Before I dive back into the world of marine Klee-diagrams I'd like to thank Mark Stoeckle who as one of the inventors of the method provided very valuable feedback especially when I worked on the datasets presented here. Thanks Mark, greatly appreciated.
Klee-diagrams can also offer ‘fingerprints’ characterizing the diversity of different regions or habitats e.g. in different bodies of water. Here, I used two COI datasets of fishes from both Polar Regions to test the power of the method in a comparative setting.
|Klee diagrams for polarfishes. A. Indicator vectors for 132 Arctic fish species. b. Indicator vector correlation (n=2) for 155 Antarctic fish species.|
Both datasets represent around 50% of the currently known species in the respective regions and span the over all major groups of fishes that have been encountered in Arctic and Antarctic waters. Almost 50 % of all Arctic fish species belong to the order of the Scorpaeniformes with 62 of them represented in my analysis (diagram a). In Antarctic waters only the one scorpaeniform family is found, the liparidae (Snailfishes). In contrast to morphological data, the COI data provide only little resolution for many liparid species. This analysis contains 13 representatives of three genera which is one third of the known diversity. The diagrams clearly document the difference in diversity and number of the scorpaeniformes including the poor resolution among the liparids in both regions.
The predominant group of Antarctic waters are the Notothenioidei with largely endemic species. There are no Arctic counterparts known. The notothenioids are often considered the only marine fish species flock and their close relationships are shown in the Klee-diagram with as high correlations.
The eelpouts (zoarcidae) occur in both regions and a large number of species exhibit profound character plasticity and are often obscuring true relationships. Species are subdivided into subspecies on the basis of characters now known to vary individually or within populations. The results are often low congeneric divergences that can be observed especially in diagram b.
Representatives of the gadiformes are found in both regions although they are known to be more speciose in Arctic waters. All members of this family exhibit shallow divergence especially among the Arctic representatives. The slightly lower correlation depicted within the Antarctic group is corroborated by recent COI barcode-based discoveries of cryptic species.
The comparison demonstrated here is limited to one taxon and one gene, thus rather serving as proof-of-concept using two sets of data that contained fully identified and verified sequences. Nevertheless known diversity patterns, similarities and discrepancies within and among groups are clearly shown, and commonalities can be retrieved in detail through the sequence order. In concert with the scalability of the method they show the applicability for a new generation of sequence data generated e.g. in environmental barcoding projects utilizing next-generation sequencing. Even with a limited ability to identify all sequences in a given sample the chance of comparing genetic diversity of different standardized samples at a single locus will help to understand and show community differences at an unprecedented scale. This could include differences between regions, changes over time, or responses to environmental changes. Eventually changes could be monitored by looking at a full picture instead of relying on single representatives such as indicator species. By looking at the DNA Barcode composition of different samples we would be able to have a birds-eye view of entire ecosystems as opposed to isolated snapshots. The challenge lies in the standardization of such datasets that allow the observation of real differences and similarities in patterns revealed through Klee-diagrams.
... to be continued