Hifiasm

Hifiasm assembles haplotype-resolved de novo genomes from long high-fidelity (HiFi) reads to produce phased assemblies for analysis of allelic variation.

Key Features:

• HiFi read support: Uses long high-fidelity (HiFi) sequence reads to capture haplotype information in assemblies.
• Phased assembly graph: Constructs a phased assembly graph that encodes genomic variations between haplotypes.
• Haplotype contiguity preservation: Preserves contiguity across all haplotypes rather than collapsing heterozygous alleles into a single consensus.
• Graph trio binning algorithm: Implements an advanced graph trio binning algorithm for improved haplotype separation compared to standard trio binning.
• Complex genome handling: Demonstrates ability to assemble large and polyploid genomes, including the ~30-Gb hexaploid California redwood.
• Empirical performance: Produces superior haplotype-resolved assemblies relative to existing tools on tested datasets including three human and five nonhuman genomes.

Scientific Applications:

• Haplotype-resolved genome assembly: Reconstruction of phased haplotypes for studies of allelic variation.
• Human genetics: Assembly and phasing of human genomes for variant analysis.
• Plant and large-genome genomics: Assembly of large or polyploid plant genomes such as California redwood.
• Comparative genomics and diversity studies: Investigation of genomic diversity and sequence variation across individuals and species.

Methodology:

Builds a phased assembly graph from long HiFi reads and applies a graph trio binning algorithm to separate haplotypes while preserving haplotype contiguity.