Basic Science and Pathogenesis.

Area of Science:

• Genomics
• Neuroscience
• Molecular Biology

Background:

• Genomic mosaicism arises from somatic mutations, influencing genetic disorders and cancers.
• Evaluating somatic mosaicism detection requires advanced techniques like donor-specific assembly (DSA).

Purpose of the Study:

• To refine somatic mosaicism detection across diverse sequencing technologies using personalized DSA.
• To establish a comprehensive approach from genome assembly to single-cell variant calling in human tissue.

Main Methods:

• Sequencing of bulk dorsolateral prefrontal cortex (DLPFC) tissue using Oxford Nanopore, NovaSeq, and linked-read technologies.
• Application of Cas9 capture with long-read sequencing (TEnCATS) for transposable elements.
• Isolation and amplification of single DLPFC neurons using MALBAC, followed by Nanopore and NovaSeq sequencing.

Main Results:

• A haplotype-resolved assembly (5.77 Gb) improved phasing rates from 11.6% to 38.0% with long-read technologies.
• DSA reduced false positive somatic calls in bulk tissue by 14.9%–72.4% and achieved high variant recall rates (97.3%–99.4%).
• Identified 468 candidate somatic heterozygous large deletions and 61 somatic transposable elements in single neurons.

Conclusions:

• The study presents a comprehensive ab initio ad finem approach for somatic variant calling.
• This work provides a valuable resource for analyzing somatic mosaicism in real human tissue at both bulk and single-cell levels.