I found interesting the observation in my reading about the Dzungaria gate that the southern Tarim route east tended to be taken by traders while the northern route towards the Dzungarian gate through the steppe tended to be taken by nomadic invaders. That might be something to chew over in terms of copper age movements east of R groups. The following maps indicate something interesting.
http://www.berzinarchives.com/web/images/en/map06.jpg
http://1.bp.blogspot.com/-wrNXSWPyDx....org+-+Map.jpg
We do have evidence for Bronze Age R1a in Tarim and lineages controlling nodes on trading networks would seem a likely explanation if that observation on Tarim being more of a trade route than a large scale migration one. However, Afansievo is located at the east end of the northern route through the Khazak steppe, the route apparently usually taken by invaders rather than traders. The culture appears to stop in the mouth of the Dzungarian gate and does not enter through it.
That is suggestive to me that they the Tarim mummies and Afanasievo may be different things altogether. I am not trying to suggest on in R1a and one is R1b as I dont think that works either but I think the two routes, two functions aspect needs to be factored in.
I think the evidence is confusing. We can see from the Altai study that today R1b (M73) is only found in a northern Altai group with links to a retreat east from further west and is close to absent in the south. R1a is more common in general in Altai. We also have both R1a and R1b in the modern Tarim area. Finally the Tarim mummies are so far R1a in terms of R.
I think there is a lot more to be teased out of the story of R's west to east phase in central Asia.
There seems to be a lot of R1a and b in Kazakhstan.
http://www.familytreedna.com/public/...ction=yresults
However the Hindu Kush study shows very little R1b compared to R1a in Kyrgyzstan and Mongolia although it only covers groups like Uzbeks, Turkmen, Tajiks etc via their element on Afghanistan its fair to say that they are also very light on R1b. The upshot of this is that contrary to some opinions R1b looks very light in the southern 'Stan' counties and appears to be more common among Iranians and Khazaks.