MTG-Link: leveraging barcode information from linked-reads to assemble specific loci

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Local assembly is crucial for sequence reconstruction, gap-filling, and structural variant analysis.
• Linked-read sequencing offers long-range information for accurate assembly but lacks dedicated local assembly tools.

Purpose of the Study:

• To develop a novel local assembly tool specifically for linked-read data.
• To address the limitations of existing tools in leveraging linked-read information for precise genomic analysis.

Main Methods:

• Introduction of MTG-Link, a local assembly tool optimized for linked-reads.
• Implementation of a unique read subsampling strategy utilizing barcode information from mapped linked-reads.

Main Results:

• MTG-Link successfully assembles large sequences (up to tens of kilobases).
• The tool significantly improves local assembly accuracy for specific loci compared to short-read methods.
• MTG-Link accurately characterizes large insertion/deletion variants and reconstructs challenging genomic regions.

Conclusions:

• MTG-Link is an effective local assembly solution for diverse linked-read sequencing technologies.
• The tool enhances the contiguity of large genomic loci, as demonstrated in butterfly genomes.
• MTG-Link is publicly available as open-source code and a Bioconda package.