Identifying contamination with advanced visualization and analysis practices: metagenomic approaches for eukaryotic...

Area of Science:

• Genomics
• Bioinformatics
• Microbial Ecology

Background:

• High-throughput sequencing enables rapid genome recovery but requires robust curation against contaminants.
• Contamination in genome assemblies can obscure true biological signals and complicate downstream analyses.

Purpose of the Study:

• To re-analyze sequencing data from Hypsibius dujardini to identify and remove bacterial contaminants.
• To develop and demonstrate a holistic approach for visualizing and curating eukaryotic genome assemblies.
• To assess the impact of different sequencing library types on contaminant profiles.

Main Methods:

• Utilized bacterial single-copy genes, k-mer frequencies, and scaffold coverage for contaminant identification.
• Employed RNA-Seq data to support the curation of the target eukaryotic genome.
• Analyzed data from two groups and eleven sequencing libraries, including Moleculo long-read libraries.

Main Results:

• Identified and characterized multiple near-complete bacterial genomes within the raw assembly.
• Successfully curated a 182 Mbp draft genome for Hypsibius dujardini.
• Found that Moleculo long-read libraries contributed most contaminant scaffolds, with variations between preparations.

Conclusions:

• Eukaryotic genome assembly curation can be significantly improved by applying tools developed for metagenomic analysis.
• Effective identification and removal of contaminants are crucial for accurate genome reconstruction.
• The study underscores the challenges and importance of distinguishing microbial genomes in complex datasets.