Some of the most significant breakthroughs in the biological sciences this century will emerge from the development of next generation sequencing technologies. The ease of availability of DNA sequence made possible through these new technologies has given researchers opportunities to study organisms in a manner that was not possible with Sanger sequencing. Scientists will, therefore, need to embrace genomics, as well as develop and nurture the human capacity to sequence genomes and utilise the 'tsunami' of data that emerge from genome sequencing. In response to these challenges, we sequenced the genome of Fusarium circinatum, a fungal pathogen of pine that causes pitch canker, a disease of great concern to the South African forestry industry. The sequencing work was conducted in South Africa, making F. circinatum the first eukaryotic organism for which the complete genome has been sequenced locally. Here we report on the process that was followed to sequence, assemble and perform a preliminary characterisation of the genome. Furthermore, details of the computer annotation and manual curation of this genome are presented. The F. circinatum genome was found to be nearly 44 million bases in size, which is similar to that of four other Fusarium genomes that have been sequenced elsewhere. The genome contains just over 15 000 open reading frames, which is less than that of the related species, Fusarium oxysporum, but more than that for Fusarium verticillioides. Amongst the various putative gene clusters identified in F. circinatum, those encoding the secondary metabolites fumosin and fusarin appeared to harbour evidence of gene translocation. It is anticipated that similar comparisons of other loci will provide insights into the genetic basis for pathogenicity of the pitch canker pathogen. Perhaps more importantly, this project has engaged a relatively large group of scientists including students in a significant genome project that is certain to provide a platform for growth in this important area of research in the future.

No abstract

Removal of introns from transcribed RNA represents a crucial step during the production of mRNA in eukaryotes. Available whole-genome sequences and expressed sequence tags (ESTs) have increased our knowledge of this process and revealed various commonalities among eukaryotes. However, certain aspects of intron structure and diversity are taxon-specific, which can complicate the accuracy of in silico gene prediction methods. Using core genes, we evaluated the distribution and architecture of Fusarium circinatum spliceosomal introns, and linked these characteristics to the accuracy of the predicted gene models of the genome of this fungus. We also evaluated intron distribution and architecture in F. verticillioides, F. oxysporum, and F. graminearum, and made comparisons with F. circinatum. Results indicated that F. circinatum and the three other Fusarium species have canonical 59 and 39 splice sites, but with subtle differences that are apparently not shared with those of other fungal genera. The polypyrimidine tract of Fusarium introns was also found to be highly divergent among species and genes. Furthermore, the conserved adenosine nucleoside required during the first step of splicing is contained within unique branch site motifs in certain Fusarium introns. Data generated here show that introns of F. circinatum, as well as F. verticillioides, F. oxysporum, and F. graminearum, are characterized by a number of unique features such as the CTHAH and ACCAT motifs of the branch site. Incorporation of such information into genome annotation software will undoubtedly improve the accuracy of gene prediction methods used for Fusarium species and related fungi. KEYWORDS Fusarium intron splicing spliceosomal introns cis-elements gene predictionDespite the increasing availability of whole-genome sequence information for diverse eukaryotes, genome annotation generally remains challenging (Yandell and Ence 2012). Apart from the computational complexities associated with the assembly of data generated by certain sequencing platforms, gene finding is particularly problematic (Loveland et al. 2012). This is mainly due to incomplete information on the inherent peculiarities associated with the structures of genes in different organisms, especially with regard to intron architecture, which severely limits optimization of ab initio gene prediction tools (i.e., methods that use characteristic DNA sequences for genome annotations) for genome annotation (Ter-Hovhannisyan et al. 2008). Here the main issues are typically restrictions, not only in the availability of preexisting and accurate gene models, but also high-quality and appropriate reference genomes for accurately predicting intron-exon structures (Ter-Hovhannisyan et al. 2008).The noncoding DNA sequences that interrupt the nuclear proteincoding genes of eukaryotes are referred to as spliceosomal introns (Bhattacharya et al. 2000). In contrast to group I and II introns with ribozymic activity (Bhattacharya et al. 2000), spliceosomal introns require the action of a lar...

No abstract