The role of machine learning in discovering biomarkers and predicting treatment strategies for neurodegenerative diseases: A narrative review

Abdullahi Tunde Aborode a, Ogunware Adedayo Emmanuel b, Isreal Ayobami Onifade c, Emmanuel Olotu d, Oche Joseph Otorkpa e, Qasim Mehmood f, Suliat Iyabode Abdulai g, Abdullahi Jamiu h, Abraham Osinuga i, Christian Inya Oko j, Sodiq Fakorede k, Mustapha Mangdow k, Oloyede Babatunde l, Zainab Olapade m, Awolola Gbonjubola Victoria n, Abosede Salami o, Idowu A. Usman p, Victor Ifechukwude Agboli p, Ridwan Olamilekan Adesola q

a Department of Chemistry, Mississippi State University, Starkville, MS, USA

b Department of Cellular & Integrative Physiology, University of Nebraska Medical Center, Omaha, NE, USA

c Department of Biology, University at Albany, New York, NY, USA

d Department of Biology, Georgetown University, Washington, DC, USA

e Department of Public Health, Faculty of Health Sciences, National Open University of Nigeria, Lokoja, Nigeria

f King Edward Medical University, Lahore, Pakistan

g Department of Biochemistry, Fountain University, Osun, Nigeria

h Department of Biomedical Sciences and Pathobiology, Virginia-Maryland College of Veterinary Medicine, Virginia Polytechnic Institute and State University, Blacksburg, VA, USA

i Department of Chemical and Biomolecular Engineering, University of Nebraska-Lincoln, Lincoln, NE, USA

j Department of Health Research, University of Lancaster, Lancaster, Lancashire, UK

k Departement of Physical Therapy Rehabilitation Science and Athletic Training, University of Kansas Medical Center, Kansas City, KS, USA

l Department of Molecular Biology, Bowling Green State University, Bowling Green, OH, USA

m Department of Biology, Lamar University, Beaumont, TX, USA

n Departement of Industrial Chemistry, University of Ilorin, Ilorin, Nigeria

o Department of Pharmaceutical Sciences, St. John's University, Jamaica, NY, USA

p Department of Mathematics and Statistics, Georgia State University, Atlanta, GA, USA

q Department of Veterinary Medicine, Faculty of Veterinary Medicine, University of Ibadan, Ibadan, Nigeria

Received 21 October 2024, Revised 11 November 2024, Accepted 8 December 2024, Available online 6 January 2025, Version of Record 6 January 2025.

https://doi.org/10.1016/j.neumar.2024.100034

Abstract

Machine learning in the field of computational intelligence continues to increase significantly and shows potential in the identification, management, treatment, and monitoring of complex diseases, such as neurodegenerative diseases, including Alzheimer's and Parkinson's diseases. Although there are currently no conclusive diagnostic and therapeutic approaches for neurodegenerative diseases, scientists have adopted machine learning algorithms alongside neuroimaging technology to examine significant symptoms and mutations in neurodegenerative diseases. Deep learning algorithms, including neural networks, have shown the ability to monitor brain structure and physiology alterations associated with infections, patients’ motor and cognitive symptoms, and their reactions to treatment. This narrative review article provides insight into the critical role of machine learning in neurodegenerative diseases, particularly in biomarker discovery and prediction of therapeutic strategies. Neurodegenerative diseases, including Alzheimer's disease and Parkinson's disease, are characterized by an abnormal accumulation of specific proteins in the brain, and these protein aggregates are believed to be the main cause of neurotoxicity and neuronal dysfunction. The article focuses on how machine learning can identify disease-related biomarkers and potential therapeutic targets by analyzing protein interaction and mutation data, emphasizing the importance of biomarker discovery in early diagnosis and disease progression monitoring. The utilization of machine learning has the potential to expedite drug discovery, pinpoint novel biomarkers, and tailor personalized treatment plans for individuals afflicted with neurodegenerative conditions. This advancement can enhance the precision of disease diagnosis and therapy, leading to improved clinical outcomes for individuals.

机器学习在发现生物标志物和预测神经退行性疾病治疗策略中的作用：叙述性综述

摘要

计算智能领域的机器学习持续大幅增长，并在复杂疾病的识别、管理、治疗和监测方面显示出潜力，例如神经退行性疾病，包括阿尔茨海默病和帕金森病。虽然目前还没有针对神经退行性疾病的确凿诊断和治疗方法，但科学家们已经采用机器学习算法和神经成像技术来检查神经退行性疾病的重要症状和突变。包括神经网络在内的深度学习算法已显示出监测与感染相关的大脑结构和生理变化、患者的运动和认知症状及其对治疗的反应的能力。这篇叙述性综述文章深入探讨了机器学习在神经退行性疾病中的关键作用，尤其是在生物标记物发现和治疗策略预测方面。包括阿尔茨海默病和帕金森病在内的神经退行性疾病的特征是大脑中特定蛋白质的异常积累，这些蛋白质聚集被认为是神经毒性和神经元功能障碍的主要原因。文章重点介绍了机器学习如何通过分析蛋白质相互作用和突变数据来识别与疾病相关的生物标记物和潜在的治疗靶点，强调了生物标记物的发现在早期诊断和疾病进展监测中的重要性。利用机器学习有可能加快药物发现，确定新的生物标记物，并为神经退行性疾病患者量身定制个性化治疗方案。这一进步可以提高疾病诊断和治疗的精确度，从而改善患者的临床疗效。