Sedimentary ancient DNA: Exploring methods of ancient DNA analysis for different taxonomic groups

Abstract

This thesis explores the potential of sedimentary ancient DNA (sedaDNA) from lakes as a tool for studying past communities. First it sets out to determine what taxa in the environment are most likely to be detected in lake sediments. For this, vegetation surveys were carried out around eleven different lakes in Norway and the taxa detected based on plant metabarcoding of the lake sediments were compared to the recorded flora. The taxa detected in the lake sediments primarily resembled those growing either in the lake or within the immediate surrounding. Furthermore, taxa abundant in the vegetation were more likely to be detected based on DNA. Three different studies were carried out to compare plant sedaDNA results to traditional palaeoecological proxies such as pollen or macrofossils. The results indicate that overall sedaDNA is capable of detecting more taxa than the alternative methods, though differences are observed between the shotgun and metabarcode methods tested, as well as the lakes themselves. A separate study aimed to detect mammals based on sedaDNA, which proved to be more problematic. Co-amplification of off-target taxa, the presence of common lab contaminants and a general low amount of template material hindered the identification of mammals present. However, the detection of “bycatch” worm taxa demonstrates their potential for palaeoenvironmental reconstructions. Finally, the shotgun sequencing of two algal rich sediment layers allowed for both the reconstruction of the algal organelle genomes, as well as the exploration of the present within lake populations.