LAMPA

"LAMPA" (LArge Multidomain Protein Annotator) is a computational tool to enhance the accuracy of estimating statistical significance in profile-profile searches of sequence similarity, particularly for multidomain proteins. Recognizing the challenges posed by using whole proteins as queries—due to the difficulty in establishing homology for similarities near cutoff levels and the complexity introduced by multidomain structures—LAMPA introduces an iterative approach to expand hit coverage gradually.

Key Features and Functionalities:

- Iterative Expansion of Hit Coverage: LAMPA systematically expands the hit coverage of multidomain proteins by re-evaluating the statistical significance of hit similarity using progressively more minor queries defined at each iteration. This approach allows for more precise domain delineation and homology detection.

- Integration with TMHMM and HHsearch: To facilitate its analysis, LAMPA employs TMHMM for the recognition of transmembrane regions and HHsearch for homology detection, combining these tools to significantly improve the annotation of protein domains.

- Pfam Database Annotation: Utilizing the Pfam database, LAMPA provided annotations that outperformed RefSeq expert annotation regarding the number of regions and annotated length for a significant portion of RNA virus polyprotein entries.

- Rationalization of Results: The improvement offered by LAMPA over traditional methods was rationalized based on the dependencies of HHsearch hit statistical significance on the lengths and diversities of query-target pairs, highlighting the importance of considering these factors in sequence similarity searches.