Pre_GI : a dynamic Catalogue and set of computational tools for the ontology and stratigraphy of horizontally transferred genomic islands in bacterial genomes

Abstract

The non-genealogical transfer of genetic information between prokaryotes is a frequent and omnipresent event. The acquisition of foreign genomic segments may aid organisms in adaptation to novel or extreme habitats with these rapid evolution events phenotypically bene cial to the recipient. These regions of atypical and foreign origin are vernacularly termed islands and are identi ed by their unique local genomic signature or composition which di ers from the global host genomic signature. The SeqWord Genomic Island Sni er program utilizes tetranucleotide frequency patterns and statistics to identify regions of probable horizontal transfer. Optimum parameter values were determined for this compositional-based island identi er to ensure acceptable levels of false negative and false positive occurrence. Post-identi cation island analysis is demonstrated with the aid of the LingvoCom package available from the SeqWord project. This island identi er was furthermore compared with other existing transfer detection packages to indicate relevance and reliability. The continued identi cation of islands in prokaryotic genomes requires a novel and functional repository with the ability to expand as newly sequenced archaeal/bacterial genomes are available. The amalgamation of a robust database, convenient interface and island analysis tools presents a novel avenue in prokaryotic island research. The Predicted Genomic Islands database currently houses 26,744 islands identi ed in 2,407 archaeal/bacterial genomes and is freely available from http://pregi.bi.up.ac.za/. The database serves as an island information hub and collection of analytical tools allowing users the ability to address a myriad of horizontal transfer and island ontology research questions. Inclusion of various novel island information criteria and analytical tools may distinguish this platform from extant island databases and tools. Novel island comparison against the current content enables a rapid yet reliable tool set in the age of brisk and economically e cient genome sequencing. The collection of all island information in a single set allows for various avenues of research and stratigraphy of islands by allowing for the deconstruction and inspection of layers in archaeal/ bacterial island communities and exchange between island hosts. The capability of this island garage was previewed with novel island identi cation results and research directions in an attempt to convey the future potential. This collection of island information and tools may prove a reliable and innovative approach in variable elds of horizontal transfer and island research with numerous applications and associations.