Human whole-exome genotype data for Alzheimer's disease

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Whole-exome sequencing (WES) data generation methods vary, complicating joint analysis.
• The Alzheimer's Disease Sequencing Project (ADSP) involves numerous studies with diverse WES data.
• Standardizing analysis across different sequencing centers, capture kits, and studies is challenging.

Purpose of the Study:

• To develop and present a bioinformatics strategy for joint-calling WES data from multiple studies.
• To address the heterogeneity in WES data generation methods.
• To create a high-quality, unified variant dataset for Alzheimer's disease research.

Main Methods:

• Joint-calling of 20,504 WES samples from nine studies.
• Utilizing variant-called format (VCF) files covering the union of capture kits.
• Processing VCF files to mitigate batch effects from different capture kits and sequencing centers.

Main Results:

• Identification of 8.2 million autosomal variants.
• 96.82% of identified variants are high-quality.
• Variants are located in 28,579 Ensembl transcripts, with 41% intronic and 1.8% having high predicted pathogenicity (CADD > 30).

Conclusions:

• The presented bioinformatics strategy effectively generates high-quality genetic data from heterogeneous WES samples.
• This approach enables robust joint analysis of WES data generated across different batches and platforms.
• The improved data integration benefits the broader genomics research community, particularly in complex disease studies like Alzheimer's.