PDA

View Full Version : When is an SNP not an SNP?



zamyatin13
10-05-2015, 09:24 PM
Not sure if the 'General' forum is the right place for this, but it seemed too broad a question for any particular sub-forum so here goes.

Have been perusing the new 'Great Tree' diagram produced by BritainsDNA, and naturally looked first at my own Y-DNA, see below:

6193

I'm the S3689 branch, traditionally known as L-21/DF63/CTS6919. However, in their diagram, there is no DF63 (otherwise known as S522 I believe).

They confirmed this in an email, stating that they...


...can confirm that your, and all S3689+ samples we have seen to date are coming out as S522- on chromo2. This indicates that the S522 SNP is not working, and we have now removed it from our list of working Y SNPs

This confuses me though. Are they saying that S522 doesn't exist upstream? In which case why are the other testing houses stating that it does exist? Who is correct? How can an SNP that seems reliable in one test, not be in another?

Sorry for the basic question. Am I misunderstanding the concept of SNPs?

J1 DYS388=13
10-06-2015, 04:57 AM
I've never heard of a "working Y SNP." It must mean that what they have set up on the Chromo2 chip isn't working.

Can you e-mail me a copy of the whole tree? I couldn't access the link ScotlandsDNA sent me, and now the link is gone. Thanks in advance. jimhoneychuck @ yahoo.co.uk

VinceT
10-06-2015, 09:33 AM
Right, they mean "working on the Chromo2 chip". The chip can only detect a known ancestral or derived state using probes that bind directly adjacent to either side the SNP position. If the probes do not bind at all, a no-call is signaled. If the probes bind to multiple locations, an ambiguous call is signaled. If the probes bind to the incorrect position, either a no-call, an ambiguous call, or an erroneous call is signaled, but they do not know if it is actually erroneous, or the erroneous call is a false positive or a false negative, without knowing what call was to be expected at that position for any given sample.