Simple questions about neurodegenerative disease: Where? When? What?

Alzheimer's disease is largely caused by genetics, rather than lifestyle factors. The probability of an identical twin developing the disease, if their co-twin already has is ~79%. By comparing the DNA of people who develop a disease against healthy people, we can identify sites in their DNA which increase the likelihood of getting the disease. In recent years it has been established that thousands of changes to the human genome can contribute to disease risk. Each DNA variant explains only a tiny part of the disease risk, but thousands of these variants can explain much of an individual's disease risk. Biological processes implicated by these loci can be understood as being causally involved in the disease. Understanding the mechanisms of neurodegenerative diseases has thus become a statistical problem: we just need to find the patterns that link the variants together.

There are many open questions about neurodegenerative diseases. Indeed, for most brain diseases, we do not even have answers to seemingly basic questions: which part of the brain has gone wrong, and at which age did this occur? For neurodegenerative diseases, which do not show symptoms until late life, it would be easy to assume that the disease causing changes occur in late life: several lines of evidence suggest that this may not be entirely the case though, and that changes which occur early in brain development may be involved.

In recent years I used genetics to identify the cell types which cause schizophrenia and to explain why it's age of onset occurs in early adulthood. This was done by showing that the variants which cause the disease preferentially affect genes which act in particular cells. Using a similar approach, I published the first paper showing that Alzheimer's genes have enriched expression in a type of cell called 'microglia' which are thought of as the immune system of the brain. This finding was a surprise to the field as Alzheimer's had traditionally been considered a neuronal disease. Contrary to this expectation, no enrichment of Alzheimer's risk genes has been found in neurons. Recently I extended the approach to Parkinson's disease: for this disease, the results confirmed the dominant theory of disease mechanism (showing an enrichment in dopaminergic neurons) but also implicated a type of cell known as oligodendrocytes, which had never previously been associated with the disease. The first part of this project will involve further investigating these two discrepancies: are neurons not involved in Alzheimer's, and are oligodendrocytes involved in Parkinson's?

Once we have identified the cell types which cause neurodegeneration, we can address the issue of the age at which the disease acts. We know that the behaviour of cells changes across the lifespan, but we don't know how or why this occurs molecularly. First we will develop a 'map' of the changes which occur in the causal cell types, then we'll test whether the disease associated genetic variants are more associated with development or old age.

What we really need to know is: what happens within a cell to cause the disease? Once we understand this, we can develop drugs to reverse this process. We can investigate this using statistics again (although we will again need plenty of data describing biological regulatory processes within the relevant cell types). Genetic variants which cause disease are known to mostly act by disrupting molecular interactions between DNA and proteins/RNA. Sites on DNA where other molecules bind tend to have distinct sequences: we can learn these sequences from publicly available datasets, then predict the effect of genetic variants on molecular interactions. We can then test statistically, whether a particular type of molecular interaction (for instance, the binding of a particular transcription factor) has been disrupted. Once identified, we can begin attempts to reverse disease effects on these binding sites.

Seek commercialisation

The first four years of this project will be focused on identifying what the neurodegenerative diseases are. The next phase will be to investigate how these mechanisms can be manipulated. The results found during the first phase will enable generation of focused efforts aimed at IP protection and commercialisation to begin. In aim 1.vii I will have identified cell types associated with rare disease genes: I will seek funding and patents for commercialising gene therapies to treat these conditions. I plan to have identified master regulators (i.e. transcription factors) involved in neurodegenerative disease during aim 2 then validated their therapeutic usage throughout aims 3--6.

Furthermore, all staff within my lab will be actively encouraged to consider founding start-ups and provided with training. I have prepared a welcome pack for my lab which emphasises my enthusiastic attitude towards on start-ups: it encourages lab members to look into early stage biotech accelerators such as Rebel Bio, Y-Combinator, Entrepreneur First and Age1.

Act as a bridge between genetics and other scientific fields

The rapid advances in genetics of recent years, should disrupt much existing disease research in biology, by putting our knowledge of disease mechanisms onto a more solid foundation. An example of how this can work comes from my work on Alzheimer's: in 2016 I published the first paper showing microglia are the primary cell type associated with the disease's genetics, and in 2019 a substantial fraction of researchers recruited to the UK's Dementia Research Initiative work on this. Today though, too many results from genetics remain trapped within papers specific to that community. By working directly with neuroscientists and writing papers intended to be read by biologists, I hope to accelerate the rate of information flow between these two disciplines. I will also work to facilitate engagement of quantitative researchers and AI specialists with biological problems.

Build on shared interests with representatives of industry and venture capital

My goal is that by after this fellowship I will be able to setup a public-private initiative, similar to Open Targets, but focused on consolidating the advances made through this project, and rolling its strategy out to other complex disorders. The purpose of this would be to share the cost of data generation required to ensure that the right mechanisms are being targeted, such that private investments can be made in genetically supported target mechanisms. If this approach could work for one complex trait, it is likely it would work for most others. For this to become feasible it will be necessary to build relationships with key figures throughout industry who understand the potential of genomics for neuroscience. To enable this, I will organise a symposium once key milestones have been achieved.

Develop VR data 'expedition' to engage the public and patient groups

To build awareness of my research amongst funders, charities and patient groups involved in neurodegenerative disease I will take part in a range of public engagement activities. A virtual reality app will be developed to showcase my work at these events. This will involve showing a 3D UMAP visualisation with clusters, marker gene expression and morphologies overlaid

Facilitate larger studies of this kind in the future by enabling consortia and support organisations

I will become involved with relevant consortia including the Human Cell Atlas (HCA), iPDGC and iGAP. I will support the HCA's open data requirements, uploading all data before publication. I will continue to make all my software available through Github, NF-CORE and TERRA where applicable. I have previously released a number of packages via Github (e.g. EWCE and MAGMA_Celltyping) and these have been used independently in many publications. Tutorials and worked examples will be provided for all published code.