Has anyone done visual phasing for DNA yet?

+11 votes
364 views
I am interested in trying this out.  I have tests for myself and three of my four siblings. I also have tests for my mother, a maternal uncle, a maternal first cousin once removed and two paternal first cousins once removed.

Do I have enough tests to get started? Would it be helpful to also get a maternal and a paternal first cousin or two before starting?
in Genealogy Help by Karen Raichle G2G6 Mach 7 (73.5k points)
Thought about it.
With the number of siblings that have taken tests, you have more than enough to start the analysis. The cousin matches are only necessary if you don't have the three sibling minimum.

As mentioned below, I have just got my third sample, so will be doing this in the New Year.
They need to be first cousins of the parents,  not first cousins of the siblings, to differentiate between grandparents.
So Simon you are saying I don't really need the cousin matches?

Just having another quick read, it would seem to me that you have a perfect combination with a brother, yourself and your sisters. The cousin matchings come in handy when doing comparisons to commonly shared matches, but don't appear relevant with "Visual Phasing".

A quick snip from the article....

Accordingly, the first step in visual phasing is to compare the DNA of the three siblings to each other in GEDmatch using the One-to-One tool. We’re going to work on one chromosome at a time, and I recommend starting with the X chromosome (especially if one of the siblings is male, since he’ll only have one X chromosome) or one of the shorter chromosomes such as 20 through 22.

2 Answers

+5 votes
 
Best answer
G'day Karen

To answer your Questions:-  >Do I have enough tests to get started?
Yes - minimum required 3 siblings.

What is happening is that the 3 siblings DNA is PHASED to parent and MAPPED to Grandparent.

So a mother (or either parent) tested is a great help and timesaver because GEDmatch can do the phasing to parent for you.

I used the mother to phase the 3 siblings (on Gedmatch) to their father's DNA and also run 3 one-to-one comparisons on the siblings phased paternal DNA as this makes working out which crossover points are maternal or paternal very easy. I found the GEDmatch phased result to the opposite (missing) parent most reliable and for an unknown reason the phased result of the phasing (present) parent unreliable.

>Would it be helpful to also get a maternal and a paternal first cousin or two before starting?

Getting a test each of one first Cousin of the mother and one of the father of the 3 siblings is a significant advantage and time saver. (NOTE - not a first cousin of the siblings but of their parent!) Even better (if they are still alive) is a grand uncle or grand aunt of the siblings.   These are used to determine which grandparent segments belong to.    It can not be a full first cousin of the 3 siblings nor their aunt or uncle as they are too close and don't differentiate between grandparents.

You generally only need one matching segment (for each of paternal side and maternal sides) on each chromosome to determine the map of both grandparents of that parental side.  Extra matches however will help confirm (or discredit) that you have got the crossover points correct.

 Have you enough to start - yes it appears likely but your language is a little vague in describing the relationships of tested relatives so I'm unsure if they fall into the "too close" category. But if you have further matches with more distant "known relationship" relatives you can work with their matching segments but there will be fewer of them.    In one Chromosome I was actually able to differentiate between maternal grandparents because of a DNA match with the mother that none of her children shared - as the siblings all share the DNA of the other maternal grandparent at that location.

My experience - during the last two months I have visually phased to grandparents, one lot of 3 siblings and now I am working on second batch of three siblings (4th cousins that are related to me by my father's mother).  

I have also bought auDNA kits to work on 3 siblings (4th cousins on my mother's Mother's side)  - which once completed will reveal most of their mother's DNA which will then complete 3 siblings on the older generation (the other 2 having already been tested.)  I'm really excited about the prospect of attempting to visually phase a sibling whose DNA was created through the same process.

While the instructions for  visually phasing are for doing it on a single chromosome (and yes each Chromosome has to been done as a single entity and some will phase easily and other's more difficult and might not even complete.)  I have been trying to create an assembly line process for the 22 chromosomes (so that it's done in assembly line steps for each of the 22 chromosomes then move onto the next step, etc.)

The instructions given in the link provided by Simon is pretty straight forward for beginners but I'm willing to help out with any problematic chromosomes - some of the GEDmatch one-to-one comparison results sometimes need a little 'reinterpreting' (or rather 'further interpretation') even when the original auDNA tests have all been with the same company.

A word of caution - one should be aware that some DNA matches may be related to multiple grandparents or if grandparents are related to each other (as in my case) one has to be careful that the known relatives used to determine the grandparents are only ones that are related via one grandparent.  

An insight bonus - After mapping I have been able to demonstrate that a known relative most closely related on the maternal Grandmother's side has matching DNA segments at different ends of the same chomosome via the different related grandparents.   Without this technique it would have been impossible to identify which grandparent the match was through and the natural assumption is that the matching segments would be via the closest relationship.

Best of luck on the adventure.

It is worth the effort.

Veni
by Veni Joyner G2G6 Mach 1 (18.5k points)
selected by Derrick Watson

Veni, Thanks so much for the detailed answer.

 Have you enough to start - yes it appears likely but your language is a little vague in describing the relationships of tested relatives so I'm unsure if they fall into the "too close" category. But if you have further matches with more distant "known relationship" relatives you can work with their matching segments but there will be fewer of them.    

I guess my details were not as clear as I thought.  The relationships I was describing were of the test subjects to me.

I have the 4 siblings (myself, my brother and two sisters), our mother and her brother, as well as their (mother and uncle's) maternal first cousin. 

A word of caution - one should be aware that some DNA matches may be related to multiple grandparents or if grandparents are related to each other (as in my case) one has to be careful that the known relatives used to determine the grandparents are only ones that are related via one grandparent.  

My father is deceased but, I have two of his first cousins. I think this might be where the problem lies.  These are my father's "double first cousins". Though as far as I can tell these are the only two of his generation still living. Since these cousins are related to both of my paternal grandparents, it would seem that I need to find someone who is only related to one grandparent. Am I understanding this correctly?

Thanks as well for the offer to help.  I just might have to take you up on that!

Karen

 

 

G'day Karen

I took a look at your wiki profile and thought for a few days that you have a grandfather tested - IF YOU DID then you can GEDmatch phase your mother to him and what's left will be the grandmother's DNA   - the result will (should be) be the same for doing the 3 siblings but much easier... 5 minutes of GEDmatch time verses maybe a week or more to do the 3 siblings.... you can do one-to-one comparisons on the between grandfather and each sibling to find their Grandfather matching segments with him and do one-to-one comparisons with your mother's maternal phased DNA and each sibling to find their maternal grandmother matching segments.   

This means you only have to visually map the paternal line by phasing each sibling with the mother to get their father's phased DNA and then do the visual phasing method on those 3 Paternal phased kits - it will be loads quicker.

But you don't ... so you have to do it the long way.... sorry my mistake - case of mistaken identity.
by the way... our family has an instance of double first cousins also.... two brothers married two sisters.
By the way I recommend that you do the final version of the mapping on dnapainter.com - it works so well
I've sent you by private message a link to an image of your 22 chromosomes mapped to your Uncle.
Thanks Veni, I sent you am email.
I haven't received any email yet... 5hrs
That's weird.  Ok, I just sent you  another one.
+2 votes
Karen, I confess that I am not familiar with the term visual phasing, or exactly what you mean by the generic term “test”. I suspect that you are talking about one specific type of DNA, autosomal. If so, and if you wanted to identify whether a DNA match was related to you in eithercyour father or mothers side, the yes, you would need atDNA from a paternal relation. Likewise if say you had a small trace of Ashkenazim or something else and  wanted to identify what side that came from.

I must say that I believe mtDNA and Y-DNA can give you more than just atDNA alone. While my atDNA makes me 99.9% Western European, without mtDNA I would not have known that my direct female line actually originated amongst the Betbers of North Africa. And without Y-DNA I could only have traced a few generations of my 1,000 year old clan from Ireland where paper records are almost non-existent.but our Y-DNA still tells who we are and how far back we were related.
by Rory Cain G2G6 (9.8k points)
The Visual Phasing that Karen is discussing is explained at https://thegeneticgenealogist.com/2016/11/21/visual-phasing-an-example-part-1-of-5/ and is related to autosomal testing.

You need a minimum of three siblings to have taken an autosomal test and this can give you the ability to determine which segments have been inherited from each of your Grandparents. This can make it substantially easier to determine which line your auDNA matches are on.

For my part, I have just got a sample from my sister so that I can undertake this analysis as well.
Yes, that is the article I have been studying.  It seems a little complicated to me but hopefully I can make some progress with it. I wish you luck in your analysis as well.
Simon and Karen - how is your phasing going?

Related questions

+6 votes
1 answer
122 views asked Jan 15, 2018 in The Tree House by Emma MacBeath G2G6 Pilot (718k points)
+6 votes
2 answers
279 views asked Nov 12, 2017 in Genealogy Help by Barry Smith G2G6 Pilot (181k points)
+5 votes
3 answers
437 views asked Feb 23, 2018 in The Tree House by Pete Toemmes G2G4 (4.8k points)
+13 votes
2 answers
469 views asked Aug 5, 2016 in The Tree House by Andreas West G2G6 Mach 5 (57.2k points)
+1 vote
1 answer
44 views asked Mar 21, 2019 in The Tree House by Milton Davis G2G6 Mach 2 (25.9k points)
+6 votes
1 answer
+49 votes
85 answers
+2 votes
1 answer
82 views asked Aug 8, 2014 in Genealogy Help by Viola Jones G2G1 (1.5k points)

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

disclaimer - terms - copyright

...