I am attempting to put a date to the emergence of a parasitic genus (Cuscuta) within a broader tree of species spanning the order Solanales using BEAST. In FigTree, there are the options to display node height and node age, which are identical except that I can scale node age using a user-inputted time scale. Node heights are in the range of about 0.080 to 0.004.
My question is how I should go about specifying that time scale? My preliminary settings in BEAUti included a prior distribution for the age of the common ancestor of Convolvulaceae and Solanaceae (mean = 66, range = about 48 - 86, in millions of years), and in my tree the corresponding node height is 0.0735. Would it be reasonable to simply set my time scale factor to 1000 to get a node age of 73.5 my (which is within the priors I set)?
It seems unlikely that I should be setting my time scale solely based on my previous ideas, because in that case I could more or less make the ages of my nodes anything I like. What would you suggest as a method of setting the time scale in BEAUti?
Are you suggesting that the inferred node age of Cuscuta should be 73.5 my? Since it is a member of Convulvulaceae, that would put the age of the genus older than your prior estimate of the mean age of the origin of its containing family. That does not seem right. Perhaps it should be 7.35 my?
In order to calibrate your time scale on the tree you need to have an idea from other evidence (not your own study) for the divergence time of related taxa, ideally the most recent common ancestor if the branch length to it is not too distant. If it is clearly a descendent of the most recent common ancestors then scale by that date and assume that the mean date for the taxa of interest is less than 66 (not more that 66 million years). The confidence interval on your estimated age of your taxa will be necessarily large but cannot be greater than the mean of your calibration point. The CIs should not overlap. Otherwise you have insufficient evidence to show when after the most recent common ancestor that your taxa appeared.
Please let me know if you do not understand my explanation.
Coming back to this project after a break, thank you for the responses!
Andrew V Z Brower I am suggesting 73.5 my for the node age of the MRCA of Convolvulaceae and Solanaceae.
Patrice Showers Corneli I plan on comparing the dates for the emergence of Cuscuta I get from various combinations of calibrations to get a final date range, I just wanted to ensure that I am using the software correctly.
Perhaps I should clarify my question: is it any more correct to use a scale factor of 1000 (to get a date of 73.5) than for example 700 (to get a date of 51.45)? How should I go about choosing a scale factor? And when I go to try other calibrations, should I continue using the same scale factor or can that change?
- Badges
- Science topic
- Similar topics
- Filing