RefKA

RefKA: Reference-guided genome assembly using k-mer binning

RefKA performs reference-guided assembly of large genomes by partitioning a closely related reference genome into bins, aligning unique k-mers from each bin to PacBio long-read sequencing data, assembling bins independently, and stitching bin assemblies into a final genome sequence.

Key Features:

• Reference-Based Binning: Divides a closely related reference genome into smaller genomic bins to localize read assignment and reduce computational complexity.
• k-Mer Alignment: Aligns unique k-mers from each bin to PacBio long reads to assign relevant sequences for bin-specific assembly.
• Parallel Assembly: Assembles genomic bins independently and in parallel to improve computational efficiency.
• Bin Stitching: Integrates independently assembled bins into a contiguous genome assembly.

Scientific Applications:

• Large and Repetitive Genome Assembly: Assembles large, complex genomes with extensive repetitive regions, including the wheat Chinese Spring (CS) genome, supporting crop improvement, evolutionary biology, and genetic diversity studies.

Methodology:

A reference genome is partitioned into bins, and unique k-mers from each bin are aligned to PacBio long-read data. Reads assigned to each bin undergo independent assembly, followed by a bin-stitching step to generate the final genome assembly while reducing computational demands compared to overlap-based methods.