So just how useful is dna?

Question

A long conversation about dna has been taking place over a couple of profiles I've been studying so perhaps this is time for me to ask the question that has been bothering me. Just what can be demonstrated by sampling the dna of living people? My understanding is this, we inherit our dna from our parents in a random manner. If neither of them has a particular gene then nor can we. Very, very occasionally a glitch happens and things can be changed, usually for the worse but sometimes for the better. In the past if it was useful more recipients would survive than otherwise, if unhelpful then fewer but this could depend on the environment. Some genes don't display physically unless you receive them from both parents, some are sex-linked (colour blindness for example). Third year biology, as far as I ever went. Statistically we inherit half from each parent but that's only statistics and statistics are only useful for studying populations, not individuals and it is possible though very very very rare to inherit all one's dna from just one parent. (That's like winning the lottery, almost impossible but people gamble anyway.) So if we test all the living members of a family we should be able to tell where they all fit in. But that's not so. Were my father also my grandfather (both dead) some say it would show in the statistics, but that is not necessarily the case. Surely all we can actually do with dna is demonstrate that the paper trail is probably wrong? Clearly I am a sceptic. I'm not entirely convinced that the remains recovered a few years ago are proven to have been Richard III, I'd bet there was a lot of unrecorded royal dna floating around then and there is now, though he did have an interesting skeleton. I have been told that it might be possible to determine who my unknown grandfather was by sampling my dna but I cannot believe it. In time it might be possible to say this guy is possible but there were probably a lot more possibilities around then. As I said earlier, statistics prove nothing much about individuals and from what I have read on here the analysis is mostly based on statistics determining cousins, etc. Wouldn't mind being enlightened.

Comments

I disagree with your comment that with DNA we can only demonstrate that the paper trail is probably wrong. On the flip side I believe that Autosomal DNA tests are most valuable when we have a paper trail and this can be used to match living siblings and/or living distant cousins to living and/or deceased common ancestors. I have used this method to connect my mothers deceased great grandparents buried in Australia to their previously unknown deceased USA parents.

---

Well for me it's been two fold: first I did the Y-DNA & it absolutly worked. Connected me to a demonstrable 3rd cousin who was a geneology nut & had us traced back to our 3x g grandfather from 1793 that was our MRCA. Secondly, the automosal test has connected me to a lot of,also demonstrable cousins that helped confirm several other tree lines. As someone else on here stated, it's another tool to use. With automosal there is a lot of chaff along with the wheat, you just have to do the leg work to find the nuggets.

Answer 1

I'm not an expert at all, but I have been studying it (for genealogical purposes enough) to correct one thing you said: you cannot inherit all of your DNA from just one parent. When you are conceived you receive one half of your DNA from each parent. You are correct, however, in that it is more or less random which genes you inherit from each parent. So, when each of your parents was conceived, they also received 1/2 of their DNA from each parent.  But because of the randomness of which genes you receive, it may be theoretically possible (although I have never heard of such a case, and don't believe it actually happens) that from one parent the autosomal DNA you receive is from only one of their parents. But the other half of your DNA is still made up of your other parent's DNA.

In my experience, DNA is extremely useful. Yes, it is all based on statistics, but the statistics are from a vast number of people and continually being improved. No, DNA cannot absolutely prove a relationship beyond immediate family, but it does provide strong clues, and used in conjuction with the paper trail can be very compelling evidence.

Answer 2

Just to clarify the random nature of "about half" of the autosomal dna.

You definitely inherit exactly half your dna from your mother and half from your father, unless you have a major anomaly which would show up as a medical condition.

The egg you came from had half of your mother's DNA and the sperm half of your father's DNA. Together they made a complete set: you.

The about half bit comes in when the egg or sperm is formed. As the copy is made of the DNA at random points it switches from one of their parents to the other parent. So on average about half comes from each of their parents but depending on where the switches take place it might be a bit more or a bit less. It would be very unusual to only get all the DNA from one parent. (The X and Y chromosomes are a special case where you get all of one of them from your father, and occasionally an exact copy of one of your mother's.)

Hope that helps a bit

Tim

Answer 3

For me, the best way to look at is that DNA evidence is like any other piece of evidence, in that you have to use it in relation to what it offers for the research question you are studying.

You have listed a number of points about DNA in general.  While some of these limitations can be relevant, it all depends on what you are trying to do.  For example - if I wanted to check to see if my Dad is my biological father and not my biological grandfather, we could each take the Ancestry DNA test - which is an autosomal DNA test that measures the autosmal DNA we get from our parents (50% from our biological father and 50% from our biological mother).  If the Ancestry DNA test results show that my Dad and me share 50% (technically 3,400 cM), then this strong evidence that we are father-son.  If my Dad and me only share 25% (around 1,700 cM), then this is strong evidence that I am actually his grandson.

Now - you might reply - "but what if I don't have a test from my father since he passed away?"  Well, like any other genealogy research question, we will have to look into the specific evidence that we have and then see if we can do DNA testing with other living relatives whose biological relationship might providence evidence toward/against a conclusion

For example - if you have a living first cousin who you expect to be the nephew of your deceased father, then the first cousin should share 12.5% of their DNA with you - around 850 cM.  If you and this cousin only share 375 cM, then this points strongly to first cousin - 1X removed.  DNA evidence of you and a first cousin actually matching at a first cousin - 1X removed level would point toward your father actually being your grandfather.

Answer 4

You inherit exactly half from each parent? How on earth does that work?

Comments

You don't necessarily inherit precisely half from each parent. You can, for example, be 49% your mom and 51% your dad. But no one is 0% their mom and 100% their dad. There is a range of shared cMs that certain degrees of relation will fall within. For example, full siblings can share from between 2209 – 3384 cMs. 2209 is on the extreme low end. But no set of full siblings will share only 1783 cMs, which is the average for what a pair of half-siblings will share.

---

Leaving out the sex chromosomes, the X has many more genes than the Y, you exactly inherit 50% from mom and 50% from dad.

Each parent gives you 23 chromosomes, which pair with the 23 chromosomes from the other parent. Now each of those chromosomes is a poorly shuffled combination from each of the maternal and paternal grandparents, which means that you can have a grandchild only be 20% of grandparent 1 and 30% of grandparent 2.

---

The autosomal DNA is in pairs of equal length with one member of each pair from the father and one member of each pair from the mother.  This is always 50% each (or so close to 50% as to make any variance irrelevant).

The 23rd pair also has members that are the same length in females.  In males, the y chromosome from the father is shorter but although it has fewer genes, studies have show that the number of 'active' genes in the y and x chromosomes is balanced.

Some would argue that the tie-breaker then, is the mitochondrial DNA that is found outside the cell.  Both the mother and father contribute mitochondrial DNA but the contribution from the father is almost always destroyed by the egg.  Ignoring those few rare cases, we would say that only the mother's contribution is in the DNA of the child and that the mother, therefore, contributes slightly more DNA than the father.

To address this, we need to be careful not to give to much weight to mitochondrial DNA with its 37 genes compared to 2800 in chromosome 1 and and still hefty 750 genes in chromosome 22.  We also need to consider whether the DNA from each parent is used in a balanced ratio (see: [http://blogs.discovermagazine.com/d-brief/2015/03/03/genetically-more-like-dad/#.XBUudlxKi0p Genetically Speaking, You’re More Like Your Dad])

So, while the purists might want to argue some of the finer points, we effectively get half from each parent.

Answer 5

I'm guilty of wooly thinking.Of course you get half from each parent. But surely your Dad could give you all or nearly all the dna he got from his dad of from his mum, likewise with your mother.

Comments

Not strictly impossible.  But the majority in each chromosome is a coin-toss, so for all of the 22 autosomal chromosomes you got from your father to be biased in the direction of his father, even 51-49, would be like tossing heads 22 times in a row.

---

A bit more complicated but essentially, when the egg or sperm is created it gets one each of each pair of chromosomes with a 50-50 chance of which one gets included. So, probability leaves a pretty wide range of which "set" is included. To make things a bit more complicated, recombinant processes can occur where the paired chromosomes can exchange part of their DNA while the egg or sperm is being formed thus creating a chromosome that isn't entirely the mother's or father's. That can skew things a bit more to one or other parent.

---

I have seen someone who shared only 18% of his DNA with one grandparent 32% with the other. Another cousin shared 20% with the paternal grandfather and 30% with the grandmother.

I think the reason is that there are really a lot fewer shuffles than most people think and the cards are very sticky.

---

But the direction of the skew is 50-50.  So it's 4 million to 1 against 22 recombined chromosomes all containing more paternal than maternal DNA.

---

It's a normal distribution curve (mostly) demonstrating that one gets 25% of one's DNA from each grandparent.  That's 75% true for 68.26% of the population, 50% true for 95.46% and 25% true for 99.72%, leaving 0.28% of the population for whom it is hardly true at all. The proportion of the population for whom it is completely true is exactly the same as the proportion for whom it is completely untrue. If there are 6 billion of us then there are in excess of 15 million for whom it's hardly true at all. That's an awful lot of people. Statistical analysis only works for entire populations.

---

When a parent create the DNA that will be transmitted in the sperm or egg, a process of meiosis will duplicate the current chromosomes and then create new strands that are random mixes of that parent's paternal and maternal chromosome.  So, the child will get 50% of its DNA from that parent but that contribution might be 33% from that parent's mother and 66% from that parent's father.  In other words, it is a balanced bell curve with the peak at the 50% and 95% of cases being no less than 33% or more than 66%.  From the perspective of the child, we could have 25% from each grandparent but possibly only 17% from one grandparent and 34% from another.

Answer 6

Ever hear of a chimera?   See Lydia Fairchild on WIkipedia.  She gave birth to children who did not match her DNA.

This is why DNA research uses statistical probabilities.  Nothing is definitive.  Or as Hamlet put it: There are more things to Heaven and Earth, Horatio, than are dreamt of in your philosophy.  Nothing is perfect nor set in stone.

I tend to think of DNA results are like a badly transcribed Census record.  While there is some truth, sometimes it just doesn't seem right. LOL

Answer 7

I'm not sure why you don't believe that it might be possible to determine your unknown grandfather through DNA testing...  Adoptees are tracking down their birth parents using DNA.  I've been able to determine my paternal great grandfather's family through autosomal DNA testing and triangulation.  It isn't usually an easy or quick task to do this, but it's certainly possible.  And the more people that test, and the better software that is out there to assist us, and the larger family trees that people put together, the easier it will be down the road.  I also recommend that people work first on creating their tree going back several generations, including all of their ancestors siblings and those siblings spouses and children, and their spouses and children.  It's very time consuming, but without an extensive tree you aren't going to have much luck in determine the common ancestral couple with your matches.

Comments

In as much as one can have an extensive tree with a very small family then I have that (living people have hardly been added to Wikitree). Close members of the family are almost entirely opposed to allowing anyone to record and broadcast their dna results. (I'm of that view). That apart I knew (and cared about) my grandmother. Although perfectly capable of running her own life she had a desperate need to have men in it. She was also adventurous so to say that mum was conceived in the place where her family lived is just guess work. I likely carry 25% of his dna but just possibly very little. With a lot of testing it should be possible to determine what came from Mr X, who was probably English and was wherever Gran was some 8.5 to 9.5 months before mum's birth. Gran kept the secret to the grave, despite the pleas of her daughter, to whom she had devoted her life, to be told the truth. Could possibly find lots of potential Mr Xs to blame for helping to create a child he could not or would not support, presuming he ever knew. Either way he risked doing it and did not care about the consequences so I guess I don't want to know about him anyway. Also note that this was 100 years ago.

---

No problem.  I appreciate your feelings.  You just had asked about how useful DNA testing was and doubted the ability to use it to track down a grandparent, so I was responding with my experience. FYI, my great grandfather was born over 130 years ago.  For me, I wanted to fill out that quarter of my mom's tree.  I've been working on my family tree for 35 years and have over 34,000 relatives in it.  I wanted to know who had fathered the grandmother that was my favorite person in the world.  I keep adding the spouses of the siblings (and all the children, etc) going back ten generations, so there are lots of people.  And DNA is the only way to break down the many brick walls of the grandmothers with only a known first name... So I love DNA and the opportunities it presents to locate these ancestors of mine whom I cherish so much.  My other grandmother was very into family history. She wrote down what she knew, and I used her notes and worked on her side when I first got started. I think I got the genealogy bug from her!  One note she wrote said, "It's a shame we know so little of mother's family."  Well, I've figured out a massive amount about that family.  Wish she were here so I could tell her about it.  She'd be thrilled...

---

You don't need to analyse your DNA.  The only way is if a relative of Mr X tests and you both have your results on a matching site.  That way you find a cousin who you can't place anywhere in your tree by known genealogy.

Outside America your chances are minimal because (a) fewer people have tested (b) people tend to have fewer cousins.

---

RJ, it's true you don't 'need' to analyze your DNA.  But then again, you don't 'need' to work on a family tree or know anything about your ancestors.  Point being, what does need have to do with it?!    Many of us do it solely because we want to.  But then again, my family would probably tell you that I definitely 'need' to do what I do with genetic genealogy.  It's my passion.  On the other point, it's easier now for people around the world to take DNA tests, so as the decades pass it will help more people throughout the world as opposed to only the U.S. and Canada.  For that, I'm very thankful.  I just wish I were a couple decades younger to be around and have the greater benefits of it.

---

RJ is correct that there are more people from the US and Canada that have tested but that is changing. My grandfather was the youngest of eleven and though I knew the names of his siblings, that was it. There was never any contact. I've recently, through DNA matches been in contact with descendants of two of his siblings, still living in the UK.

I've also been able to find the common ancestor to more distant relatives in Canada ,the US and Australia. My direct ancestors were mainly Ag labs and though they obviously stayed in England, there were siblings that emigrated .

Recently, I came across a new wiki tree member from Australia who made a profile for someone who I knew to be the sibling of an early 19th C ancestor. This member does not know the parents of his grandmother ( I think she , like your mother didn't know) .However, using DNA, it has apparently  been possible to work out  who her grandmother's, grandparents were. Through gedmatch we were quickly able to confirm that we also match at the expected level ( common ancestor 7 generations ago) The paper trail to Australia is also strong. So at the moment, the unknown g grandparent is still unknown but it looks as if the brickwall has been hurdled.

Answer 8

Never mind -- commented elsewhere.

Answer 9

DNA evidence is best through the y-DNA test (males and mt-DNA test for females).  X-DNA testing tracks alternate generations for males as a male usually inherits only one X (from mother).  Au-DNA testing covers the other inherited DNA sequences.  Even so given the nature of gene mutations DNA sequences will change over the generations.  Some sequences necessary for continued species existence have remained stable for millions of years.

Answer 10

I have been told that it might be possible to determine who my unknown grandfather was by sampling my dna but I cannot believe it.

I don't know why you would believe that; it's certainly possible that DNA can tell you who your unknown grandfather was, and depending on whether you have any unknown close relatives related through that grandfather who've tested with any of the commercial DNA services, it's almost 100% certain that DNA can answer that question. For instance: My mother tested with Ancestry. Her top match was a person with a name I didn't recognise, with an estimated relationship of "1st cousin". This person's shared matches included people who were descendants of her maternal grandfather's ancestors, and her maternal grandmother's ancestors. Based on the amount of shared DNA (which tends to be consistent within an average range for a given relationship), one of this person's parents was a full sibling of my grandmother. Talking to this person, I discovered that their mother had been given up for adoption at birth; their mother was the previously-unknown full sister of my grandmother. My great-grandmother left my great-grandfather in the early 1940's, and it turns out that she was pregnant at the time she did it. And this is something I never would've discovered without DNA testing.

Answer 11

One can make very strong assertions using DNA but rarely in isolation.  Having a good knowledge of the families and medical histories involved can be very important. Typically, the closer the family relationship, the more one can rely on DNA.  For example, a birth certificate might say who the father is, but if the father and child do not match with DNA, we can ALMOST be certain that the birth certificate is in error.  I say ALMOST because some bone marrow transplant recipients will not match.  Conversely, you can have a false positive involving monozygotic (identical) twins.

Generally speaking, autosomal DNA is good for third cousin and closer but can also be useful for more distant relationships with more than just two cousins matching (triangulation, clustering).  The yDNA and mitochondrial-DNA (mtDNA), however, can go back hundreds of years with great reliability.  They follow only the paternal and maternal lines respectively but are less prone to recombination offering a deeper look at a person's ancestry albeit focused on a single line for each.  It is yDNA and/or mtDNA that is/are used for these inspections of royal DNA. For Richard III it was yDNA.

It bears repeating that the DNA alone is limited but it can help to support or destroy certain claims. It should also be mentioned that some will misconstrue what the DNA implies.  For example, some would like to say that Thomas Jefferson had male offspring with one of his slaves based on the fact that some of her descendants have yDNA that matches some of his brother's descendants.  While this does imply that Thomas could POSSIBLY have been the source of that yDNA, it really only says that one of several Jefferson men were the source (other evidence points to the brother).

To your question, I would answer that it is very useful but must be handled in context so as not to be misused.

Answer 12

I think it is extremely useful, especially in the closer cousins range (2-4), and sometimes even farther out (4-6) using Gedmatch.com utilities for comparison. There one can see the specific chromosome interception points and overlaps that prove an inherited SNP of DNA particular to a line that may descend from different siblings within the same family. It helps identify that specific SNP that is passed on within that family.. and validates, substantiates the paper trail of the line. A successful triangulation is priceless!

Answer 13

In my case it certainly helped. I did the Y-DNA and Mt DNA deal through family tree and the Ancestry DNA.   While I already had a good paperwork trail going back to the late 1600s. I was hoping to gain some knowledge about the family origins before 1600s Virginia.  ( WHICH I HAVE NOT) ...

However, I did discover some previously unknown relatives who all link via DNA back to the same common ancestor.   ( Capt Christopher Clark 1678 to 1754)   I also discovered that there were a two or three other  families who were genetically Clarks .  ( They were neighbors of my Clarks' back in old Virginia and one of the Clark boys liked to visit any gal within a days ride. )     I was also able to disprove a couple very old family legends on both my fathers side of the family and my mothers side, after no native American Indian DNA was found.   Plus I think I have been able to disprove another question regarding Penelope Clark the wife of Capt Christopher Clark back in the late 1600 and early 1700s.   One camp likes to say she was a Johnson and the other camp claims (hopes) she was a Bolling so they could claim descent from Pocahantus.    Well I have over two pages of Johnsons' who are DNA cousins and NO Bolling DNA cousins.  That combined with no Indian DNA would sort seem to seal the deal.      If you can find another living relative to ALSO submit a DNA sample it would help add to any info you might gain.

Answer 14

If you are trying to find family it can be useful.

An example. This year on ancestry a second cousin appeared as a match to me. I recognised the surname as the married name of one of my paternal grandfathers sisters. I made contact with the cousin and he was able to give me the story of his grandmother. But the big revelation was for him. He did not know his grandmothers maiden name, which I provided, as well as all the research that had been done on the family which he had no idea existed. We now keep in contact via emails and phone calls between Canada and Australia.

A second point about this is DNA can keep on giving. I first tested years ago but as more cousins test they appear as matches.

I also have a distant 8th cousin who did a Y-DNA test this year. He appeared as a very close Y-DNA match to me. As it turns out he was adopted, and who he thought was his maternal grandfather turns out to be his paternal grandfather. Same surname as mine.

Answer 15

In my experience it has been both useful and useless. My reported ethnicity is 0% Native American in spite of having a large number of autosomal genetic matches with Native American haplotypes. The system is flawed and prejudiced IMHO. This is where science becomes science fiction.

On the usefulness note, I have located and confirmed 100's of Native American blood relatives. Many of them would have gone unrecognized had I not had a very extensive family tree already established.

So yes, if you have an extensive family tree, an autosomal test can be very useful.

Comments

Really, it just sounds like you could stand to learn a bit more about it - most tools are practically useless if you don't know how to use them! The ethnicity results have almost no place in serious genealogical analysis, and Native American ancestry is often missed (especially when it is a very low percentage). It's all about the matches!

---

Bud, the accuracy of ethnicity results depends on the populations tested, how many testers of each populations there are, and the amount of effort that has been put into identifying DNA bits characteristic of a given population.  Unfortunately, Native American DNA is disproportionately under-represented, very few have tested, and that means there's probably insufficient data to draw meaningful results from.  Once there are more full blooded Native Americans tested, and identified as such, there should be better data to identify characteristic DNA bits (both individual SNP's and full segments).

The other problem is that autosomal DNA gets diluted so quickly with each generation back.  Unless your Native American ancestry is recent, it may have been overwhelmed by all your other ancestry.

---

Nobody is partially native American.  One is either born in America and thus 100% or not and thus 0%.  One could be partially aboriginal though.  

N.B. the term 'indigenous' is synonymous with 'native.'  Note too, that some will interpret 'America' as the USA and others as the two continents, the lands linking them, and nearby islands.  Canada, not being satisfied with the single word 'aboriginal' when it can be expanded to two, uses 'first people.'  One is left to ponder whether their legislature is paid by the word or simply have a limited vocabulary (there are other possibilities but they are quite Orwellian).

---

Really Frank? I did learn all I need to know about it. The Anzick sample was altered because it was argued that certain haplotypes were European and not native American which is total BS. The Anzick child was found in Montana, not in Europe and he lived before all of the European migrations. All of the haplotypes that matched his sample were therefore not European. With the unaltered sample, I tested as an 8.83% match to Anzick boy, that is sufficient genome to be regarded as his 3rd cousin and is more in keeping with the rest of the matches that I have encountered. DNA testing can be useful as I previously stated if you're a serious genealogist, but if you're just curious about your past you will not get an accurate report especially if you have Native ancestry.

---

Thom, if you really want to get technical about it, the indigenous regard themselves as the inhabitants of "Turtle Island" aka. North America. So truly they are North Americans despite our Canadian, American and Mexican borders and any label that has been placed upon them.

---

I do not mind being technical as words with ambiguous meaning hamper correct understanding.  Churchill once said that Brits and Americans were two people divided by a common language.  Very apt as we share many words that have different, and sometimes, opposite meanings.

To your point about the indigenous regarding themselves as inhabitants of "Turtle Island", I would have to say that although there might be a few indigenous persons who think in that way, I can safely say that I, as an indigenous person do not and would go further and say that I know many indigenous persons who also do not think in that way.

---

Yes, really, Bud.

You can get lost in the weeds of the details of how they try to identify specific ethnicities, but the Big Picture is that you don't get exactly the same amount of DNA from each ancestor, once you get past the generation of your parents. If you have just a small amount of aboriginal ancestry, that small amount of DNA may not contain whatever genetic markers they look for. In some cases, you may have simply missed the boat on getting DNA from that ancestor altogether.

The idea that there is some sort of conspiracy or bias within the testing community is unfair to them, and I would add - just kind of ridiculous.

---

Well it seems that they have altered or replaced the Anzick sample with the explanation that several of his genetic matches seem to have European haplotypes. As I said previously, he was found in Montana and his age was estimated at 12.5k years which predated ANY European migrations.

Ridiculous and unfair? I think not.

So what is a person supposed to think when the scientific evidence is altered because of a difference of opinion? This wouldn't be the first time this has happened either.

---

I would certainly be the last  person to claim that there was no such thing as corruption within the scientific community, but it sounded like you suggested that this was some sort of act motivated by "prejudice" against aboriginal people, which is not consistent their tribal nature these days.

But that's all a side point, and that in itself is my point: If the analysis of the remains of a single individual who died 12.5k years ago is critical to what you're doing with genetic genealogy - you're just doing it wrong.

---

As I said before, the genetic matches I have are inconsistent with the ethnicity report of the same sample...They are inconsistent with each other! Trying to make sense of all this and coming to the conclusion that all aspects (including oral histories) of my research neatly fall into place with the original sample is "doing it wrong"? Deductive reasoning isn't "wrong"!

Prejudice: preconceived opinion that is not based on reason or actual experience (or scientific evidence).

Answer 16

Matching YDNA tests are the ONLY way to sort out paternal lines and who is related to whom for common surnames such as Smith, Jones or Johnson.  But your DNA in any format doesn't give you a ready made family tree.  You still have to do the paperwork/research. The DNA is just a more specific research tool.  

YDNA has definitely shown us where old books had the wrong Smith men related.  This happened occasionally with colonial Smith lines, and yDNA has helped us connect these early American families to their biological ancestors.

Comments

That's certainly true for many generations back, but auDNA can do a lot of that too, if you're considering the last 100-200 years.

My gt-gt-gt grandfather was a Johnson, and had a son named William. Since he moved away (or possibly died, as far as I knew) there was basically no chance to find out what happened to William Johnson from my end. But his descendants showed up in my brother's DNA matches, plain as day. Some of the sisters' descendants have emerged n our DNA matches as well.

Answer 17

For determining your ethnicity, DNA testing is pretty much useless. The different companies are widely different in their interpretations. which is why I say it is useless - or very very wooly. Their interpretations are colourful and nice, but not always accurate.

For finding relatives on the other hand, DNA is very very useful.

My mother was adopted as a child, and she and I both did a DNA test this year (2018). We have been very fortunate to find some close cousins and we are slowly making contact and arranging times to meet.