Modulation at age of onset in tunisian huntington disease patients: implication of new modifier genes

Explainable artificial intelligence based on feature optimization for age at onset prediction of spinocerebellar ataxia type 3

Existing treatments can only delay the progression of spinocerebellar ataxia type 3/Machado-Joseph disease (SCA3/MJD) after onset, so the prediction of the age at onset (AAO) can facilitate early intervention and follow-up to improve treatment efficacy. The objective of this study was to develop an explainable artificial intelligence (XAI) based on feature optimization to provide an interpretable and more accurate AAO prediction. A total of 1,008 affected SCA3/MJD subjects from mainland China were analyzed. The expanded cytosine-adenine-guanine (CAG) trinucleotide repeats of 10 polyQ-related genes were genotyped and included in related models as potential AAO modifiers. The performance of 4 feature optimization methods and 10 machine learning (ML) algorithms were compared, followed by building the XAI based on the SHapley Additive exPlanations (SHAP). The model constructed with an artificial neural network (ANN) and feature optimization of Crossing-Correlation-StepSVM performed best and achieved a coefficient of determination (R2) of 0.653 and mean absolute error (MAE), root mean square error (RMSE), and median absolute error (MedianAE) of 4.544, 6.090, and 3.236 years, respectively. The XAI explained the predicted results, which suggests that the factors affecting the AAO were complex and associated with gene interactions. An XAI based on feature optimization can improve the accuracy of AAO prediction and provide interpretable and personalized prediction.

Investigation of the Influence of TBP CAG/CAA Repeats in Conjunction with HTT CAG Repeats on Huntington's Disease Age at Onset in a Brazilian Sample

Huntington's disease (HD) is a genetic neurodegenerative progressive and fatal disease characterized by motor disorder, cognitive impairment, and behavioral problems, caused by expanded repeats of CAG trinucleotides in the HTT gene. The aim of this study was to investigate the influence of TBP gene CAG/CAA repeats in conjunction with HTT gene CAG repeats, on the age at HD onset in Brazilian individuals. Individuals diagnosed as molecularly negative for HD presented 29-39 TBP CAG/CAA. Their most frequent allele had 36 repeats. In individuals diagnosed as molecularly positive for HD, a range of 25-40 TBP CAG/ CAA was found. The most frequent TBP allele had 38 repeats. We also conducted TBP direct Sanger sequencing of some samples which demonstrated other four TBP structures different from the basic TBP structure and others reported in the literature. The HTT expanded CAG and TBP CAG/CAA repeat sizes jointly explained 66% of the age at onset (AO) in our HD patients. The strongest variable in the model associated with AO was the number of expanded HTT CAG repeats. The difference between the association of HD AO with HTT expanded CAG together with TBP CAG/CAA and the association of HD AO with HTT expanded CAG was 0.001 (∆R²). Therefore, we found a weak association (0.1%) of TBP CAG/CAA repeats on HD AO, if any.

Co-expression Patterns between ATN1 and ATXN2 Coincide with Brain Regions Affected in Huntington's Disease

Cytosine-adenine-guanine (CAG) repeat expansions in the coding regions of nine polyglutamine (polyQ) genes (HTT, ATXN1, ATXN2, ATXN3, CACNA1A, ATXN7, ATN1, AR, and TBP) are the cause of several neurodegenerative diseases including Huntington’s disease (HD), six different spinocerebellar ataxias (SCAs), dentatorubral-pallidoluysian atrophy, and spinobulbar muscular atrophy. The expanded CAG repeat length in the causative gene is negatively related to the age-at-onset (AAO) of clinical symptoms. In addition to the expanded CAG repeat length in the causative gene, the normal CAG repeats in the other polyQ genes can affect the AAO, suggesting functional interactions between the polyQ genes. However, there is no detailed assessment of the relationships among polyQ genes in pathologically relevant brain regions. We used gene co-expression analysis to study the functional relationships among polyQ genes in different brain regions using the Allen Human Brain Atlas (AHBA), a spatial map of gene expression in the healthy brain. We constructed co-expression networks for seven anatomical brain structures, as well as a region showing a specific pattern of atrophy in HD patients detected by magnetic resonance imaging (MRI) of the brain. In this HD-associated region, we found that ATN1 and ATXN2 were co-expressed and shared co-expression partners which were enriched for DNA repair genes. We observed a similar co-expression pattern in the frontal lobe, parietal lobe, and striatum in which this relation was most pronounced. Given that the co-expression patterns for these anatomical structures were similar to those for the HD-associated region, our results suggest that their disruption is likely involved in HD pathology. Moreover, ATN1 and ATXN2 also shared many co-expressed genes with HTT, the causative gene of HD, across the brain. Although this triangular relationship among these three polyQ genes may also be dysregulated in other polyQ diseases, stronger co-expression patterns between ATN1 and ATXN2 observed in the HD-associated region, especially in the striatum, may be more specific to HD.