A while ago, I said that it sure looked as if David Reich and Nick Patterson were working out the origin and scope of the Indo-European invasion of Europe. They have more to say on that.
The abstract for their upcoming talk (SMBE 2014) makes several points.
First, they looked at a drudge of skeletons from Samara, in Russia, dating between 3,000 to 9,000 years before present. These samples had the Ancient North Eurasian admixture (Sibermen), and thus are a possible source for that late-arriving genetic component in Europe. There was population turnover here too: the oldest skeletons had U4 and U5 mtDNA, while the later ones included W, H, T, I, K and J.
They go on to model how the steppe samples may relate to ancient and contemporary DNA samples from Europe, the Caucasus, and South Asia – sounds like a model of Indo-European dispersal. And, probably a model of the Sibermen dispersal. Details are not in the abstract.
I would guess that their evidence indicates that the Indo-Europeanization of Europe entailed massive population turnover in northern Europe, less in southern Europe, none in Sardinia and among the Basques. I think that you see this genetic trace of the Indo-Europeans in the Uighurs, but very little in Iran/Pakistan/India: that late expansion of charioteers may have involved elite dominance, or perhaps a multistep process that diluted away much of the original genetic signature. Maybe not quite diluted to zero: we do see the European lactase-persistence allele in India. Of course a wildly advantageous allele like that could be transmitted by the smallest bit of admixture.
I suspect that the SNP next to KITLG that causes blondness was fairly common in that invasion: today it is quite rare in the Basques and Sardinians, fairly common in northern Europe and the Uyghurs, very rare in India.