The majority of predicted proteins contained in the genomes of novel bacteriophage show little or no similarity to characterized proteins in the databases, and as such can only be described as hypothetical. This genetic diversity is surprising given that structures of phage and the functions that they carry out are relatively conserved. I was interested in the possible use of the lambda repressor fusion system as a tool in confirming our phage gene prediction. Bcep781 and Bcep56 were targeted for the identification of homotypic (self-self) interactions. To identify homotypic interactions in the proteins of Bcep781 and BcepMu a combined genomics and bioinformatics approach was used. Interactions were identified using a lambda repressor fusion system on a genome wide scale. A perl script was then written to extract and analyze the sequences of amino acids expressed in the fusion proteins of positive clones, and to compare them to predicted Bcep781 and BcepMu proteins. Fourteen potential homotypic interactions in ten different predicted Bcep781 proteins were identified, and sixteen different potential homotypic interactions were identified in fifteen different BcepMu proteins. The lambda repressor fusion system is thus useful both in identifying and localizing homotypic interactions in proteins, and in confirmation of predicted proteins for which no other data is available.
More information for the labmda repressor fusion system can be found at the Hu Lab's website.
A brief description of how the system works can be found here.
Last modified: December 16 2005 19:24:46.