DNA barcodes and meiofaunal identification

Abstract

In recent years there has been a desire to definitively catalogue the life on our planet. In light of the increasing extinction rates that are driven by human activities, it is unlikely that this will be achieved using traditional methods. Whilst most organisms which have a body size of more than 1cm have been described, the vast majority of animal life is smaller than this, collectively known as meiofauna, and is yet to be catalogued. Meiofaunal organisms present a range of problems for traditional taxonomy. Firstly they are microscopic, meaning that morphological features are often difficult to resolve. Secondly these creatures often exhibit cryptic diversity meaning that different species often look the same. Thirdly, it is often the case that the organisms are poorly described in the literature making it very difficult to confirm identification, assuming that someone has already described it. It is possible, however, to obtain DNA sequences from these organisms. DNA barcoding, the use of short sequences of DNA to identify individuals, is now commonly used in a wide range of applications. It has been proposed that a single target gene should be sufficient to describe all organisms this way. Barcodes can be acquired from individuals or from bulk extractions from environmental samples. In the latter case, many of the sequences obtained are novel and unlikely to ever have a type specimen associated with them. When this is the case, assessing the diversity of a sample becomes a computational exercise. However, as yet, there is no agreed standard method adopted for analyzing the barcodes produced. Indeed most methods currently employed lack objectivity. This thesis investigates the efficiency of a range of gene targets and analysis methods for DNA barcoding, with an emphasis on meiofaunal organisms (nematodes, tardigrades and thrips). DNA barcodes were generated for up to three genes for each specimen. Sequences for each gene were analysed using two programs, MOTU_define.pl and DOTUR. These programs use different methods to assign sequences to operational taxonomic units (OTU), which were then compared. An objective method for analysing sequences such as MOTU_define.pl, which relies on only the information contained in the sequences, was found to be most suitable for designating taxa. It does not attempt to apply evolutionary models to the data, and then infer taxa from the derived data. In addition to barcoding, some samples were pre-processed using video capture and editing (VCE). This creates a virtual slide of a specimen so that a sequence can be linked to a morphological identification. VCE proved to be an efficient method to preserve morphological data from specimens.