Dissecting genomic landscape of somatic mutations in human brain

Abstract

1.49 variants per individual with average 6.17% of VAF in the early-stage vs 8.46 variants per individual with average 1.5% of VAF in the late-stage. Each of the early- and late-stage mutations has a distinct mutational signature compared to tumor-originated mutations. Regarding brain somatic mosaicism, I found that mutational signatures were specific to brain regions. Thus, this study provides new insight into somatic mutations arising in early and late developmental stages and their biological implication. Focal cortical dysplasia (FCD) is a cortical developmental malformation presenting intractable epilepsy, which is largely caused by somatic mutations that abnormally activate the mTOR pathway. This condition is the most cause of epileptic surgery in childhood. Even intensive diagnostic effort, many children with FCD type II remain without a molecular diagnosis. Then, in the part 2, I optimized the affected cell sorting system with subsequent bioinformatic analysis of whole genome sequencing. I was able to identify the causative mutations in 4 of 14 patients in genes encoding mTOR pathway components

Most somatic mutations arising during the normal development present at various levels in single or multiple tissues, depending on the developmental stage and the affected organs. However, it remains unclear how the human developmental stages or mutation-carrying organs affect somatic mutations’ features. In the first part, I performed a systemic and comprehensive analysis of low-level somatic mutations using deep whole-exome sequencing (average read depth ~500x) data in a total 498 of matched multiple tissues (e.g. brain, blood, and others) from 190 individuals. Interestingly, I found that the early-stage mutations (shared in multiple organs) showed significantly lower number but higher variant allele frequency (VAF), compared to the late-stage mutations (in a single organ)

2 ultra-low-level somatic mutations and 2 germline structural variations. The cell sorting method and comprehensive bioinformatic algorithm is a great way to find the cause of the disease in patient without molecular diagnosis using conventional sequencing in FCDII.