Health behavior risk prediction in metabolic syndrome patients: development and validation of an interpretable machine learning model via multisource heterogeneous data integration.

Huntington’s disease (HD) is an autosomal dominant genetic disorder characterized by age-dependent progressive neurodegeneration predominantly in the striatum and in the cortex of the brain [, ]. Although HD is considered a rare disorder, studies show a consistent increase in HD incidence and prevalence over the past decades, posing a growing burden on families, healthcare systems, and society [–]. HD manifests with progressing cognitive impairment, involuntary movements, psychiatric, metabolic and sleeping problems, and eventually death of the patient typically within about 20 years after the symptoms first appear [, ]. Despite decades of research, no curative treatment is currently available. The disease is caused by expanded cytosine–adenine–guanine (CAG) repeats in one of the alleles of the () gene, translated to a mutant HTT (mHTT) protein containing an abnormally long poly-glutamine (polyQ) tract near the N-terminus []. This mHTT protein exhibits toxic gain-of-function properties and has a greater tendency to misfold, fragment, and form aggregates when compared with the wild-type protein [, ]. However, beyond the toxicity of mHTT in HD, numerous pathways, associated with autophagy, intracellular trafficking, energy metabolism, synaptic function, and protein homeostasis, are disrupted, leading to the pathophysiology and progression of the disease [, ]. A better understanding of the molecular changes involved in such pathways in HD could lead to alternate therapeutic interventions that target the affected pathways. Protein phosphorylation, the most prevalent post-translational modification (PTM), is a critical player in cell signaling pathways and is known to be dysregulated in various neurodegenerative disorders. In addition, dysregulation of phosphorylation has previously been shown to play a pivotal role in the pathology of HD [–]. However, there is a lack of in-depth characterization of phosphorylation events in HD, which could offer deeper insights into the altered signaling pathways and help discover potential new therapeutic targets.