JTK: targeted diploid genome assembler.

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Accurate diploid assembly, separating homologous chromosomes, is crucial for understanding genetic variations.
• Reference-based phasing methods struggle with segmental duplications and structural variations, leading to unreliable alignments.
• Existing long-read technologies like PacBio HiFi require additional data (ONT or Hi-C) for complete haplotype phasing.

Purpose of the Study:

• To develop a single long-read sequencing technology capable of accurate diploid genome assembly.
• To address the limitations of current methods in assembling complex genomic regions, such as segmental duplications and structural variations.

Main Methods:

• Introduced JTK, a megabase-scale diploid genome assembler.
• Employs random sampling of kilobase-scale 'chunks' from long reads to phase variants and generate haplotypes.
• Captures both single nucleotide variants (SNVs) and structural variations (SVs) simultaneously using these chunks.

Main Results:

• Achieved fully assembled haplotypes with approximately 99.9% accuracy in the histocompatibility complex (MHC) and leukocyte receptor complex (LRC) regions using 60-fold Oxford Nanopore Technology (ONT) reads.
• Demonstrated superior contiguity in the LRC region compared to assemblies derived from high-coverage PacBio HiFi + Hi-C data for a Japanese sample.
• Successfully assembled difficult-to-map regions that were previously intractable with reference-based approaches.

Conclusions:

• JTK provides a robust solution for accurate diploid genome assembly using a single long-read sequencing technology.
• This assembler is particularly effective for challenging genomic regions, complementing existing reference-based phasing methods.
• JTK is poised to advance pan-genomics research, especially for medically significant but hard-to-assemble genomic areas.