Skip to navigation Skip to main content Skip to footer

Approved Research

Precision medicine for ME/CFS using omics data and machine learning

Principal Investigator: Dr Christopher Armstrong
Approved Research ID: 79568
Approval date: February 24th 2022

Lay summary

Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS)  is a complex and disabling illness with a diverse range of symptoms that fluctuate in severity and duration. The unknown cause of ME/CFS has created diagnosis and treatment challenges for health care providers as there are no underlying physical attributes to separate ME/CFS sufferers and non-ME/CFS individuals. The uncertainty in diagnosis delays the implementation of appropriate treatment plans and places a significant burden on society and to those suffering with the disease.

Personalised medicine technologies are being developed for most complex chronic diseases as research realise treatment decisions can be made by predicting disease development from one's genetic make-up. Metabolism is the basis of life; it consists of networks of small molecules that keep the body functioning. Hence the combination of genetic and dynamic metabolite information can be used to holistically map the health status of an individual. 

Our hypothesis is that the different fatigue presentations in ME/CFS are caused by multiple points of stress in energy pathways occurring in various levels in the biological system. Different symptom severities and fluctuations caused by similar metabolism mechanisms may be further separated based on the individual's genetic material. 

This project aims to develop an analytical tool to rapidly detect the biological pathways that have been altered due to disease state based on the genetic and metabolic profile of ME/CFS individuals. The analytical tool will be developed using the UK Biobank data, and validated with an external dataset that we will generate in our institute. This tool has the potential to offer several benefits including, reduced medical tests and examinations for diagnosis and the development of personalised treatment strategies.

The estimated duration of this project is 3 years. During this time, we will perform exploratory data analyses, iterate, and continually adapt our analytical tool using machine learning.