Assessing structural variation in a personal genome-towards a human reference diploid genome

Area of Science:

• Genomics
• Bioinformatics

Background:

• Characterizing large genomic variants, or structural variants (SVs), is crucial for advancing genomic research and clinical applications.
• SV detection remains challenging due to their diversity, complexity, and size, compounded by experimental and computational demands.

Purpose of the Study:

• To characterize the structural variant content of a personal genome using Parliament, a consensus SV-calling infrastructure.
• To demonstrate the efficacy of Parliament in integrating multiple data types and SV detection methods.

Main Methods:

• Integrated analysis of whole-genome array comparative genomic hybridization, short-read next-generation sequencing, long-read sequencing, long-insert sequencing, and whole-genome architecture data.
• Utilized Parliament, a publicly available infrastructure, for consensus SV calling.

Main Results:

• Parliament identified 31,007 genomic loci inconsistent with the hg19 reference assembly in the HS1011 personal genome.
• 9,777 putative SVs were supported by hybrid local assembly, long-read data, or multi-source heuristics, spanning 59 Mbp (1.8%) of the reference genome.
• 3,801 SV events were uniquely identified using long-read data, highlighting its importance in SV discovery.

Conclusions:

• The analysis revealed the strengths and limitations of various sequencing technologies, particularly the impact of long-read data on SV discovery.
• The HS1011 data and Parliament infrastructure provide a public resource for SV discovery, software calibration, and personal genome analysis.