Using AnnoTree to Get More Assignments, Faster, in DIAMOND+MEGAN Microbiome Analysis

Area of Science:

• Microbiome Research
• Bioinformatics
• Genomics

Background:

• Metagenomic analysis commonly aligns sequencing reads to protein databases like NCBI-nr for taxonomic and functional profiling.
• Existing pipelines, such as DIAMOND+MEGAN, rely on large databases that are computationally intensive and not optimized for microbiome studies.

Purpose of the Study:

• To evaluate the AnnoTree protein database as a more efficient alternative to NCBI-nr for metagenomic analysis.
• To assess the impact of AnnoTree on taxonomic and functional assignment rates and computational performance.

Main Methods:

• Alignment of metagenomic reads against the AnnoTree protein database using DIAMOND.
• Taxonomic and functional binning of assigned reads using MEGAN, with support for both NCBI and GTDB taxonomies.
• Comparison of AnnoTree performance against the prokaryotic subset of NCBI-nr.

Main Results:

• A twofold speedup in analysis time was achieved using AnnoTree compared to NCBI-nr.
• Increased read assignment rates were observed, with a doubling of assignments to KEGG pathways.
• MEGAN now supports binning to the Genome Taxonomy Database (GTDB) in addition to NCBI taxonomy.

Conclusions:

• The AnnoTree database offers a computationally efficient and effective alternative for alignment-based metagenomic analysis.
• AnnoTree improves read assignment accuracy and expands taxonomic profiling capabilities within pipelines like MEGAN.
• The AnnoTree database is well-suited for metagenomic analysis and should be integrated into standard bioinformatics pipelines.