HyLight: Strain aware assembly of low coverage metagenomes.

Area of Science:

• Genomics
• Bioinformatics
• Microbial ecology

Background:

• Microbial communities exhibit significant strain-level variation impacting phenotypes.
• Accurate strain reconstruction is challenged by sequencing errors and read length limitations.
• Existing metagenome assembly methods often compromise accuracy, strain awareness, or cost.

Purpose of the Study:

• Introduce HyLight, a novel metagenome assembly approach.
• Address challenges in strain-level genome reconstruction using complementary sequencing data.
• Improve accuracy, contiguity, and strain awareness in microbial community genome assembly.

Main Methods:

• Implemented a hybrid approach combining third-generation sequencing (TGS) and next-generation sequencing (NGS) data.
• Utilized strain-resolved overlap graphs (OG) for accurate individual strain reconstruction.
• Employed low-coverage TGS data to mitigate costs while maintaining assembly quality.

Main Results:

• HyLight achieved strain-aware and contiguous assemblies with minimal errors.
• Demonstrated an average improvement of 19.05% in preserving strain identity.
• Showcased near-complete strain awareness across diverse microbial datasets with reduced costs.

Conclusions:

• HyLight effectively integrates TGS and NGS data for superior metagenome assembly.
• Offers significant advances in strain awareness, contiguity, and accuracy.
• Provides a cost-effective solution for high-fidelity microbial community genome reconstruction.