The German site also had either haplo F or H - undetermined.
The F could also be IJ, which wasn't tested.
And the German site had an entirely different branch of G than the Iceman had.
This is to be expected considering the large diversity of G subclades found in the Levant.
In believe an earlier study of Neolithic farming skeletions found haplogroup N.
I am not aware of anything of the sort, except for mtDNA N. You may want to check my summary of all the ancient Y-DNA and mtDNA studies to date (I still have to add the sources for the Bronze and Iron ages).
The authors of the Spanish article published today refer to "the probable high level of heterogeneity of the Neolithic dissemination in Europe," and indicate that their study was inadequate to evaluate the arrival of the men in the Neolithic transition. (Lacan et al., Ancient DNA Suggests the Leading Role Played by Men in the Neolithic Dissemination, PNAS Early Edition)
Their opinion on the subject is not more informed than ours. In fact they may be more amateurish since they stated that the mtDNA lineages found were typically pre-Neolithic, when only 2 out 7 samples were (U5 and H3).
In addition, agriculture was already widespread over all southwest Asia, not just the Levant by at least several millennia before the non-Spanish men mentioned above which greater increases the possibilities for their origins. Because the G subgroups involved here may have been in existence for multiple millennia before before the movement to Europe, leaving many opportunities for intermixing with other haplogroups.
This is all the more surprising to find such homogeneity of G2a in Late Neolithic sites as well.
You also speculate on the origins of haplo G in Egypt. This is one country where G origins have been a black page due to the scarcity of detailed haplotype samples. There are only two Egyptian samples in the Haplogroup G Project. I found one in the YHRD database, and El-Sibai, 2009, Annals of H Genetics) had 8 G haplotypes. These samples seem to have bits and pieces of known G subgroups. There is only one DYS390=21, typically a sign of G2a3a so common in Greece and Turkey. There is no DYS392=10 found in all G2a1a men and ultra common in the mid-Caucasus Mtns. There are no men with DYS392=12, found in all G1 persons, most common in Iran & its western borders. There were no men with a double value for DYS19, found in half of G2a4 men (such as the Iceman) There were no men with the very low DYS385a values seen in European G2a3b1a1a men. There is one man with the DYS388=13 value seen in the large G2a3b1a2 men of Europe. The rest are miscellaneous patterns that could represent paleolithic arrivals, but one cannot ignore the isolated G2a3a & G2a3b1a2 samples which suggest a subsequent arrival. So one can argue anything one wants with regard to the Egyptian samples.
That's very interesting. Judging from this, these G2a3a and G2a3b1a2 could very well be of Greco-Roman or Anatolian (Hittite ?) origin, since Egypt was conquered and occupied by all of these.
With regard to your comments that haplo G is more common in the northeast of Spain, this seems in conflict with your haplogroup G map which shows a relatively low percentage there. In the Adams Iberian study, they found an average amount of G in that area. In the samples in the YHRD database for Barcelona, by my rough estimate, these also have an average amount of G, but the samples from the Pyrenees have a below average amount.
Sorry, typo error. I meant Northwest (Cantabria, Asturias, Northwest Castille, North Portugal).