LRScaf

LRScaf improves genome assembly contiguity by scaffolding NGS-derived draft assemblies with long reads from third-generation sequencing to bridge gaps and resolve ambiguities.

Key Features:

• Improved Contiguity: Increases NGA50 of the CHM1 human genome from 127.1 kbp to 9.4 Mbp using a 20-fold coverage PacBio dataset and from 115.3 kbp to 12.9 Mbp using a ~35× coverage Nanopore NA12878 dataset.
• Efficiency: Demonstrates the shortest runtime among compared scaffolders while exhibiting peak RAM usage of 20.3 GB on CHM1 and 62.6 GB on NA12878.
• Versatility Across Organisms: Consistently improves assembly contiguity and achieves the best or moderate scaffold contiguity and accuracy across organisms including E. coli, S. cerevisiae, A. thaliana, O. sativa, S. pennellii, Z. mays, and H. sapiens.
• Mapper: Uses the minimap mapper for rapid processing of long reads.
• Cost-Effectiveness: Boosts assembly quality of large genomes with relatively low computational resource requirements.

Scientific Applications:

• Draft genome scaffolding: Enhances contiguity and accuracy of draft assemblies generated from short-read NGS data by incorporating long reads.
• Large and complex genome assembly: Applicable to improvement of assemblies for large or repeat-rich genomes such as Z. mays and H. sapiens.
• Downstream genomic analyses: Supports studies in genetics, evolutionary biology, and precision medicine by producing more contiguous reference assemblies.

Methodology:

Employs a scalable algorithm that integrates long-read sequencing data into NGS-based draft assemblies to bridge gaps and resolve ambiguities, with processing accelerated by the minimap mapper.