Mantis

Mantis annotates protein sequences by homology-based methods to generate consensus-driven functional annotations across integrated reference datasets for (meta-)omics analyses.

Key Features:

• Integration of Multiple Reference Data Sources: Dynamically integrates multiple high-quality reference datasets to generate unified annotations.
• Customization: Supports inclusion of custom reference data sources to tailor annotations to specific research projects.
• Advanced Annotation Methodology: Combines text mining with consensus integration and uses a depth-first search algorithm for domain-specific annotation, producing an average precision increase of 0.038 compared to sequence-wide methods.
• Efficiency and High-Quality Output: Annotates an average genome in 25–40 minutes and reports average coverage of 81.4% and precision of 0.892.
• Reproducibility: Designed to produce reproducible annotation results across different environments and datasets.

Scientific Applications:

• Genomics: Assigns functional annotations to protein-coding genes in isolate genomes and genome assemblies.
• Proteomics: Identifies functional domains within proteins to support proteomic functional interpretation.
• Systems Biology: Provides high-coverage functional annotations useful for systems biology and pathway analyses.
• Metagenomics: Scales from single genomes to large metagenomes to annotate proteins from complex microbial consortia in (meta-)omics studies.

Methodology:

Performs homology-based annotation integrating multiple reference datasets, applies text mining for annotation reconciliation, and employs a depth-first search algorithm for domain-specific annotation to produce consensus-driven outputs.