Computational analysis of SOD1-G93A mouse muscle biomarkers for comprehensive assessment of ALS progression

Pedro Gómez-Gálvez,Victoria Navarro,Ana M. Castro,Carmen Paradas,Luis M. Escudero
DOI: https://doi.org/10.1101/2024.03.11.584407
2024-03-14
Abstract:Amyotrophic lateral sclerosis (ALS) is, to date, a fatal neurodegenerative disease that currently presents significant challenges in both diagnosis and treatment. The rapid progression and critical prognosis of the disease make accurate and prompt diagnosis crucial for enhancing ALS patient outcomes and quality of life. In this regard, the identification of ALS biomarkers is imperative for evaluating potential treatments, being SOD1 murine models widely employed for their validation in preclinical studies. Here, we introduce and apply an upgraded version of NDICIA (Neuromuscular DIseases Computerized Image Analysis), an image analysis tool to test and quantify the presence of neuromuscular disease image biomarkers. We applied this method to examine morphological and network features in histological images of skeletal muscle biopsies acquired from SOD1 and wild type (WT) mice at different stages of the disease progression: first we quantified the level of difference among mutant and WT groups of images, second we filtered out those features being the primary factors in every separation and finally we analysed how those differences and features change as the disease progresses through its different stages. This analysis revealed the way the pathology of mutant muscles evolves, differentiating from WT in muscle fibres arrangement (graph-theory properties) at presymptomatic stages, and presenting incremental atrophic-like morphological properties at later stages. In consequence, our assay pointed out muscle organization features as an important defining factor in ALS, which could be an excellent outcome measure to early detect a beneficial effect of therapeutic interventions in preclinical trials.
Neuroscience
What problem does this paper attempt to address?