Identification of contributing genes of Huntington's disease by machine learning

Area of Science:

• Neuroscience
• Genetics
• Computational Biology

Background:

• Huntington's disease (HD) is an inherited neurodegenerative disorder caused by polyglutamine mutations in the HTT gene, leading to motor, cognitive, and psychiatric impairments.
• The precise mechanisms underlying HD pathogenesis remain incompletely understood, despite extensive biological data.
• Current machine learning approaches for HD lack sufficient data density for comprehensive mechanistic insights.

Purpose of the Study:

• To apply machine learning to identify genes involved in Huntington's disease pathogenesis.
• To analyze gene expression profiles from postmortem prefrontal cortex samples of HD patients and controls.
• To uncover novel molecular mechanisms contributing to HD.

Main Methods:

• Utilized gene profiling ranking to reduce dimensionality of expression data from 157 HD and 157 control samples.
• Employed machine learning algorithms including decision tree, rule induction, random forest, and generalized linear model.
• Focused on postmortem prefrontal cortex tissue for gene expression analysis.

Main Results:

• Identified 66 potential genes implicated in Huntington's disease pathogenesis.
• Achieved high cross-validated accuracies for the machine learning models, ranging from 89.49% to 97.46%.
• Enrichment analysis revealed involvement of identified genes in transcriptional regulation, inflammatory response, neuron projection, cytoskeleton, and cognitive/sensory/perceptual systems.

Conclusions:

• The study identified key genes potentially contributing to HD pathogenesis through machine learning analysis.
• Mutant HTT may disrupt the expression and transport of these identified genes, driving neurodegeneration.
• These findings provide a foundation for further research into HD mechanisms and therapeutic strategies.