I suspect that the comparison of physical and genetic distance is duplicating
e.g. pop1 vs pop2 AND pop2 vs pop1 both being retained.
This would be a problem for the model (overloading data points that are identical), and for the plot.
There should be a total number of comparisons that are shown as a QC step when this is done so the user gets informed on what contrasts are being made.
I suspect that the comparison of physical and genetic distance is duplicating e.g. pop1 vs pop2 AND pop2 vs pop1 both being retained. This would be a problem for the model (overloading data points that are identical), and for the plot. There should be a total number of comparisons that are shown as a QC step when this is done so the user gets informed on what contrasts are being made.