Benchmarking genome choice in functional genomics analyses

Area of Science:

• Genomics
• Functional Genomics
• Bioinformatics

Background:

• The human genome reference provides a coordinate system but lacks diversity.
• Current references are incomplete and not representative of global human genetic variation.
• Understanding genome representation's impact on functional genomics is crucial.

Purpose of the Study:

• To benchmark genome representation's influence on functional genomics analyses.
• To compare analytical frameworks across different genome references.
• To assess the impact of individual genome assemblies versus standard references.

Main Methods:

• Utilized chromatin accessibility sequencing (ATAC-seq), RNA sequencing, whole-genome bisulfite sequencing, and chromosome conformation capture (Hi-C).
• Analyzed data from lymphoblastoid cell lines of five individuals with phased genome assemblies.
• Compared functional genomics results across hg38, CHM13, a draft human pangenome, and individual assemblies.

Main Results:

• Individual identity explained the largest variance (57.52–78.47%) in functional estimates.
• Genome choice had a minor contribution (0.002–7.85%) to variance.
• Approximately 2% of biological signals were only detectable with personal assemblies, including a novel AluY5a insertion impacting a TNKS enhancer.

Conclusions:

• Individual genetic variation is the primary driver of functional genomic differences, not the genome reference.
• Current linear references miss biologically important features, highlighting the need for pangenome approaches.
• Personalized genome assemblies are essential for capturing the full spectrum of human functional genomics.