Development of molecular markers for studying population structure in marine fishes : 'coding versus non-coding DNA' : which is best?
This thesis set out to assess whether variants identified within either the "coding" or the "non-coding" regions of the genome were more effective in resolving population structure in marine fishes. The assessment was carried out on three marine fish species - haddock (Melanogrammus aeglefinus), the lesser sandeel (Ammodytes marinus) and the Atlantic salmon (Salmo salar). The study was extended to compare variation within the "coding and "non-coding" regions of a section of the transferrin gene directly with nucleotide sequencing of the cDNA isolated from haddock and sandeels. Variants were identified both the "coding" and "non-coding" regions of this section of the transferrin gene selected and a population analysis was carried out to assess which source of variants was more effective for resolving population differences. Overall, the findings of this study were inconclusive. In some cases variation identified within coding regions was better than variation identified in non-coding regions of the transferrin gene for resolving population structure. However, in other cases the opposite was true, or both sources of variation were found to be uninformative with regards to genetically differentiating between populations. This inconclusiveness was partially attributable to the small number of variants that were identified in coding region for both haddock and Atlantic salmon and the fact that variation within only a single gene was studied. This study did, however, indicate that the methodology employed for screening individuals has clear advantages over previous methods used for population analysis, e.g. protein electrophoresis and mini/microsatellite analysis. All the variants present in the transferrin gene were detected using nucleotide sequencing. In contrast, using protein electrophoresis only those variants that cause an amino acid change and can be detected using histochemical stains are revealed. Difficulties were, however, experienced in accurately assigning allele sizes using mini- and microsatellites and led to binning of the data. This had the disadvantage that by increasing the accuracy of the data set, much of the resolution is likely to be lost.