Genotyping TOMM40'523 Poly-T Polymorphisms Using Whole-Genome Sequencing

Area of Science:

• Genetics and Genomics
• Computational Biology
• Neuroscience

Background:

• The TOMM40'523 poly-T repeat polymorphism (rs10524523) is linked to Alzheimer's disease (AD) and cognitive decline.
• Accurate genotyping of this polymorphism is essential for neurodegeneration research.
• Current methods for genotyping TOMM40'523 from whole-genome sequencing (WGS) data are often insufficient, requiring additional PCR and targeted sequencing.

Purpose of the Study:

• To develop and validate a novel computational pipeline for accurate genotyping of the TOMM40'523 poly-T repeat polymorphism directly from WGS data.
• To assess the performance of this pipeline against traditional PCR-based genotyping methods.

Main Methods:

• Developed an ensemble machine learning model using XGBoost, integrating predictions from multiple short tandem repeat (STR) detection tools.
• Utilized STR tool predictions, k-mer counts, and other features for enhanced poly-T repeat length estimation.
• Benchmarked the computational pipeline against PCR-based genotyping in 1,202 participants from four cohort studies.

Main Results:

• The ensemble computational model significantly outperformed individual STR tools in repeat length estimation.
• Achieved high accuracy in repeat length estimation (R² = 0.92) and genotype accuracy (93.2%) compared to PCR-derived gold standards.
• Validated WGS-derived genotypes by replicating known associations between TOMM40'523 variants and cognitive decline.

Conclusions:

• Computational genotyping of TOMM40'523 from WGS data is a scalable and reliable alternative to PCR-based assays.
• This WGS-based approach facilitates broader investigations of TOMM40 variation in large-scale genetic studies.
• The developed pipeline enhances the utility of existing WGS data for Alzheimer's disease and cognitive decline research.