mkdom

mkdom identifies and clusters protein domains using recursive PSI-BLAST homology searches to delineate conserved protein domain building blocks across Bacteria, Archaea, and yeast.

Key Features:

• Domain Clustering: Clusters protein domains by assuming the shortest amino acid sequence represents a single domain and using that representation for querying and clustering.
• Recursive Homology Searches: Employs recursive PSI-BLAST searches to detect homologous sequences and delineate frequent domain building blocks.
• Cross-Domain Analysis: Identifies domains unique to specific domains of life and those shared among Bacteria, Archaea, and yeast, marking commonly shared domains as fundamental building blocks.
• Multi-Domain Protein Statistics: Computes statistics on multi-domain proteins and highlights organisms with unusually high numbers of large multi-domain proteins such as Bacillus subtilis and Mycobacterium tuberculosis.
• Domain Shuffling and Permutation Analysis: Detects examples of highly shuffled domains and circular permutations to inform studies of protein evolution and functional diversity.
• Genome-scale Analysis: Applied to analysis of protein domain shuffling across 17 completed microbial genomes.

Scientific Applications:

• Evolutionary Biology: Investigates conservation and diversification of protein domains across Bacteria, Archaea, and yeast.
• Structural Bioinformatics: Studies protein architecture and the modular nature of domains for structural interpretation.
• Comparative Genomics: Compares domain distribution and composition across multiple genomes to identify lineage-specific and shared domains.
• Protein Evolution Studies: Examines domain shuffling and circular permutation events to understand mechanisms of protein functional diversification.

Methodology:

Performs recursive PSI-BLAST homology searches and clusters domains using the shortest amino acid sequence as the representative single domain; applies these analyses across genomes (17 completed microbial genomes) to delineate domain building blocks, detect domain shuffling and circular permutations, and compute multi-domain protein statistics.