Can i determine the threshold of interspecific distance according to my own data? I've read lots of publications about DNA barcoding and i know the interspecific distance is almost clear for COI. (for vertabrates, especially fishes). But, what if my interspecific distances are lower than the cited publications? (and i'm definitely sure that i've done species identification correctly). Is that possible and is it a sign of cryptic diversity?
The idea of a genetic threshold to delimit species boundaries is tragically flawed. To answer your first Q: I do not think you can or should cherry-pick your own data when you know there are other data out there that are contrary to your main hypothesis. Unfortunately, there is nothing we can do to get away from the simple fact that any measure or threshold gleaned from genetic data is going to be 100% arbitrary. Unless you have a full phylogeny with data from every species (or at least a reasonable taxon sampling effort) for that particular marker, you are going to have some degree of assumptions. If you have all those data, then you can easily make the argument. If you do not have those data, I would say "approach with caution!". If you are trying to get this published, any solid reviewer is going to nail you on this point. Here is some more advice: get more data from a diverse set of characters (morphology, calling/behavior, etc.). If you have lower than expected P distances, then your argument is going to be weak if you think those should be different species. I would suggest that either you have a heterogenous meta-population scenario and/or you have some degree of cryptic species complex going on. Most likely, the former - and only because people have a 'splitting' bias these days. Don't be afraid to embrace the biological truth that is messy and ungainly for scientific analysis: species are hazy, floppy, and nebulous entities = nature doesn't care about our pigeon-holes :-) If you can, embrace the mess, get as many different data (i.e., not just genetic) and be transparent about your results. Good luck!
Dear Dilruba Seyhan,
Standard limit between intra- and inter-species divergence ( 2% of divergence or the 10x rule ) for COI gene has been reported in many research findings in fish species. It is normal to find intra- and interspecific distances frequently overlap, and visually defining a threshold becomes difficult as there is mutation rate varies considerably among genes and species. Results obtained from the COI gene is known to possibly be affected by several biases, ideally it should be combined with the analysis of other mitochondrial genes as well as nuclear along with morphological, geographical or ecological data to clearly delimit species. Combination of genes for analysis will help in case of cryptic species. Best wishes.
It all depends on which group you are working with. The best way to start is finding papers that deal with groups that are very close to the one you are researching. Try to use them as a reference for your data. As mentioned in the two replies, try to include other elements, especially morphological characters. It is also advisable to get the opinion of an expert who deals with classical taxonomy of your group, if you want to propose cryptic species with interspecific distances between 2 and 3%.
I agree with the first two responses to this question. Higher resolution markers (microsats, NGS) often show very strong differentiation between sister species that have diverged comparatively recently, while COI will not pick up any differentation. The problem is that this marker evolves quite slowly (e.g. 1% per million years in some inverts). New species may already be reproductively isolated, adapted to different environmental conditions and look different, but COI will only reflect this a few 1000-100 000 years from now. So while strong differentiation indicates that you probably deal with different species, lack of differentiation does not mean that the study populations are members of the same species.
- Badges
- Science topic