Regional sequence expansion or collapse in heterozygous genome assemblies

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• High heterozygosity in sequencing data poses significant genome assembly challenges.
• Such challenges can negatively affect downstream biological analyses, particularly in non-model organisms.
• Current genome assembly metrics may not fully capture the impact of heterozygosity on assembly quality.

Purpose of the Study:

• To investigate how varying assembly parameters and algorithms affect genome assemblies from heterozygous datasets.
• To assess the impact of assembly differences on protein annotations and gene ontology (GO) enrichment.
• To identify limitations of standard assembly quality measures and propose improvements.

Main Methods:

• Re-assembly of a heterozygous sequencing dataset using diverse parameters and assembly algorithms.
• Analysis of resulting genome assemblies for fragmentation, length, and local sequence variations (collapse/expansion).
• Statistical assessment of protein annotations and GO category enrichment across different assemblies.

Main Results:

• Assembly methodologies significantly altered protein annotation enrichment in specific GO categories.
• While total assembly length remained stable, fragmentation levels varied, indicating local assembly collapse or expansion.
• Statistically significant GO group deviations were observed even in assemblies with high N50 values, stemming from subtle local sequence variations.

Conclusions:

• Genome assembly methodology choice critically impacts functional annotation of genes, especially in heterozygous genomes.
• N50 is insufficient for assessing assembly quality in heterozygous datasets due to potential local sequence misassemblies.
• Development of new quality metrics to detect local assembly collapse and expansion is crucial for reliable genomic analyses.