Computationally identifying novel essential genes, including non-coding RNAs, in intergenic sequences in bacterial genomes

The increasing availability of bacterial genome sequences and genome-wide laboratory analyses has opened the door for high-throughput bioinformatic approaches to accelerate the discovery of antimicrobial drug targets. To expand the list of such possible drug targets, I have developed an approach to identify novel essential genes encoded in unusually large intergenic regions between previously annotated genes. Applying this approach to the analysis of the intrinsically antibiotic resistant pathogen P. aeruginosa PAO1, I computationally predicted at least 5 novel protein-coding genes that were also confirmed to be transcribed by RT-PCR. In addition, at least 5 novel non-coding RNAs were predicted. I used the same computational pipeline to predict such novel putative genes in Mycobacterium tuberculosis H37Rv, reflecting the general applicability of the method. Finally, I analyzed the phylogenetic distribution of the essential genes, providing insight into their evolutionary origins.