LINKS

LINKS scaffolds genome assemblies using long-interval nucleotide k-mer links from long reads to improve assembly contiguity and resolve repetitive regions.

Key Features:

• Long-Read Utilization: Uses long reads from Oxford Nanopore Technologies and PacBio, as well as other draft genomes, as linking evidence to scaffold assemblies without requiring read alignment or base correction.
• Error-Tolerant Scaffolding: Leverages sequence properties of nanopore and other error-prone data to perform scaffolding directly from raw reads without computational base correction.
• Scalability: Demonstrated applicability to large genomes, including re-scaffolding a 20 Gbp Picea glauca assembly, indicating utility across a wide range of genome sizes.
• Improved Contiguity and Correctness: Increased contiguity of an ABySS Escherichia coli K-12 assembly more than five-fold using Oxford Nanopore long reads and produced assemblies for Saccharomyces cerevisiae W303 with contiguity and correctness comparable to or exceeding competing applications.

Scientific Applications:

• Genome scaffolding: Improving scaffold length and resolution of repeats in draft assemblies using long-read linkage information.
• Re-scaffolding large and complex genomes: Applying long-interval k-mer linking to increase contiguity in very large genomes such as Picea glauca.
• Enhancing short-read assemblies: Augmenting assemblies produced by short-read assemblers (e.g., ABySS) with long-read-derived long-range connectivity.
• Analysis of error-containing long-read data: Using raw Oxford Nanopore and other error-prone long reads as direct input for scaffolding without prior base correction.

Methodology:

Constructs scaffolds by linking nucleotide k-mers at long intervals using long reads (Oxford Nanopore, PacBio, or draft genomes), leverages long-range information from nanopore reads, and operates without read alignment or base correction by exploiting sequence properties of error-prone data.