STR vs SNP YDNA testing?

+6 votes
728 views
I'm curious about the difference between STR  &  SNP testing of YDNA... pros and cons.

STR (Short Tandem Repeats) is standard.  From what I can tell, it can provide only a rough approximate estimate for the age of ones common ancestor.

SNP (Single Nucleotide Polymorphism)  is more specific and can determine relationships with a much higher level of precision.  However, there must still be some uncertainty with it too.

SNP testing is extra and of course cost more.  However, it does not seem to be prohibitive.  Has the cost come down?  I have noticed that not many people have had SNP testing.
in Genealogy Help by Andrew Ross G2G6 Mach 3 (37.1k points)
retagged by Andrew Ross
I know you've used several tags already, but adding simply "dna" might bring your post to more people's attention.

Here are some practical points regarding haplogroups and haplotrees that I've gleaned from working with the Petty Project at FTDNA over the past year or so.

 

The first thing to recognize about haplogroups, is that

( (  (a) each of us has an essentially unique genetic family tree, represented as a tree-structured sequence of haplogroups – called a haplotree –  which is entirely analogous to a physical family tree.

Although a person’s haplotree consists of a fixed sequence of haplogroups,

b)    (b) it takes specific genetic (SNP) testing, of the person as well as a number of his cousins, to uncover and identify them all.

Rather than a new generation being formed simply as a result of a man having a son, as is the case with physical generations,

c)    (c) a new genetic generation is formed only when a mutation occurs in a man’s Y chromosome, and that mutation is then passed along to his son. 

Another difference between a physical family tree and a haplotree is that

d)    (d) a mutation does not necessarily occur in every “physical” generation; it can be quite variable, but on average, a new genetic generation is formed only about every 2 – 3 physical generations. 

This is somewhat of an oversimplification, but basically,

e)    (e) each mutation results in the formation of a new haplogroup.

It’s very important to recognize that

f)      (f) once a mutation occurs, and a new haplogroup is created, those changes continue to be passed down to every subsequent (physical) generation.

Finally,

g)    (g) the global male human haplotree is constantly growing and evolving, with new haplogroups being discovered every day, and because of this, downstream haplotrees are always subject to adjustment: unlike a grave marker, our haplotree is not etched in stone.

 

4 Answers

+9 votes
 
Best answer
Andrew, until someone comes along who can better answer your question, here is one reference that you might consult, Roberta Estes's January 2, 2020 post about Y-DNA.  She does discuss STRs and SNPs.

https://dna-explained.com/2020/01/02/y-dna-part-1-overview/
by Living Kelts G2G6 Pilot (555k points)
selected by Leake Little
And here is her Part 2:

https://dna-explained.com/2020/01/27/y-dna-part-2-the-dictionary-of-dna/

In my opinion, Roberta Estes is one of the best DNA bloggers.  Her explanations are always understandable by an average person without advanced scientific knowledge.
Thank-you Julie;

I noticed the following key line from Roberta's article:

"DNA haplogroups are quite reliability predicted by STR results...

Haplogroups are only confirmed or expanded from the estimate by SNP testing..."
Andrew, I can't respond intelligently right now.  Have been meaning for a long time to refresh my understanding of Y-DNA but haven't got around to it, and probably won't in the immediate future.  However, I have great confidence in Estes, and would not be hasty to conclude that she has been inconsistent (if that's what you're saying).
Thank-you Julie;

I wasn't trying to imply that her statement was inconsistent; rather she is making a good point:

STR testing predicts...

SNP testing confirms...
Your last statement is not a feature of STRs vs SNPs. It’s just the definition of haplogroup — haplogroups are by definition groups of people who share the same alleles at certain clusters of SNPs. So you will automatically confirm if you test all of your SNPs, but STRs by themselves won’t tell you for sure what your SNPs are.
Thanks, Barry.
Andrew, have you considered upgrading to the Big Y?  It isn't as expensive as it used to be.

Blaine Bettinger's 2016 book "The Family Tree Guide to DNA Testing and Genetic Genealogy" has a good chapter on Y-DNA testing (which I'm sorry to say I haven't taken the time to re-read this morning), and it a very good book in general for anyone who wants to work with DNA.  it is available on-line for a reasonable price.
+3 votes

Andrew,

you should first ask what questions you are trying to answer, but in either case you should be prepared for neither test to tell you much. This is basically because so few people do Y tests compared to autosomal testing.

If you want to test whether you and another person are descended through a specific male ancestor then both of you taking an STR test would either confirm or reject it with a very high degree of confidence - much better than autosomal tests. Both taking big-Y tests would be even better, but not cheap.

If you are just fishing for cousins then you should expect a low rate of return.

If you are simply curious about where your male line came from long ago then SNP testing will help.

And you can take it from me, because I am expert in doing Y testing the wrong way around. I don't regret it, because I am basically curious, but it definitely cost more than it could have.

STR testing is better for close generational differences, comparing to other testers within, say, 10 generations. It will give an estimate of haplogroup, but you should not expect that to be more precise than a few thousand years old.

Big-Y testing will give you an almost complete description of your SNP mutations (and many STRs) and possibly reveal one of two SNPs that were not even known to science.  My nearest tester to date shared a common ancestor about 600 years ago, and that person wasn't clan royalty, so there was no recorded tree and we are never going to know.

You might also be thinking about targeted SNP testing, (since you mentioned it was not very expensive). I would not bother with it, unless it has changed lately -  the panels that I chose only brought my haplogroup down to about a thousand years ago.

A final option, depending why you are testing in the first place, is the autosomal tests that include the Y-chromosome. One example is living-DNA, which gave me basically the same result as two panels of ftDNA SNP tests.

by Cameron Davidson G2G6 (7.7k points)
+3 votes
I'm not quite sure what question you are asking because SNP testing is available in two very different variants.

Autosomal DNA testing is a SNP test. This is what ancestry/MyHeritage/LivingDNA/23andMe all sell you as a DNA test. It will tell you about multiple types of relationships for recent generations - about 5 generations back though with a fair wind and a bit of luck it can go a little further.

There is another type of SNP testing - most often now called BigY - to the best of my knowledge it is only offerred by two companies, Family Tree DNA and ySeq.

That one will deliver very deep information but on your male line only - back to the year dot.

The BigY test will also give you all of the ySTR markers that you could ever wish for.

So if you specifically want to explore your Ross surname and connections to other Rosses, then either of the Y routes is the way to go. If you can afford the price tag, then the BigY test or ySeq's equivalent is a better buy in the long run compared with the STR tests, but in either case you may also want to look at the yFull service and sign up for that too.

In answer to your other question about price, the cost of all these tests has fallen over the years, but it is always worth looking out for things like father's day sales.
by Derrick Watson G2G6 Mach 4 (49.5k points)
+3 votes
Use your STRs to investigate. Use your SNPs (predicted or otherwise) to confirm. Test the SNP when you desire definitive proof. To test all SNPs would run into the 10's of thousands of dollars... Better to be more selective, which is why SNPs Packs are a better value than picking and choosing one here or there. Because the submitted sample is of a limited supply, it is usually recommended to upgrade to BigY700 to get the max result possible. In my case the L1065 Pack was haldf the price of BigY700. So I waited a month and saved and went for BigY700. I'm not really expecting much from the upgrade however other than confirming what i already suspect frommy research. I looked at it as more of an investment going forward for the next several years. fwiw
by Chris Campbell G2G6 (6.7k points)

Related questions

+6 votes
2 answers
568 views asked Oct 7, 2020 in Genealogy Help by Andrew Ross G2G6 Mach 3 (37.1k points)
+4 votes
4 answers
299 views asked Jul 18, 2020 in The Tree House by Alan Upritchard G2G6 (6.8k points)
+16 votes
2 answers
464 views asked Apr 7, 2015 in Policy and Style by William Pool G2G3 (3.3k points)
+27 votes
5 answers
448 views asked Feb 18 in The Tree House by Edison Williams G2G6 Pilot (451k points)
+6 votes
2 answers
287 views asked Aug 5, 2022 in Genealogy Help by Dsw Sayne G2G1 (1.1k points)
+6 votes
2 answers
472 views asked Nov 23, 2018 in Genealogy Help by Kevin Daniel G2G1 (1.6k points)
+9 votes
2 answers
+3 votes
1 answer
273 views asked Feb 11, 2020 in The Tree House by J. Salsbery G2G6 Mach 3 (32.4k points)

WikiTree  ~  About  ~  Help Help  ~  Search Person Search  ~  Surname:

disclaimer - terms - copyright

...