I think all data should be considered and gene diversity is a valuable piece of data whether it is STR or SNP diversity. I think all data should be considered in context, though. Are you saying that some data should be discarded if it disagrees with a particular point of view?
We are talking about this in another thread.
I don't know which thread you are talking about... STR Wars? If so let's talk about that over there so as not to bog down discussions about David Anthony's writings.
Diversity alone is not reliable,
I agree. In fact I just said that we should look at all
the data. Is there anyone who thinks we should look at diversity alone in a vacuum?
because the calculations often contain different lineages. Diversity is meaningful only if it for certain contains haplotypes from only one lineage.
I agree that measuring variance across different not closely related clades is not meaningful. I wasn't doing that. In this thread I was talking about U106 in its entirety, not just U106*.
Diversity of different R1b branches together is just as meaningless as the diversity of R1a, R1b and N1c together.
Your example is wrong and not at all equivalent. R1b encompasses a clade in its entirety. what I was talking about - U106, in its entirety, is a relatively young clade.
R1a, R1b and N1c together are not one clade. You have to go way, way back to get to a common ancestor for N and R. You'd have to include Q and P and more to get to an entire clade with a common ancestor.
I agree that diversity calculated by region is suspect. It's just another piece of data. I also agree that we should breakdown each clade into as many subclades as possible, analyze each and look at the components as well as the whole. I try to do that, whenever enough data is available.
It is not correct to say diversity is useless and totally discard it arbitrarily. There may be good reasons to dis-count (for some reason this word gets deleted so I had to stick in the hyphen) it for a for a particular comparison. It could also be a red herring as it could indicate a pooled/destination population rather than a launch pad population. The data may not be representative. However, it is not wise to summarily dismiss it. What do you say if diversity for a clade in one region is .1 and in another it is .9 ? That's a huge difference and diminishes the probabilities that the .1 region is the source.
... but okay, so you don't like STR diversity. I think it is more valuable than to analyze just frequency. Frequency is a better indicator oftentimes of the end of a trail rather than the beginning. If you disagree, that's fine but lets discuss over on your other thread.
Back to the topic here. What evidence do you offer that U106 or its predecessor lineages did not get to Northern Europe from an easterly source? Do you have any genetic evidence? Polako says he has no idea of U106's origin?
What do you say? Where did U106 originate?