dbCAN2

dbCAN2 annotates carbohydrate-active enzymes (CAZymes) in genomic and metagenomic datasets to characterize CAZyme repertoires involved in complex carbohydrate metabolism.

Key Features:

• HMMER Search: Utilizes the dbCAN hidden Markov model (HMM) database to identify CAZyme sequences based on protein family HMMs.
• DIAMOND Search: Performs rapid alignments against the pre-annotated CAZy sequence database to identify known CAZymes.
• Hotpep Search: Searches against a curated database of conserved short peptides specific to CAZymes to enhance detection sensitivity.
• Multi-tool consensus filtering: Combines outputs from HMMER, DIAMOND, and Hotpep and filters out CAZymes detected by only one method to improve annotation accuracy.
• Nucleotide sequence input: Accepts nucleotide sequences directly for CAZyme annotation.
• CAZyme Gene Cluster (CGC) prediction: Predicts physically linked CAZyme gene clusters to aid identification of putative polysaccharide utilization loci (PULs).

Scientific Applications:

• Genome and metagenome CAZyme annotation: Annotating CAZymes within genomes and metagenomes to define CAZomes.
• Plant and microbial community mining: Mining plant genomes and associated microbial communities for CAZymes involved in plant carbohydrate metabolism.
• CAZyme gene cluster and PUL discovery: Identifying CAZyme gene clusters and putative polysaccharide utilization loci in microbial genomes or metagenomes.
• Biomass and feedstock studies: Characterizing enzymes relevant to complex carbohydrate degradation for research on renewable energy, materials, and biological utilization of plant biomass.

Methodology:

Computational methods explicitly include HMMER searches using the dbCAN HMM database, DIAMOND searches against the pre-annotated CAZy sequence database, Hotpep searches against a curated conserved short-peptide database, combination of outputs with filtering of single-tool hits, and prediction of physically linked CAZyme gene clusters (CGCs).