dementia medicine’ at the Alan Turing Institute. The symposium included three sessions on Mendelian Randomisation with molecular exposures, Mendelian Randomisation methods, and Mendelian Randomisation applications, with presentations from 9 speakers.
Programme
Session 1 – MR with molecular exposures (talks)
- Exploring mechanisms in a Mendelian randomization framework - Stephen Burgess
- Using plasma proteomics and Mendelian randomization to identify
- Causal proteins for Alzheimer's disease - Lazaros Belbasis
- Genetic Apo-E and dementia: Dual pathways through lipids and cytokines - Roy Maclean
- Drug-targeted Mendelian randomization (MR) is a popular approach for exploring the effects of pharmacological targets - Benjamin Woolf
Session 2 – MR methods (talks)
- Using genetic and pharmacogenetic instruments for precision medicine in Mendelian randomization - Jack Bowden
- Using clustering of genetic variants in Mendelian randomization to investigate the causal mechanisms underlying multimorbidity - Xiaoran Liang
- Negative control analyses for non-linear Mendelian randomization - Fergus Hamilton
- Do my sensitive analyses agree? A statistical test to judge the similarity of several Mendelian Randomisation estimates - Matthew Tyler
- Interactions in MR framework: how the environment modulates causal effects - Leona Knüsel
Session 3 – MR applications (talks)
- Methodological challenges when using Mendelian Randomization in dementia research - Kate Tilling
- High body mass index, intermediate metabolic risk factors, and risk of vascular-related dementia: observational and Mendelian randomization studies - Liv Tybjærg Nordestgaard
- The causal relationship between gut microbiome composition and Alzheimer's disease: A two sample Mendelian Randomization analysis - Genevieve Monaghan
Talking points
- Should we abandon the clinical concept of mild cognitive impairment in favour of a data-driven alternative?
- Will insights from AI lead us to reconsider the concept of dementia itself?
- Are conventional analytic methods fit for purpose?
- Given the lack of diversity in dementia studies will AI make things worse?
- How do we unlock data more effectively and in particular tackle the problem of data hording?
- Do we have the skills we need and what knowledge transfer activities should be prioritised?
- How do we overcome information overload and gain insights from burgeoning scientific literature?
How to get involved?
Organisers
Dr Janice Ranson
University of ExeterContact info
Researchers
Professor Richard Everson, University of Exeter
Dr Ahmad Al-Khliefat, King’s College London
Dr Nathan Skene, Imperial College London
Dr Conceição Bettencourt, UCL
Dr Sarah Marzi, Imperial College London
Dr Zhi Yao, LifeArc
Dr Tim Rittman, University of Cambridge
Dr Michele Veldsman, University of Oxford
Dr Laura Winchester, University of Oxford
Dr Petral Proitsi, King’s College London
Dr Danielle Newby, University of Oxford
Dr Isabelle Foote, Queen Mary University of London
Dr Donald Lyall, University of Glasgow
Dr Andrey Kormilitzin, University of Oxford
Dr Peter Bagshaw, NHS
Dr Neil Oxtoby, UCL
Dr Eugene Tang, Newcastle University
Dr Magda Bucholc, Ulster University
Professor Zuzana Walker, UCL