Frederic Bertels

Bioinformatics and computational biology

I completed my Diplom (Masters equivalent in Germany) in Bioinformatics at the Friedrich Schiller Universitaet in Germany in July 2007. In September 2007 I moved to New Zealand to work at the Bioinformatics Institute at the University of Auckland with Allen Rodrigo on projects related to metagenomics. Through Allen I met Justin O’Sullivan (IMBS, Massey Auckland) who needed help dealing with the output of Solexa genome re-sequencing data aimed at defining chromosome-chromosome interactions. Through Justin I became aware of the work in Paul Rainey's lab to find mutations in re-sequenced lineages of their experimental bacterial populations. To assist I devised strategies to enable data to be interrogated in order to identify the mutations of interest.

The short (36 nucleotide) reads from the Solexa sequencer (housed at the Allan Wilson Centre Genome Service) were mapped back to a reference P. fluorescens genome using Eland. This output, when appropriately displayed (both in Artemis and Gbrowse) and interogated made it easy to find point mutations.

Unfortunately this strategy did not work for short insertions or deletions (some of these were known before I began my data analysis). To overcome this problem I found a powerful way to interogate the sequences that Eland was unable to align and to map these discarded sequences back onto the genome.

PhD studies: repetitive extragenic palindromic sequences (REPs) in bacteria

Initial studies whetted my apetite for further research and aided by a studentship from the Allan Wilson Centre for Molecular Ecology and Evolution I have begun a PhD on the evolution and maintenance of repetitive extragenic palindromic sequences (REPs) in bacteria

Paul and Xue-Xian initially discovered REPs in the genome of P. fluorescens SBW25 when sequencing portions of the genome in their studies of plant-induced gene fusions. These sequences, which are especially numerous in the SBW25 genome, were reported in the genome paper by Silby et al. 2009, but numerous unanswered questions have led to my project.

I began by examining the frequencies of short sequences in the genome of SBW25. I then went on to describe the distribution and structure of highly over-represented words (short sequences). This unbiased starting position led me to conclude that the most abundant short sequences in the genome of SBW25 are REP sequences. During these analyses I discovered a protein family that is closely associated with REP sequences. Interestingly, this protein family shows similarities (albeit weak) to a known class of transposons that use short palindromes (a defining feature of REP sequences) as recognition sequences. Together, the observation that homologues of this family are associated with REP sequences in most completely sequenced Pseudomonas strains (also in Escherichia coli, Salmonella and other bacteria), has led me to conclude that REP sequences originate from flanking sequences of a new class of transposon. At the moment I am testing the hypothesis experimentally.

Publications from previous work

Rodley CDM, Bertels F , Jones B, and O'Sullivan JM. Global identification of yeast chromosome interactions using Genome conformation capture. Fungal Genetics and Biology 2009; 46(11): 879-886.

Rodrigo A, Bertels F, Heled J, Noder R, Shearman H, Tsai P. The perils of plenty: what are we going to do with all these genes? Philos Trans R Soc Lond B Biol Sci. 2008 Dec 27;363(1512):3893-902.