BOMP

BOMP predicts integral beta-barrel outer membrane proteins from fasta-formatted polypeptide sequences of Gram-negative bacteria.

Key Features:

• Input Format: Accepts one or more fasta-formatted polypeptide sequences from Gram-negative bacterial sources.
• C-terminal motif search: Identifies a characteristic C-terminal pattern present in many integral beta-barrel proteins as an initial filter for candidates.
• Integral beta-barrel score: Calculates an "integral beta-barrel score" by analyzing sequence stretches for amino-acid patterns typical of transmembrane beta-strands to quantify beta-barrel likelihood.
• Performance metrics: Achieved 80% precision and 88% recall against SwissProt-annotated subcellular localization datasets for Escherichia coli K12 (788 sequences) and Salmonella typhimurium (366 sequences).
• Proteome analysis: Identified 103 candidate integral beta-barrel proteins in the predicted E. coli proteome of 4,346 sequences, including 36 predictions with no significant similarity (E-value < 1e-10) to SwissProt proteins with annotated subcellular localization.
• Cross-species estimation: Estimated integral beta-barrel protein content across predicted proteomes of ten bacteria, finding a range of 1.8% to 3%.

Scientific Applications:

• Microbial genomics: Enables genome-scale identification of outer membrane beta-barrel proteins in Gram-negative bacteria for annotation of predicted proteomes.
• Structural biology: Provides candidate lists of integral beta-barrel proteins for studies of transmembrane beta-strand architecture and folding.
• Pathogenesis and therapeutic targeting: Supports discovery of novel beta-barrel candidates relevant to bacterial physiology and pathogenesis and the selection of potential therapeutic targets.

Methodology:

Combines pattern recognition and sequence analysis by integrating a C-terminal motif search with evaluation of transmembrane beta-strand propensity to compute an integral beta-barrel score.