Project ID NS-MH2026_09

ThemeNS-MH

Co Supervisor 1A Dr Dr Ahmad Al Khleifat Institute of Psychiatry, Psychology & Neuroscience, School of Neuroscience, Department of Basic & Clinical NeuroscienceEmail

Co Supervisor 1B Dr David Watson Faculty of Natural, Mathematical & Engineering Sciences, Department of InformaticsEmail

Third Supervisor Dr Sarah Marzi

Harnessing DNA Methylation and Machine Learning to Personalise Clinical Trials

(a) Scientific Basis and Translational Aspects
ALS is a clinically and biologically heterogeneous disease, complicating trial design and treatment targeting. While genetics has been partially integrated into trials, epigenetic profiling offers a dynamic layer to capture both disease progression and treatment response.

This project leverages a unique resource: for the first time in ALS, blood-derived DNA methylation has been collected at three time points (baseline, midpoint, end of treatment) in ALS clinical trial. This enables:

Tracking disease progression via longitudinal methylation shifts
Identifying treatment response biomarkers
Integrating epigenetic, clinical, and genetic data to improve patient stratification
The project addresses a major unmet need in neurodegeneration: using real-time biological markers to optimise trial design, interpretation, and personalisation.

(b) Techniques and Skills the Student Will Develop
The student will gain skills in:
Epigenetics and data analysis, statistical modelling, machine learning, biomarker discovery (predictive modelling, multi-omic integration), clinical trial design (virtual arms, stratification), and reproducible research (Git, documentation, HPC/cloud).

(c) Overarching Aim
To evaluate epigenetic profiles as dynamic biomarkers of progression and treatment response in ALS, and develop a framework for their integration into clinical trial design.

(d) Specific, Measurable Objectives

Year 1
Literature review and training in epigenomic pipelines
QC and preprocessing of three trial methylation timepoints
Initial exploration of methylation shifts and clinical associations
Present preliminary findings; submit conference abstract

Year 2
Identify differentially methylated regions linked to progression/response
Cluster patients by longitudinal methylation patterns
Integrate with genomic and clinical features
Develop and cross-validate early predictive models
Submit first-author manuscript

Year 3
Validate models in external datasets (e.g. MND Biobank)
Run trial simulations using biomarker-driven stratification
Collaborate with trial teams on implementation impact
Present at an international meeting

Year 4
Refine biomarker framework; engage regulatory/industry partners
Submit thesis and publish further findings
Contribute to grants, mentor others, and document pipelines

(e) 3-Month Rotation Project

Title: Epigenetic Age Estimation in ALS: Methylation Clocks and Data Quality Control

Objective: To process baseline DNA methylation data from ALS trial and apply established methylation clocks to estimate biological age.

Outline:
Weeks 1–2: Intro to methylation biology and hands-on training in QC tools
Weeks 3–5: QC and normalisation of baseline EPIC array data
Weeks 6–8: Apply clocks to estimate biological age
Weeks 9–12: Assess age acceleration vs clinical variables; summarise in internal report

Expected Outcome:
A clean baseline methylation dataset
Epigenetic age estimates for ALS patients
A short written report and figures to support future longitudinal work

Representative Publications

Grant, O. A., Iacoangeli, A., Zwamborn, R. A. J., van Rheenen, W., Byrne, R., Van Eijk, K. R., Kenna, K., van Vugt, J. J. F. A., Cooper-Knock, J., Kenna, B., Vural, A., Topp, S., Campos, Y., Weber, M., Smith, B., Dobson, R., van Es, M. A., Vourc’h, P., Corcia, P., … Al Khleifat, A. (2024). Sex-specific DNA methylation differences in Amyotrophic lateral sclerosis. https://doi.org/10.1101/2024.11.22.624866 Doherty, T., Yao, Z., Al Khleifat, A., Tantiangco, H. M., Tamburin, S., Albertyn, C., Llewellyn, D. J., Oxtoby, N. P., Ranson, J. M., & Duce, J. (2023). Artificial intelligence for dementia drug discovery and trials optimization. Alzheimer’s Dementia: The Journal of the Alzheimer’s Association, 19(12), 5922-5933. https://doi.org/10.1002/alz.13428 Bucholc, M., James, C., Khleifat, A. A., Badhwar, A., Clarke, N., Dehsarvi, A., Madan, C. R., Marzi, S. J., Shand, C., Schilder, B. M., Tamburin, S., Tantiangco, H. M., Lourida, I., Llewellyn, D. J., & Ranson, J. M. (Accepted/In press). Artificial Intelligence for Dementia Research Methods Optimization. Alzheimer’s Dementia: The Journal of the Alzheimer’s Association.

Padh, K., Zeitler, J., Watson, D., Kusner, M., Silva, R., & Kilbertus, N. (2023). Stochastic Causal Programming for Bounding Treatment Effects. In Proceedings of The 2nd Conference on Causal Learning and Reasoning (Vol. 213, pp. 142-176). (Proceedings of Machine Learning Research). https://proceedings.mlr.press/v213/padh23a.html Watson, D. S., Tax, N., O’Hara, J., Mudd, R., & Guy, I. (2023). Explaining Predictive Uncertainty with Information Theoretic Shapley Values. In Advances in Neural Information Processing Systems (Vol. 36). (ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS). https://proceedings.neurips.cc/paper_files/paper/2023/hash/16e4be78e61a3897665fa01504e9f452-Abstract-Conference.html Watson, D. S., Krutzinna, J., Bruce, I. N., Griffiths, C. E. M., McInnes, I. B., Barnes, M. R., & Floridi, L. (2019). Clinical applications of machine learning algorithms: Beyond the black box. BMJ (Online), 364, Article ll886. https://doi.org/10.1136/bmj.l886