LAMPA

LAMPA annotates multidomain proteins by iteratively refining profile-profile searches to improve domain detection and delineation in RNA virus polyproteins.

Key Features:

• Iterative Approach: LAMPA iteratively re-evaluates the statistical significance of profile-profile hits using progressively smaller query segments to refine domain delineation when borders are unknown.
• HHsearch integration: LAMPA employs HHsearch for profile-profile homology detection during iterative searches.
• TMHMM integration: LAMPA uses TMHMM to identify transmembrane regions and incorporate them into protein annotations.
• Pfam database annotation: LAMPA annotates multidomain proteins using the Pfam database, focusing on polyproteins exceeding 1000 amino acids common in RNA virus proteomes.
• Statistical significance handling: LAMPA addresses challenges in estimating statistical significance during profile-profile searches for protein domains.

Scientific Applications:

• RNA virus polyprotein annotation: Applied to RNA virus polyproteins to improve structural and functional characterization of viral proteins.
• Improved homology detection: Compared with using intact polyproteins as queries in HHsearch, LAMPA identified 507 additional homologous regions in 14.4% of analyzed polyproteins under strict cutoffs.
• Superior annotation performance: LAMPA outperformed RefSeq expert annotations by region number and annotated length for 69.3% of RNA virus polyprotein entries.

Methodology:

LAMPA performs iterative refinement of query segments for profile-profile searches, progressively adjusts queries based on statistical significance assessments, and uses computational experiments to evaluate dependencies between local alignment similarity scores and query-target lengths and diversities.