Researchers at The Neuro (Montreal Neurological Institute and Hospital, Canada) and the Ludmer Centre of McGill University (Montreal, Canada) have used an artificial intelligence algorithm to analyze samples from individuals with Alzheimer’s and Huntington’s, in order to identify molecular patterns specific to these diseases. The results, published in Brain, could lead to a more individualized approach to neurodegenerative disease treatment.
Most prevalent neurodegenerative disorders take decades to develop. Owing to a lack of long-term chronological data regarding neurodegenerative disease progression, along with the variation in disease progression and expression between individuals, it is often difficult to evaluate the effectiveness of therapies for neurodegenerative diseases.
This study aimed to uncover long-term chronological information contained in large-scale data by covering decades of disease progression among individuals with such diseases. The researchers used a machine-learning algorithm to detect unique patterns, in order to reveal how changes in gene expression over time are related to changes in the individual’s condition.
Using a novel gene expression contrastive trajectory inference (GE-cTI) method, the team analyzed the blood and post-mortem brain samples of 1969 individuals with Alzheimer’s and Huntington’s disease. Unlike previous studies, which have often used static data, the algorithm was able to detect how their genes expressed themselves in unique ways over decades.
You might also like:
Further, the researchers observed comparable results for the blood test and post-mortem brain test, with 85–90% of the of the top predictive molecular pathways identified in the brain also detected in the blood. This reveals a striking similarity between molecular alterations both in the brain and peripheral body.
“This test could one day be used by doctors to evaluate patients and prescribe therapies tailored to their needs,” commented Yasser Iturria-Medina (McGill University), the first author of the study. “It could also be used in clinical trials to categorize patients and better determine how experimental drugs impact their predicted disease progression.”
The team plan to carry out further testing of these models in other diseases such as Parkinson’s disease and amyotrophic sclerosis.
Sources: Iturria-Medina Y, Khan A, Adewale Q et al. Blood and brain gene expression trajectories mirror neuropathology and clinical deterioration in neurodegeneration. Brain doi:10.1093/brain/awz400 (2020); www.mcgill.ca/neuro/channels/news/ai-analyzed-blood-test-can-predict-progression-neurodegenerative-disease-317675