Ken Nordtvedt created a STR based tree for haplogroup I-L38. It is displayed at: https://http://knordtvedt.home.bresnan.net/
In her book "Ancestral Journeys - the peopling of Europe from the first venturers to the Vikings" (Thames & Hudson, 2013, p.149) Jean Manco relates haplogroup I-L460 (from which I-L38 is a branch) to the Neolithic Usatovo culture and the villages around the Dniester. According Manco this relation explains why I-L38 appears alongside R1a(11) after apparently migrating up the Dniester and around the Carpathians into present-day Germany.
This far 8members of the FTDNA I-L38 project shared their SNP results of the National Genographic's Geno 2.0 project to our project. Most of them shared their raw data file for comparision. More results of I-L38 Geno 2.0 participants should be compared to find out more about the structure of haplogroup I-L38.
This far the comparision of the more than 12.000 Ydna-SNPs can be summarized as follows:
PF112: Terry D. Robb interpretes PF112 as a mutation. Since Holmes belongs to the I-L38 "21-9" subgroup it would be interesting to know the PF112 result of other members of this subgroup. National Geographic does not interprete PF112 as a mutation since they describes the paternal haplogroup of Holmes as I-L38 and not as I-PF112.
PF3590: Ken Nordtvedt does not interprete PF3590 as a mutation because he thinks the mis-reading is due to a calibration change between the 12316 ysnps reports from Geno2 and the later 12062 ysnps reports. All haplogroup I customers got the same result (T,C) in the first batch of results, and all those in later batch got the same different result (T,T).
F780: One of the FTDNA I-L38 project members (Deterding/confirmed L38) took the Geno2.0 test. His paternal line is typed by the Genographic project as I-F780. Deterding had A,G for SNP F780. All other I-L38 members with known F780 23andme results tested G,G on this SNP. SNP F780 probably indicates the third branch of I-L38.
The 23andme results of two I-L38 project members: Prata and Marotta (both belonging to the "22-14" cluster), show that they share the same L69/S163/rs9786274 = T,T mutation. It seems that SNP L69 (after L533) determines the second branch of I-L38.
At the YDNA SNPS Comparison Project (on Facebook: https://www.facebook.com/pages/YDNA-SNPs-Comparison-Project/166307390055000) the (23andme) SNPs of 10 I-L38 samples can be found. One of these samples, named Marotta, tested L69/S163/rs9786274 positive.
Pilgrim Captain Myles Standish (http://en.wikipedia.org/wiki/Myles_Standish) turns out to be a member of haplogroup I-L38 (http://www.familytreedna.com/public/mayflowersociety/default.aspx?section=ycolorized).
A frequency map of I-L38 in Europe is displayed at: https://sites.google.com/site/haplogroupil38/i-l38-distribution-map
During the years 2005 and 2006 approximately 2000 archaeological finds ranging from the Neolithic Period to Late Antiquity were found on the Kletthamer Feld (Erding, Upper Bavaria). Out of this context a burial site was examined comprising 13 individuals, some of them rich in precious grave goods. The inhumations were dated to the second half of the 4th to the first half of the 5th century – a time of upheavals in relation to the demographic structure of the former Roman province Raetia (today southern Bavaria).
Inserting the 9 displayed STR values of samples 1664 and 1704 in Cullen's Haplogroup I Predictor, leads to the following haplogroup estimations: Ix-S23 =>25% I-S24 =>25% J2-M102 =>22% I-S31* =>14% I-M253 =>9% I-P37.2 =>3% G , G2 =>2%
Ix-S23 is the name Cullen's predictor uses for I-L38. Although the DYS393 (=16) value of the samples 1664 and 1704 is very atypical for I-L38, it might be possible that these samples belong to haplogroup I-L38 (which would not be that surprising since Bavaria is not that far from the Upper Rhine region).
In february 2012 Terry D. Robb presented frequency maps for several I2 haplogroups, among them “I2.0110*” (which is his code name for I-L38).
According to Terry, France and Switzerland come out the dominant (highest frequency) places. But also England, and even Norway, still have a significant frequency of I-L38.
This frequency map differs from the frequency map made by Hans De Beule in 2010: http://sites.google.com/site/haplogroupil38/i-l38-distribution-map
According to Hans, I-L38 mainly spread in an area that can be described as a diagonal “belt” going from the NW of Europe to the SE. In his view, the highest concentrations (and STR diversity) are to be found in the Upper and Middle Rhine area.
Terry agrees that finer geographic binning will conclude that certain parts of France have higher frequencies than others. Also on Hans’ map France, (and especially the NE) has a significant I-L38 frequency.
Overall conclusion is that more (and especially French) samples are needed to finetune the current frequency maps.
In 2012 it became clear that SNP L533 separates I-L38 in a modest way. Until now only few I-L38 families turned out to be L533 positive. Members with DYS385a,b=13, 14 AND DYS454=11 are encouraged to test for L533 since this combination looks like the "STR-signature" of I-L533.
So far for the results returned for L38,L39, and L40 there has been no separation with members testing positive for all three SNP's.