As part of our virtual Partners & Platforms Seminars series, the UK DRI is delighted to welcome an exciting line-up of talented DEMON Network members (including Dr Nathan Skene, UK DRI Group Leader) to demonstrate how they are using data to advance dementia research.
“DEMON Network & UK DRI: driving forward experimental dementia research with data science and artificial intelligence”
The recently launched Deep Dementia Phenotyping (DEMON) Network incorporates over 400 scientists, clinicians and industry partners. The Network’s vision is to revolutionise dementia research and healthcare by connecting innovators and harnessing the power of data science and artificial intelligence. Funded by Alzheimer’s Research UK and the Alan Turing Institute, the DEMON Network is free to join and aims to facilitate novel analytic approaches for interdisciplinary collaborative research. The Network is led by Professor David Llewellyn at the University of Exeter, with 5 National Theme Leads and 27 Regional Leads across the UK. Benefits to members include opportunities to contribute to collaborative papers and funding applications, and access to a Clinical Advisory Panel and a Public Patient Involvement Group. It provides a new national infrastructure with training, networking, seminars and workshops, and coordinated engagement with industry for real-world impact. Members have wide-ranging interests including genetics, brain imaging, diagnostic technologies, experimental medicine, analytic methods development and the optimisation of clinical trials, making it a perfect partner for the UK DRI.
Friday 17 July 2020, 12:00 - 13:30 BST
This event is open to UK DRI & DEMON Network. Registration links for the event will be distributed internally by both networks.
Speakers include:
- What is the UK DRI and why should we work together?
Dr Giovanna Lalli UK DRI Director of Scientific Affairs
- Introduction to the DEMON Network
Professor David Llewellyn Turing Fellow (the Alan Turing Institute); Professor of Clinical Epidemiology and Digital Health (University of Exeter)
- MRI markers of cerebrovascular cognitive impairment
Dr Michele Veldsman Postdoctoral Research Scientist in Cognitive Neurology (University of Oxford)
Cerebrovascular risk factors increase the likelihood of dementia. High cerebrovascular burden leads to vascular dementia, accounting for around 20% of dementia cases. Less well appreciated, is that up to 70% of patients with Alzheimer’s disease (AD) also have cerebrovascular disease pathology at post-mortem. The impact of mixed pathologies is likely greatly underestimated. Studies of neurodegenerative dementias rarely control cerebrovascular burden, beyond age and obvious magnetic resonance imaging (MRI) markers like white matter hyperintensities (WMHs). In this talk, I will show work investigating MRI markers of cerebrovascular burden in healthy ageing in 22 000 people from the UK Biobank. I will demonstrate new methods for the estimation of the spatial distribution of cerebrovascular risk-related WMHs and their impact on cognition. I will also present work looking at the importance of microstructural integrity of normal appearing white matter and integrity of grey matter in distributed brain networks for the preservation of cognitive function in healthy ageing and after ischaemic stroke. Together, I will build up a picture of the important MRI markers of cerebrovascular burden that may act as transdiagnostic markers of cognitive impairment.
- Genetic identification of cell types underlying brain disorders and cognitive traits
Dr Nathan Skene UK DRI Group Leader; Lecturer in Dementia Research (Imperial College London)
Using a cellular taxonomy of the brain from single-cell RNA sequencing we have evaluated whether schizophrenia associated genomic loci are linked to particular brain cell types. We found that the common-variant genomic results consistently mapped to pyramidal cells, medium spiny neurons (MSNs) and cortical interneurons. These enrichments were due to sets of genes that were specifically expressed in each of these cell types, suggesting that each cell type plays a biologically distinct roles in schizophrenia. This approach has now been used to map the cell types associated with Parkinson's disease, educational attainment, insomnia and neuroticism. A perspective will be offered on how this approach will enable neural circuits underlying diseases to be mapped out.
- Machine learning for biomarker discovery, pathway modelling and potential causality determination in Alzheimer ’s disease.
Professor Graham Ball Professor of Bioinformatics (Nottingham Trent University)
Dementia in all of its form is a complex set of diseases with a myriad of forms. We have limited understanding of the molecular basis and causality of these diseases, previous hypotheses providing limited success in terms of identifying potential druggable targets. Often treatment strategies are devised based on targeting a specific molecules or biological processes but without wider understanding of the broader implications in a pathway setting. Methodologies such as mass spectrometry-based proteomics, RNASeq and gene expression arrays offer the potential for characterisation of disease derived samples using a huge number of proteins or genes and providing wide coverage over the human molecular system. These data sources are however challenging in their complexity.
The human mind is very good at finding patterns in a system but is not able to conduct the task repetitively for large numbers of parameters. Conversely computers are very good at searching for features in such a data space, but previously defined statistical methods are not able to cope with the high complexity. Here we present the application of Artificial Neural Networks (ANNs, a form of artificial intelligence having the characteristics of both human pattern recognition and computer automated searching) to finding molecular features in Alzheimer’s disease. A suite of methods is presented some with an Alzheimer’s case study application, some in other domains but with the potential for applicability in dementia. Thus, using machine learning techniques, it is possible to interrogate the biology of disease in a non-reductionist approach using current pathway models as a framework for discovery of new biology, adherence to pathways and the perturbations that occur in a disease state. We also present a range of statistical and artificial intelligence-based machine learning techniques to interrogate pathways and find new potentially druggable features.
