LRRfinder: a web application for the identificationC of leucine-rich repeats and an integrative Toll-like receptor database

Homology modelling is considered the most accurate technique for computational prediction of protein structure. However, this technique comes with fundamental caveats of dependency on template quality, identification of structural features and accuracy of alignment. Leucine-rich repeats (LRRs) characterise a diverse family of proteins. Recently resolved structures reveal a highly conserved region in LRRs that assemble into the curved parallel beta-sheet lining the inner circumference of their solenoid structure. Thus, prediction of these structurally important regions is essential in the comparative modelling of LRR proteins and their interactions. Here, we describe the generation of tLRRdb, a database of selected Toll-like receptor (TLR) sequences with annotated co-ordinates. Derived from this is LRRfinder, a web application for the identification of LRRs within user-defined sequences to facilitate identification of structurally important regions, particularly relevant for protein-protein interaction studies and classification of novel sequences. LRRfinder is available at: www.lrrfinder.com.