| dc.description.abstract | The dramatic drop in sequencing costs has created many opportunities for novel biological research. Many research questions depend on comparing sequenced reads to a “reference genome” to characterise genes, regulatory regions, genetic variation and gene expression. Typically, the reference genome is computationally assembled de novo from reads generated by “shotgun sequencing” and is often the first step in the molecular characterisation of a species. Unfortunately, this process is prone to errors, which results in several regions of the genome to be missing, fragmented, or mis-assembled in the reference. I review the sources of these errors, the challenges in evaluating the quality of de novo genome assemblies, present new metrics and a tool to overcome some of the limitations. Furthermore, I show that fine tuning the parameters of assembly software is an effective way to obtain higher quality genome assemblies. However, the quality of genome assemblies is ultimately tied to the length and the error profile of sequenced. I present a tool to facilitate rapid transfer of gene annotations to a new genome assembly with high confidence so that known regions of the genome do not need to be recharacterized. Furthermore, intermediate output of the tool can also be used to transfer variant annotation and other data formats. Finally, I present a graphical interface for the popular BLAST software. Among other things, it is useful for qualitative quality assessment of genome assemblies and gene annotations or to characterise regions of interest in newly assembled genomes. I believe the approaches presented here can play a key role in genomic characterisation of previously understudied organisms. As examples, I present two studies on social evolution in ants where I led specific analyses through the application of above-mentioned tools and related concepts. | en_US |
