Using graph networks to identify microbiome-based therapeutics for neurodegenerative diseases
Award Number
2424027Award Type
StudentshipStatus / Stage
ActiveDates
27 September 2020 -29 September 2024
Duration (calculated)
04 years 00 monthsFunder(s)
EPSRC (UKRI)Funding Amount
£0.00Funder/Grant study page
EPSRCContracted Centre
University College LondonPrincipal Investigator
Eva-Katherine LymberopoulosWHO Catergories
Models of DiseaseUnderstanding Underlying Disease
Disease Type
Dementia (Unspecified)CPEC Review Info
| Reference ID | 768 |
|---|---|
| Researcher | Reside Team |
| Published | 24/07/2023 |
Data
| Award Number | 2424027 |
|---|---|
| Status / Stage | Active |
| Start Date | 20200927 |
| End Date | 20240929 |
| Duration (calculated) | 04 years 00 months |
| Funder/Grant study page | EPSRC |
| Contracted Centre | University College London |
| Funding Amount | £0.00 |
Plain English Summary
Neurodegenerative diseases such as Parkinson’s disease, Motor Neuron Disease, or dementia, are the leading cause of global disability yet there are no interventions that can alter the steady progression of the disease, dampen its severity, or prevent it. The human gut microbiome has come into focus as a potential mechanism and thereby target for treatment interventions. A major advantage over other intervention targets is that the gut microbiome can be altered relatively easily and cost-effectively, meaning that it is ideal for reaching as many patients as possible. However, an obstacle so far has been that the gut microbiome is extremely complex – there are one hundred trillion individual microbes of over one thousand species which interact with each other and the host in complicated and subtle ways. That is one of the reasons why research so far has had limited success in identifying specific targets for therapeutic intervention.
In this project, we want to use AI methods that are able to account for this complexity. We will use graph network models together with Machine Learning techniques to analyse large-scale datasets comprised of tens of thousands of stool samples. We hope this will lead to the development of a network-based model of the typical disease gut microbiome makeup, enabling us to try out therapeutic interventions computationally. This combination of analysis methods has not yet been used to investigate the gut microbiome, much less specifically in neurodegenerative disease. The research will thus produce a new methodology that can be extrapolated to other diseases with gut microbiome involvement, as well as hopefully tangible targets for subsequent drug development.
Aims
In this project, we want to use AI methods that are able to account for this complexity. We will use graph network models together with Machine Learning techniques to analyse large-scale datasets comprised of tens of thousands of stool samples.