[Update: please read my 2012 independent analysis of Western Jews, Palestinians and other populations of West Asia, which adds some interesting nuances but in essence confirms what is said here].
This does not seem just another paper on Jewish genetics but more like The Paper. While it is pay per view and hence I haven't been able to read it in full , the material I could see at Dienekes' blog (same as in the supplementary material) is most revealing. I think this research will mark a before and after in Jewish genetics and also gives some interesting hints on other populations, specially in West Asia and North Africa.
Doron M. Behar, Bayazit Yunusbayev, Mait Metspalu et al. The genome-wide structure of the Jewish people. Nature, 2010. Pay per view but supplementary material freely accessible.
One of the good things is that finally comparison with Turks and other populations from that area where early Jewish Diaspora in the Hellenistic-Roman era is known to have lived, rather than in Palestine, in a time when Judaism (several sects including eventually Christianism) was still actively proselytizing.
Unlike what I used to think, Western Jews (Sephardi and Ashkenazi) do not cluster too well with the Turkish sample but they cluster almost perfectly with Cypriots and Lebanese. They seem to have no particular relation with Palestinians (nor Druze nor Negev Bedouins) but these also appear clearly different from other Arabs and in general any other sampled population.

This K-means analysis is for me the answer to all these endless discussions on the origin of Jews and Palestinians. Western Jews seem essentially to have coalesced in the Cypriot-Phoenician area probably by, essentially, conversion. Palestinians seem to be a uniquely distinct population, albeit somewhat admixed with their neighbors, which may well originate with the local Neolithic and certainly must have been there in early historical times.
In other words: Palestinians are most likely to be the true descendants from the Jews of the Biblical period, rather than modern Western Jews who seem more as originating from a Phoenician-Cypriot population which converted to Rabbinic Judaism for whichever reasons.
Other Jewish populations also seem to originate in conversion episodes but from different genetic pools. Hence Yemeni Jews appear as genetically Arab, Ethiopian Jews as Ethiopian, Indian Jews as Indian and Iraqi-Iranian Jews as Iranians. There may be some fine threading to enrich this overall picture but the essentials seem very clear in any case.
Anyhow, Moroccan Jews appear as just another branch of Western Jews (no particular relation with Moroccans is apparent) and I have not been able so far to identify the closest population to the small sample of Uzbek Jews. Also Turco-European Jews seem at least somewhat admixed with Europeans, something that was already well known.
As for other populations, I find interesting that an specific autosomal component of NW Africans has been detected (for the first time as far as I can tell). Many specific clusters have obviously not been detected because of the relative shallowness of the K-means analysis.
The supplementary materials have other interesting graphs, PC analysis of autosomal, Y-DNA and mtDNA genetics and a global K-means analysis of relevance when observing some peripheral Jewish populations specially but also providing some general hints on other populations (a very wide sample, specially in Eurasia).
I must cheer and congratulate the authors of this research for finally addressing the debate on Jewish (and Palestinian) origins with a worthy extended sample of many many populations, including key ones such as Turks, Cypriots and Lebanese (among others). The wording of the abstract doesn't say things as clearly as I do but their data speaks volumes.
Update: A reader, Joe, tells me that H.G. Wells already suggested this Phoenician true origin of Jews. In his book A Short History of the World, chapter XXII, he wrote, speaking of Semitic peoples, once all powerful but then defeated by the Indoeuropeans (Persians, Greeks, Romans):
Is it any miracle that in their days of overthrow and subjugation many Babylonians and Syrians and so forth, and later on many Phoenicians, speaking practically the same language and having endless customs, habits, tastes and traditions in common, should be attracted by this inspiring cult [Judaism] and should seek to share in its fellowship and its promise? After the fall of Tyre, Sidon, Carthage and the Spanish Phoenician cities, the Phoenicians suddenly vanish from history; and as suddenly, we find not simply in Jerusalem but in Spain, Africa, Egypt, the East, wherever the Phoenicians had set their feet, communities of Jews.
Is it any miracle that in their days of overthrow and subjugation many Babylonians and Syrians and so forth, and later on many Phoenicians, speaking practically the same language and having endless customs, habits, tastes and traditions in common, should be attracted by this inspiring cult [Judaism] and should seek to share in its fellowship and its promise? After the fall of Tyre, Sidon, Carthage and the Spanish Phoenician cities, the Phoenicians suddenly vanish from history; and as suddenly, we find not simply in Jerusalem but in Spain, Africa, Egypt, the East, wherever the Phoenicians had set their feet, communities of Jews.
See also the discussion on Atzmon 2008.
Update Nov 21: a free copy of the paper is available here (PDF).
Update Nov 21: a free copy of the paper is available here (PDF).