Prelude:

To understand the Pressburg ancestry a little more, initially we proceeded via the Schoenberg Barrow DNA segment matches. This led us identify which of the matching segments that Julian, Ed and John have with Randy and his sister have their origin with Randy's mother or father - that is, which matches with Randy and his sister are complemented by matches with Randy's parents:

We note that Randy's parents have a significant total match of 128.4cM with each other, and thus they are related to each other. Above, Randy's parents make independent contributions by way of the segment matches of Julian, Ed and John. The Cr17 match is of course interesting; there is a likelihood of false positives of course, as this overlap has not been subjected to Q-match. We have explored the Pressburg ancestry of Scott's wife and MWaas (see here) but no obvious pattern emerges.

Schoenberg/Lousada matches

We explore the Schoenberg matches at the 4 Lousada indicator sites with Q-match on GEDmatch set at 3cM segment size and initially Precision=7 then Precision = 3, and observe that:

1. Chromosome 21

Some matches were not shown above (because Randy's parents were not present). Thus, at chromosome 21 we first found Randy's matches with Julian, John and Ed (noted here) as elucidated here. These matches acquired greater significance since the 21st chromosome segment 36,873,163 - 37,833,931 appears to be a 4th Lousada indicator site. To clarify a somewhat confusing situation around this Cr21 segment a further set of comparisons was run as shown here. From this it can be clearly seen that Randy's 4 or perhaps 5 matches are due to a co-operation between his parents; that is, Randy's matching segments are not explicable without segments having been received from each parent. In note 5 below we go on to suggest that the contribution of Randy's father is not a multiple statistical fluke but arises because Randy's father also has Baruch Lousada ancestry as has Randy's mother (as we suggest in note 4 below)! A further observation about use of the Cr21 Lousada indicator site may be found in note 5 below.

2. Chromosome 8

At this Lousada indicator, as shown here, Randy's mother matches Scotts wife (P=7). Randy matches TPhelan with no parental match though at P=3 his mother's match emerges - a peculiar situation like we saw in note 1 above where the parents' genes seem to cooperate to produce a match. Randy's sister at P=3 (matching with Ed) shows the same peculiar matching, and additionally (matching Allan with no parental match) both siblings together share a peculiar match. A further observation about use of the Cr8 Lousada indicator site may be found in note 5 below.

3. Chromosome 5

There are no matches as shown here. With no matches with the control sample either, it seems that the Cr5 Lousada indicator can be suspended from use for the moment.

4. Chromosome 2

At the Cr2 Lousada indicator, using Q-match set at 3cM, with Precision = 7, we find matches not previously evident (see here). GEDmatch continually tweaks its methodology for small matches, and at the time we first generated Schoenberg/Baruch Lousada match data, chromosome 2 matches were not detected. Now, matches at this site appear between both Randy and Randy's mother with Julian, Jeannine and the Papacin-Lousada descendant TPhelan and are hard to ignore. (The match involving Randy's father together with Randy and his mother is discussed in note 5 below). The 3 matches suggest genealogical enquiry will be worthwhile to see whether Randy's mother has Baruch Lousada ancestry. Of course, if she does have Baruch Lousada ancestry, the obvious source would be the wife of Baruch - perhaps as shown here.

We show here further details of our segment matches at the Cr2 indicator site (ie around the 219m location). The weight of this segment data, especially the astonishing segment boundary coincidences, encourages us.

 5. Observations

Naturally, in the light of the Cr8 and Cr21 complications discussed in notes 1, 2 and 4 above, we compared indications of Randy's parents' Baruch Lousada ancestry using Q-match as shown here. Some control matching was run as shown here. As will be seen, we include all our Baruch Lousada sample (not just those descending from Baruch). We show that the Schoenberg matches extend to the non-Baruch Baruch Lousadas with an increase in matches as we go from P=7 to P=3 and covering 3 of the 4 Baruch Lousada indicator sites. While this supports our suggestion in note 4 that Randy's mother has Baruch Lousada ancestry, we can go beyond that because it is possible that Randy's father also has Baruch Lousada ancestry. This could explain the mysterious cooperation between the genes of both parents on Cr8 (3 times; see note 2) and on Cr21 (4 or 5 times; see note 1). Then, at Cr2, as we saw in note 4 above, Randy's match with TPhelan coincides with that of both parents. Though a match between Randy's parents at the Cr2 site does not appear at Precision = 7, it emerges at Precision = 3.

From all this, some observations about the 4 Lousada indicator sites can be made. As suggested in note 3 above, it is hard to see what use the chromosome 5 indicator can be - with neither Schoenbergs nor controls showing a single match. Thus we turn our attention to the chromosome 8 and 21 indicators. The controls show matching here at Precision = 7 and more at Precision = 3. The chromosome 8 indicator appears often, sometimes with the chromosome 21 indicator, for 3 people who are not Baruch Lousadas but no chromosome 2 indicator matches arise. The obvious conclusion is that the chromosome 8 and 21 indicator matches only apply when in conjunction with a chromosome 2 indicator match. It remains to be seen whether they can appear separately in this context, or only together as a trio of coincident indicators!

Though we have been operating at 3cM, Q-match removes most of the false positive problem, but there is still a chance of what the genetic genealogists call IBS (Identical By State) or pre-genealogical matches. IBD (Identical By Descent) is what we are seeking. And of course, with common ancestors being so far back, we lose many IBD matches. In addition we are plagued by what we called above ''mysterious cooperation'' between chromosome pairs, and what some genetic genealogists call pseudo-matches. Accordingly we have been unable to arrive at a rigorously structured account of the segment matches - for if we were able to completely detect and remove the pseudo-matches, we could presumably demonstrate absolutely the reality of matches within our set of distant Baruch Lousada relatives. Of course pseudo-matches are not always antithetical to our thesis - for as will be seen from note 1, 2 and 4 above, some of Randy Schoenberg's Baruch Lousada matches emerge both directly from a parent and indirectly via contributions from both parents.

Despite the reservations our data retains plausibility, especially because of the astonishing segment boundary coincidences at the Cr2 indicator site.