Comprehensive computational analysis of leucine-rich repeat (LRR) proteins encoded in the genome of the diatom Phaeodactylum tricornutum.

We have screened the genome of the marine diatom Phaeodactylum tricornutum for gene models encoding proteins exhibiting leucine-rich repeat (LRR) structures. In order to reveal the functionality of these proteins, their amino acid sequences were scanned for known domains and for homologies to other proteins. Additionally, proteins were categorized into different LRR-families according to the variable sequence part of their LRR. This approach enabled us to group proteins with potentially similar functionality and to classify also LRR proteins where no characterized homologues in other organisms exist. Most interestingly, we were able to indentify several transmembrane LRR-proteins, which are likely to function as receptor-like molecules. However, none of them carry additional domains that are typical for mammalian or plant-like receptors. Thus, the respective signal recognition pathways seem to be substantially different in diatoms. Moreover, P. tricornutum encodes a family of secreted LRR proteins likely to function as adhesion or binding proteins as part of the extracellular matrix. Additionally, intracellular LRR-only proteins were divided into proteins similar to RasGTPase activators, regulators of nuclear transport, and mitotic regulation. Our approach allowed us to draw a detailed picture of the conservation and diversification of LRR proteins in the marine diatom P. tricornutum.