ConDiGA

ConDiGA optimizes taxonomic annotation of genes from metagenomic data to construct protein sequence databases for metaproteomic analysis.

Key Features:

• Contigs-directed annotation approach: Uses the most confident species from contig annotation results as reference to annotate genes, rather than direct gene-level or contig-level taxonomic assignment.
• Optimized pipeline benchmarking: Demonstrates improved performance compared to alternative metagenomics-based database construction strategies (MD1 and MD2) using synthetic microbial communities, fecal samples, and in silico mixed datasets.
• Multiple annotation tool compatibility: Supports integration with BLAST, Kaiju, and Kraken2 for taxonomic annotation, with Kaiju showing optimal performance when combined with the ConDiGA (MD3) approach.
• Benchmarking datasets: Generates publicly available unique metagenomic and metaproteomic datasets of 12 bacterial species as a standard benchmarking sample for evaluating analysis pipelines.

Scientific Applications:

• Metaproteomic database construction: Builds accurate protein sequence databases from metagenomic data for compositional and functional characterization of microbial proteins in microbiota.
• Microbiome analysis: Enables in-depth metaproteomic analysis of complex microbiota samples by addressing taxonomic annotation reliability in metagenomics-derived protein sequence databases.
• Pipeline evaluation: Provides standardized benchmarking data for comparing and evaluating various metagenomics-based database construction and annotation pipelines.

Methodology:

ConDiGA (MD3) takes the most confident species from contig annotation results as reference to annotate genes, in contrast to MD1 where genes are directly annotated against the whole bacterial genome database, and MD2 where contigs are annotated against the whole bacterial genome database and the taxonomic information of contigs is assigned to the genes.