A Machine Learning Approach for Highlighting microRNAs as Biomarkers Linked to Amyotrophic Lateral Sclerosis Diagnosis and Progression

Abstract

Amyotrophic Lateral Sclerosis (ALS) is a fatal neurodegenerative disease characterized by the progressive loss of motor neurons in the brain and spinal cord. The early diagnosis of ALS can be challenging, as it usually depends on clinical examination and the exclusion of other possible causes. In this regard, the analysis of miRNA expression profiles in biofluids makes miRNAs promising non-invasive clinical biomarkers. Due to the increasing amount of scientific literature that often provides controversial results, this work aims to deepen the understanding of the current state of the art on this topic using a machine-learning-based approach. A systematic literature search was conducted to analyze a set of 308 scientific articles using the MySLR digital platform and the Latent Dirichlet Allocation (LDA) algorithm. Two relevant topics were identified, and the articles clustered in each of them were analyzed and discussed in terms of biomolecular mechanisms, as well as in translational and clinical settings. Several miRNAs detected in the tissues and biofluids of ALS patients, including blood and cerebrospinal fluid (CSF), have been linked to ALS diagnosis and progression. Some of them may represent promising non-invasive clinical biomarkers. In this context, future scientific priorities and goals have been proposed.

Keywords

• ALS
• miRNAs
• degenerative diseases
• clinical markers
• prognosis
• text mining