Do you have adblock enabled?
 
If you can read this, either the style sheet didn't load or you have an older browser that doesn't support style sheets. Try clearing your browser cache and refreshing the page.

(Fark)   Are there any genetic scientists online today? Given the ubiquity of genetic testing services available now. Is there a way for someone looking at an anonymous DNA sample to tell if a person is the result of incest?   ( fark.com) divider line
    More: Unlikely, Cascading Style Sheets, Politics Ivanka Trump, Business Teen Vogue, sexual predator father, nearly naked bottle, Geek Tim Cook, Ivanka Trump, HTTP cookie  
•       •       •

142 clicks; posted to Discussion » on 03 Nov 2017 at 2:20 PM (2 weeks ago)   |   Favorite    |   share:  Share on Twitter share via Email Share on Facebook   more»



37 Comments     (+0 »)
 
View Voting Results: Smartest and Funniest
 
2017-11-03 11:50:51 AM  
Not a scientist...but it's the whole "Anonymity" thing that's a huge problem with DNA testing.  I think you'd also have to have a sample from the other party.

This may be a bit dated, but it may be worth a try.
https://venturebeat.com/2013/09/20/ho​w​-to-use-23andme-without-giving-up-your​-genetic-privacy/
 
2017-11-03 11:55:09 AM  
We're all related.
 
2017-11-03 12:05:21 PM  
you'd have to have comparison samples, just like for paternity tests
 
2017-11-03 12:06:24 PM  
Not a geneticist, but it's theoretically possible to say that it was highly likely incest.

So, as most people know, you get set of chromosomes from each parent.  That means that for any two children (let's call them S for Son and D for Daughter), each has half of each parent's genes.  Other than the sex chromosomes (in which S has dad's Y and one of mom's X, but D has dad's X and one of mom's X), which gene each has out of the two from their parent is random.  So, in general, there's a 1 in 4 chance that any particular gene from parents is shared between two siblings.

That means there's a 1/16 chance for both kids to have the same gene from mom and the same gene from dad.  With 23 chromosome pairs, there's a pretty good likelihood that at least one pair will match like this for any son/daughter combination.  That means there's (at least) a 1/2 chance that any offspring from the two will get at least one set of identical genes from both parents - which is why genetic abnormalities show up more easily in cases of incest.

If you want to look at parent/child incest, then it's still about 1 in 4 for each pair - meaning only a 1 in 747 chance (3/4 to the 23rd power) that such an identical matching *doesn't* occur.

Now, if there's no pair with matching genes, it's almost impossible to tell.  But if a set matches, it's pretty likely (there's a very low chance of some random stranger having an identical gene, not just due to distribution but genetic drift).
 
2017-11-03 12:09:36 PM  
Is the DNA a single helix?
 
2017-11-03 12:13:02 PM  
Looking at just a single DNA sample? No, not really. Maybe the high degree of homozygosity across the genome might suggest there was something funky going on (especially if you found a bunch of rare homozygous variants), but without anything to compare it to you can't know with any certainty.

Comparing to parents' DNA? Quite possibly, although depending on the set of markers used, you might not have complete confidence especially if the individuals have come from a population isolate (for example, Iceland). That being said, you'd likely need to look at a set of markers: a single rare homozygous variant doesn't tell you much in that regard, as it potentially could have arisen independently multiple times.

But without knowing the pedigree, I don't think you can't say with 100% certainty that a person is the product of incest just by looking at DNA sequences.

/geneticist
 
2017-11-03 12:15:26 PM  
... or listen to no1curr, since they know what they're talking about :)
 
2017-11-03 12:17:26 PM  
Yes.

By the way, your sister called... something about today being your turn to babysit.
 
2017-11-03 12:21:00 PM  

obenchainr: ... or listen to no1curr, since they know what they're talking about :)


I liked your explanation though :)
 
2017-11-03 01:25:09 PM  

kevlar51: Is the DNA a single helix?


Is the sample from Alabama?
 
2017-11-03 01:37:03 PM  

kevlar51: Is the DNA a single helix?


Isn't that RNA?
 
2017-11-03 01:39:28 PM  
If there was, all of West Virginia would be a good place to research.
 
2017-11-03 01:48:07 PM  

Marcus Aurelius: We're all related.


According to the Ancestry family tree I developed, my wife and I are cousins, 29X removed.  We decided not to have children, just to be safe.  That and I'm a lawyer and she thought it best to end that genetic error for the sake of the species.
 
2017-11-03 01:48:48 PM  

Marcus Aurelius: We're all related.


Then why don't I get the Family Discount everywhere?
 
2017-11-03 02:28:22 PM  
I saw a documentary about this on Pornhub.
 
2017-11-03 02:29:58 PM  
One easy delimiter is birthplace. Anything under 36 degrees 30 minutes is suspect for incest.
 
2017-11-03 02:31:18 PM  
Asking for a friend, Morty Subby?

vignette.wikia.nocookie.netView Full Size
 
2017-11-03 02:38:01 PM  
I was glad my small rural town was close enough to a military base. Plenty of new blood coming through on a regular basis.

Avoid the South Mountain, though. https://en.wikipedia.org/wiki/Goler_c​l​an
 
2017-11-03 02:39:05 PM  
Ph.D.  Medical Geneticist here.

No.  Not from a single individual's sample.  23andMe and these other companies use a DNA chip that looks at specific genetic loci, and doesn't sequence the entire genome.  And depending on the loci they look at, there may be a high degree of homozygosity at that loci in the general population  (admittedly, the heavy duty math of population genetics puts me to sleep).  But it would require genetic testing of the parents to confirm such a question and by simply taking the sample from purported parents, you'd have answered it.  (Unless paternity is in question, in which case standard paternity testing would also answer it, but would require testing multiple individuals.)
 
2017-11-03 02:39:15 PM  

no1curr: Looking at just a single DNA sample? No, not really. Maybe the high degree of homozygosity across the genome might suggest there was something funky going on (especially if you found a bunch of rare homozygous variants), but without anything to compare it to you can't know with any certainty.

Comparing to parents' DNA? Quite possibly, although depending on the set of markers used, you might not have complete confidence especially if the individuals have come from a population isolate (for example, Iceland). That being said, you'd likely need to look at a set of markers: a single rare homozygous variant doesn't tell you much in that regard, as it potentially could have arisen independently multiple times.

But without knowing the pedigree, I don't think you can't say with 100% certainty that a person is the product of incest just by looking at DNA sequences.

/geneticist


There are some human populations that are more homozygous than the average 21st century American, European, Chinese, etc - I'm thinking the Amish, some Ashkenazik Jews, isolated populations in places like Australia or Micronesia, right?  That term you used 'population isolate' in reference to Iceland - is that a geographical/historical description, or can you measure that statistically based on some indicator of homozygosity?
 
2017-11-03 02:40:52 PM  

Marcus Aurelius: We're all related.


^^True.^^

And as your 278,415th cousin (twice removed) I feel obligated to say you have been a huge disappointment to Aunt Dorothy.

/Marcus is the reason she drinks...
 
2017-11-03 02:53:01 PM  

Mad Scientist: Ph.D.  Medical Geneticist here.

No.  Not from a single individual's sample.  23andMe and these other companies use a DNA chip that looks at specific genetic loci, and doesn't sequence the entire genome.  And depending on the loci they look at, there may be a high degree of homozygosity at that loci in the general population  (admittedly, the heavy duty math of population genetics puts me to sleep).  But it would require genetic testing of the parents to confirm such a question and by simply taking the sample from purported parents, you'd have answered it.  (Unless paternity is in question, in which case standard paternity testing would also answer it, but would require testing multiple individuals.)


Full sequencing could let you find the increased homozygosity and get an estimate of how likely the individual is to be the product of incest, but not very reliably.

Linkage disequilibrium might be a better tool. Identifying long runs of homozygosity suggest the chromosomes are recently identical by descent. This has been used to assess inbreeding in populations https://www.ncbi.nlm.nih.gov/pmc/arti​c​les/PMC3415543/ , but I don't know how sensitive it would be in an individual.
 
2017-11-03 02:57:58 PM  

wejash: Marcus Aurelius: We're all related.

According to the Ancestry family tree I developed, my wife and I are cousins, 29X removed.  We decided not to have children, just to be safe.  That and I'm a lawyer and she thought it best to end that genetic error for the sake of the species.


Good call. It's tough having a kid when one of you is already several centuries old.
 
2017-11-03 03:01:34 PM  
We all sprang from a ridiculously small family of people in Africa. They were rammblers and spread out boning each other and eveen neanderthals when one was slower than our ancestors. European humans almo0st didnt make it and it came down to maybe 25 survivers that got busy boning again . If it werent for their version of :Your Mom" honkies would have gone extinct
 
2017-11-03 03:03:30 PM  
Just send it to 23 and me and they will tell you if your mother/sister is your real mom
 
2017-11-03 03:56:29 PM  

basemetal: If there was, all of West Virginia would be a good place to research.


In the Southern USA we call them inbreed Hillbillies.
In England you call them the Royal Family.
 
2017-11-03 04:29:27 PM  

mikaloyd: We all sprang from a ridiculously small family of people in Africa. They were rammblers and spread out boning each other and eveen neanderthals when one was slower than our ancestors. European humans almo0st didnt make it and it came down to maybe 25 survivers that got busy boning again . If it werent for their version of :Your Mom" honkies would have gone extinct


Sometimes you don't even need the genetic testing. You can just look at typos and punctuation errors to determine where the genetic abnormalities lay.

/I keed!
 
2017-11-03 04:32:22 PM  

FrancoFile: There are some human populations that are more homozygous than the average 21st century American, European, Chinese, etc - I'm thinking the Amish, some Ashkenazik Jews, isolated populations in places like Australia or Micronesia


How about dem Habsburgh?
http://www.nature.com/news/inbred-roy​a​ls-show-traces-of-natural-selection-1.​12837
In a 2009 paper2, Ceballos and Álvarez showed that the extent of inbreeding increased over the reign of the Habsburgs, so that Charles II was more inbred than is typical of a child of a brother and sister.

img.fark.netView Full Size

 (and this was painted by someone trying to make Charles look good)
 
2017-11-03 06:29:20 PM  

wejash: my wife and I are cousins, 29X removed.


Geez, one of you is REALLY old.

Seriously though, assuming you really meant "29th cousins", there probably isn't a person in the world you aren't at least that closely related to.

Really though, you're probably full of shiat. Most people can't reliably trace back that many generations unless they get on a royalty train.
 
2017-11-03 06:45:15 PM  

Hollie Maea: wejash: my wife and I are cousins, 29X removed.

Geez, one of you is REALLY old.

Seriously though, assuming you really meant "29th cousins", there probably isn't a person in the world you aren't at least that closely related to.

Really though, you're probably full of shiat. Most people can't reliably trace back that many generations unless they get on a royalty train.


If you have English, Scandinavian, or French Canadian ancestry, you are no more than 10th cousins with close to 40 million other Americans, and probably closer to eighth (once removed, or so). I'm ninth cousins or slightly closer with my wife, Madonna, Celine Dion, and the Bush twins, and Obama.
 
2017-11-03 07:13:13 PM  

Unbridled Apathy: Hollie Maea: wejash: my wife and I are cousins, 29X removed.

Geez, one of you is REALLY old.

Seriously though, assuming you really meant "29th cousins", there probably isn't a person in the world you aren't at least that closely related to.

Really though, you're probably full of shiat. Most people can't reliably trace back that many generations unless they get on a royalty train.

If you have English, Scandinavian, or French Canadian ancestry, you are no more than 10th cousins with close to 40 million other Americans, and probably closer to eighth (once removed, or so). I'm ninth cousins or slightly closer with my wife, Madonna, Celine Dion, and the Bush twins, and Obama.


Just doing the math, you would have about a half a billion ancestors in the 29th generation. That's assuming of course that there aren't any loops in your tree, which wouldn't be the case. But still, that's more people than lived on Earth at that time. And that number doubles with every generation, so 29 generations might not quite be the level at which you are related to everyone, but it's probably within a couple generations of it.
 
2017-11-03 07:59:07 PM  
Human geneticist here...Actually yes, from a single DNA sample you can tell is a person is a product of incest.  If you do a test called a SNP array, you can identify regions of homozygosity across the genome and depending on the number and length of these you can determine the degrees of relatedness of the parents.  Without other samples you may not be able to identify an individual (e.g.  if the degree of relationship is consistent with a sibling and the girl has several brothers you couldn't pin it on a specific bother without a  samples from candidate relatives) but if was father-daughter you could tell that close a relationship.  At children's hospitals that perform SNP assays on newborns with congenital abnormalities if consanguinity is identifies there is mandatory reporting.
 
2017-11-03 08:04:30 PM  

jengen: Human geneticist here...Actually yes, from a single DNA sample you can tell is a person is a product of incest.  If you do a test called a SNP array, you can identify regions of homozygosity across the genome and depending on the number and length of these you can determine the degrees of relatedness of the parents.  Without other samples you may not be able to identify an individual (e.g.  if the degree of relationship is consistent with a sibling and the girl has several brothers you couldn't pin it on a specific bother without a  samples from candidate relatives) but if was father-daughter you could tell that close a relationship.  At children's hospitals that perform SNP assays on newborns with congenital abnormalities if consanguinity is identifies there is mandatory reporting.


I assume this is much easier with XX than XY?
 
2017-11-03 08:20:05 PM  
It doesn't matter, because most of the genome is autosomes (chromosomes 1-22).  There is a nice paper that describes the types of information you can tell from array data PMID:22118739, in case you're interested in more detail.
 
2017-11-04 04:14:08 PM  
It is possible to make a rough estimate of relationship by the amount of autosomal DNA shared, but it is not possible to determine exact paternity. For example, those descendants of President Jefferson might actually be descendants of his brothers or paternal uncles. His DNA is not so specifically marked that you can tell Tom from Dick or Harry.

You can eliminate the possibility of incest if the Y chromosome does not match (a non-paternity event) or the mitochondrial DNA is not that of the Mother's (a more unusual case in which your Mother is not your real Mother or even an aunt, cousin or grandmother). Otherwise, heredity is too random to be 100% certain in all cases. There are normally ancestors from whom you do not inherit DNA, and even your siblings share only about 25% of your DNA, as do your grandparents.

But autosomal DNA can only give you probabilities. In the case of incest you have fewer grandparents than you ought to. Most of us have four, two of each sex. In a case of Father-daughter or Mother-son incest, the child will be missing one grandparent and have one duplicated. Because it is quite common for cousins to marry (even 1st and 2nd cousins, but certainly more distant cousins), there could be extra DNA from one or more great-grandparents and so forth. Also, you can have "double cousins" when two brothers marry two sisters. This strengthens some lineages as well.

But I do not believe there is an infallible test for incest, although some recessive traits might be a dead giveaway if one parent or the other does not carry a gene to match the other parent's recessive gene.

It usually takes eight or nine generations for "normal" or "legal" inbreeding to produce genetic defects based on recessive genes. But in the case of incest, the short circuit is much shorter and more direct. Your father is your grandfather, or your mother is your grandmother. You are missing a second grandparent of the same sex. In the case of a family with multiple generations of incest, things fall apart genetically pretty fast genetically.  A recessive gene borne by only one ancestor can start to cause problems in as little as tthree generations I believe.
 
2017-11-04 04:27:21 PM  
I think what i just said was right as far as it goes, but it occurs to me that XX and XY do come into play.

Sons inherit one X chromosome, from their Mother. Daughters inherit two, one from their Mother, one from their Paternal Grandmother. Man on boy incest can not be determined genetical from the XY chromosomes. Man on girl incest and woman on boy incest would shift the odds of inheritance. The Y chromosome would be the same in the case of a grandson, when it ought to be that of some other man unless it is a very common type. Y chromosomes only pass in the paternal line. Daughters do not inherit them. X chromosomes pass only from the maternal line to sons, and 50% paternal-maternal in the case of daughters. So an X and Y chromosome might give way more information than the other 22 chromosomes pairs. In the same way, a son inherits mitochondrial DNA but does not pass it on. If it passes on to a grand-daughter, there is a non-paternity event of a different sort.

Other forms of incest are harder to prove or track. Mothers, fathers, sons and daughters are the main and firmest definition of incest. In some societies, the concepts of kinship are completely different. Clans may determine incestuous relationships rather than propinquity.

In any event, mutations occur at a fairly fixed rate, although rarely, in Y chromosomes and mitochondrial DNA which are inherited whole. This makes things slightly different for those two chromosomes. Other chromosomes may tell no tales.
 
2017-11-04 07:06:46 PM  
Thereʻs a link in gedmatch.com that tests your genetic data to see if your parents are related - you have to upload your ancestry/23andme/whatever data to them though.
 
Displayed 37 of 37 comments

View Voting Results: Smartest and Funniest

This thread is closed to new comments.

Continue Farking

On Twitter





Top Commented
Javascript is required to view headlines in widget.
  1. Links are submitted by members of the Fark community.

  2. When community members submit a link, they also write a custom headline for the story.

  3. Other Farkers comment on the links. This is the number of comments. Click here to read them.

  4. Click here to submit a link.

Report