UMGAP: the Unipept MetaGenomics Analysis Pipeline

Area of Science:

• Microbiology
• Bioinformatics
• Computational Biology

Background:

• Shotgun metagenomics generates vast datasets on environmental microbial communities.
• Efficient and accurate biodiversity analysis tools are crucial for extracting insights from this data.

Purpose of the Study:

• To develop a novel taxonomic profiling method for shotgun metagenomics data.
• To address limitations of DNA-based approaches, particularly for uncultivated organisms.

Main Methods:

• Developed the Unipept MetaGenomics Analysis Pipeline (UMGAP), a suite of open-source tools in Rust.
• Implemented a taxonomic profiling strategy based on translated protein-coding regions.
• Benchmarked UMGAP against state-of-the-art shotgun metagenomics tools using six preconfigured pipelines.

Main Results:

• UMGAP demonstrates competitive performance compared to existing state-of-the-art tools.
• The protein-based approach achieves low runtime, manageable memory usage, and high accuracy.
• UMGAP successfully profiles archaea, bacteria, eukaryotes, and viruses.

Conclusions:

• UMGAP provides an effective alternative for taxonomic profiling in shotgun metagenomics.
• The tool's design balances performance, accuracy, and usability.
• Interactive visualizations enhance the exploration of complex community data.