Machine learning suggests polygenic risk for cognitive dysfunction in amyotrophic lateral sclerosis.

Area of Science:

• Neuroscience
• Genetics
• Computational Biology

Background:

• Amyotrophic lateral sclerosis (ALS) is a progressive neurodegenerative disease primarily causing muscle weakness.
• Cognitive dysfunction is frequently observed in ALS patients, but the underlying risk factors are not well understood.

Purpose of the Study:

• To investigate the association between genetic polymorphisms and cognitive performance in ALS.
• To explore the relationship between genetic risk scores, longitudinal cognitive decline, and neuroimaging findings in ALS.

Main Methods:

• Sparse canonical correlation analysis (sCCA), an unsupervised machine-learning technique, was used to analyze genetic data from a large ALS cohort.
• A polygenic risk score was derived and correlated with cognitive performance, longitudinal cognitive decline, and neuroimaging data (cortical thinning and neuronal loss).

Main Results:

• Single nucleotide polymorphisms were collectively associated with baseline cognitive performance in ALS patients.
• The polygenic risk score derived using sCCA predicted longitudinal cognitive decline.
• This genetic risk score also correlated with in vivo cortical thinning in specific brain regions and post-mortem neuronal loss in ALS cohorts.

Conclusions:

• Common genetic polymorphisms contribute polygenically to the risk of cortical disease vulnerability in ALS.
• These genetic factors are associated with cognitive dysfunction and neurodegenerative changes in the brain, offering insights into ALS pathogenesis.