PDA

View Full Version : Apparently Geno 2.0 goes back only 5-6 generations



Missouri1455
01-06-2017, 03:40 PM
A month ago i talked to a person who worked for Geno and he told me straight up in terms of ethnic percentages geno 2.0 only goes back 5-6 generations; their biggest focus is the subclades; which got me thinking that the company calculates the ethnic percentages differently from 23ndme/ftdna; so what i noticed is geno 2.0 basically inflates any minority admixtures you might have, thats why Europeans sometimes score around 50% european and 40% meditreanean when in other tests(ftdna/23ndme) they will be given a more accurate results of 90% european or more.

I believe the calculation differs significantly between geno 2.0 and 23andme/ftdna but to me 23ndme/ftdna seems more accurate when it comes to ethnicity results.

I just wanted to share this with people who were not aware of this.

Thank you

jpb
01-06-2017, 03:54 PM
Isn't the ethnicity on age Geno 2.0 NG the same pretty much as MyOrigins?

Missouri1455
01-06-2017, 04:01 PM
Isn't the ethnicity on age Geno 2.0 NG the same pretty much as MyOrigins?

Tbh all i know is that ethnic results from ftdna/23ndme are more reliable than geno 2.0; and the reason for that is the calculations between ftdna, 23ndme, and even ancestry tend to be very close and similar in terms of ethnic results.

Missouri1455
01-06-2017, 04:27 PM
I feel i've been ripped off in terms of my ethnicity results(Geno 2.0); but anyways i have already applied a ftdna kit so hopefully that will settle things down asap

kingjohn
01-06-2017, 04:31 PM
i saw some results of members here and other forum
and there ftdna my origins and geno2 next generation components % are almost the same ...
best regards
adam

Missouri1455
01-06-2017, 04:38 PM
i saw some results of members here and other forum
and there ftdna my origins and geno2 next generation components % are almost the same ...
best regards
adam

Ill post my ftdna results here and compare it with my geno 2.0 as soon as possible :thumb:

Cofgene
01-06-2017, 04:38 PM
Be careful in your use of terms "accurate" and "reliable". There are not ethnically pure representatives, or definitions thereof, for reference populations in these calculations. They are sets of results which have been grouped and labeled for having characteristics of the region attributed to them. A result cannot be considered "accurate" but should be considered a "similar to." For reliable the calculations are done based upon similar mathematical models and different datasets. Reliability is considered within a framework of repeated measurements/calculations. So it would be more appropriate to state that the results exhibit more concordance or similarity than to infer improper methodology between the models. There is no gold reference standard so differences should be expected and results should be referenced in a manner that supports the lack of reference standards.

Missouri1455
01-06-2017, 04:52 PM
Be careful in your use of terms "accurate" and "reliable". There are not ethnically pure representatives, or definitions thereof, for reference populations in these calculations. They are sets of results which have been grouped and labeled for having characteristics of the region attributed to them. A result cannot be considered "accurate" but should be considered a "similar to." For reliable the calculations are done based upon similar mathematical models and different datasets. Reliability is considered within a framework of repeated measurements/calculations. So it would be more appropriate to state that the results exhibit more concordance or similarity than to infer improper methodology between the models. There is no gold reference standard so differences should be expected and results should be referenced in a manner that supports the lack of reference standards.

I agree i might have misused the word accuate/reliable but ill use the word similarities; from what i have seen the similarities i found in the different calculation results; ftdna, 23ndme, and ancestry results tend to be much closer than geno 2.0.

In my case i do believe geno 2.0 inflated my minority admixtures and the reason for such claim is based on my ethnic results geno 2.0 gave me 20% for east african; yet in gedmatch when it came to population distances i barley found any east african groups listed; at most ill find 1(east african) out of 20 and in some cases ill find none; which i believe reveals how much geno 2.0 tends to inflate with ethnic admixtures but with time we'll know; so at the moment i do believe a company like ftdna/23ndme will help me shed light on this issue.

kingjohn
01-06-2017, 07:06 PM
test with ftdna
it is not expensive now
there autosomal test..... only 79$much less than geno2 ....
best regards
adam

p.s
geno2 have there things
in the clades as you say the y dna and mtdna + the neaderthall %

Missouri1455
01-06-2017, 07:13 PM
test with ftdna
it is not expensive now
there autosomal test..... only 79$much less than geno2 ....
best regards
adam

p.s
geno2 have there things
in the clades as you say the y dna and mtdna + the neaderthall %

Thank you; the kit is on its way im hoping it arrives as soon as possible.
About the neanderthal i scored around 0.6%; i was wondering what do ethiopians, egyptians, levantines score on their neanderthal %?

Thank you in advance

Akra20m
01-06-2017, 07:35 PM
i saw some results of members here and other forum
and there ftdna my origins and geno2 next generation components % are almost the same ...
best regards
adam

I guess that is only the case if you transfer the raw data from geno2 to ftdna.

Principe
01-06-2017, 08:26 PM
A month ago i talked to a person who worked for Geno and he told me straight up in terms of ethnic percentages geno 2.0 only goes back 5-6 generations; their biggest focus is the subclades; which got me thinking that the company calculates the ethnic percentages differently from 23ndme/ftdna; so what i noticed is geno 2.0 basically inflates any minority admixtures you might have, thats why Europeans sometimes score around 50% european and 40% meditreanean when in other tests(ftdna/23ndme) they will be given a more accurate results of 90% european or more.

I believe the calculation differs significantly between geno 2.0 and 23andme/ftdna but to me 23ndme/ftdna seems more accurate when it comes to ethnicity results.

I just wanted to share this with people who were not aware of this.

Thank you

hman92, before I transferred to ftdna, I called nat geo for this reason, they told me the percentages we get are from the last 6 generations, but could represent your admixes in the last 1000 years, this is why the admixes seem off, 23&me from my opinion gives you the last 500 years, while Nat Geo/Ftdna gives you the true ancestral components, to say I was not shocked to get 39% Middle Eastern, 30% (29.5% is the true number when I called) Asia Minor, was a bit high, you figure it would be something recent, but it seems to be a component of Southern Italians to get, as I see it most of it might represent an ancestral component (Neolithic/Bronze Age), and some being recent (Roman era/Byzantine), I have a Turkish FF match whose reported at 4th to distant cousin, and if I go on Gedmatch I have a couple more plus some Asia Minor Greek matches. The North African is coming from the Emirate of Sicily period predominately and for the Jewish it comes from my father's side with my subclade and additional IBD sharing with Jews, which is rare (I think) for Southern Italians. I prefer Nat Geo/My Origins because it shows the ancestral component for people (it is nowhere near perfect but still fascinating), also you see people from across the Mediterranean getting some % of other Mediterranean groups which goes to show how busy the ancient Mediterranean was.

For Y-dna Nat Geo is not good, they incorrectly gave me J-CTS2906, it was only ftdna that could give me the most accurate results for my Y-DNA, they did get my mt-dna right however. In the end it is up to what you find to be the best test, for Y Ftdna is definitely the best, for autosomal it's all about preference.

Oleg (Rus)
01-06-2017, 08:53 PM
They just look at your alleles combinations and compare them to reference samples. What generations are you talking about? I don't think that "generations" in this case make sense at all. It could mean something, if they used ancient samples, but, AFAIK, they actually use samples of modern people.

Missouri1455
01-06-2017, 10:48 PM
They just look at your alleles combinations and compare them to reference samples. What generations are you talking about? I don't think that "generations" in this case make sense at all. It could mean something, if they used ancient samples, but, AFAIK, they actually use samples of modern people.

generations as in the past 5-6 generations so approximatley around last 100-200 years; but as principe said they can pick up an admix from the last 1000 years but thats not guarantee; so yeah geno 2.0 is not ancient as they call themselves to be; the only ancient thing about them is the subclade information they offer.

kingjohn
01-06-2017, 11:22 PM
I guess that is only the case if you transfer the raw data from geno2 to ftdna.

read is case
http://www.theapricity.com/forum/showthread.php?199562-NatGeo-FamilyTreeDNA
i agree that each case is individual
khandboshi from this forum
also got very similar autosomal components % in both ftdna and geno2 .
regards
adam

wombatofthenorth
01-07-2017, 01:44 AM
Ill post my ftdna results here and compare it with my geno 2.0 as soon as possible :thumb:

I think it will be identical, only FTDNA will give less accurate maps for what each region means and no background reference info.

wombatofthenorth
01-07-2017, 01:56 AM
I guess that is only the case if you transfer the raw data from geno2 to ftdna.

So people who tested at FTDNA first get a different result from MyOrigins than those who then later or only transfer over from Geno 2.0 NG?
So what? FTDNA doesn't run their own test but just copies over the % from Geno 2.0 NG??

wombatofthenorth
01-07-2017, 01:59 AM
generations as in the past 5-6 generations so approximatley around last 100-200 years; but as principe said they can pick up an admix from the last 1000 years but thats not guarantee; so yeah geno 2.0 is not ancient as they call themselves to be; the only ancient thing about them is the subclade information they offer.

Not really, what they mean is they only list down to 2% and beyond 6 gens anything coming from just one single line would probably total to under 2% and might not register BUT at the same time the way they compare what you do get does look back a ways, and deeper than23 which tries to use larger blocks to try to filter results more down to 500 years.

vettor
01-07-2017, 02:14 AM
So people who tested at FTDNA first get a different result from MyOrigins than those who then later or only transfer over from Geno 2.0 NG?
So what? FTDNA doesn't run their own test but just copies over the % from Geno 2.0 NG??

I agree for copying , but

ftdna do or did the testing for Geno , so as for myself........I did ftdna first, then did geno ....but never transferred
my son only did geno and transferred to ftdna...................so our ancestry are more different than I would like.( my ftdna with son's geno )...............this is because although ftdna do all the testing, geno ask for different areas to be tested than what ftdna test.

But.........they do come together in gedmatch..............maybe due to the phasing

jortita
01-07-2017, 04:31 AM
Is this the newest Geno 2.0 with the new company as I tested with the FTDNA one in Feb 2016, and to me both my tests are related to my deep ancestry and not recent ancestry as I get Finland and Northern Siberia 5% and North Africa 2% on my Geno 2.0 next generation test

khanabadoshi
01-07-2017, 05:46 AM
read is case
http://www.theapricity.com/forum/showthread.php?199562-NatGeo-FamilyTreeDNA
i agree that each case is individual
khandboshi from this forum
also got very similar autosomal components % in both ftdna and geno2 .
regards
adam


I transferred my Geno 2.0 to FTDNA. So my "FTDNA" results are from Geno 2.0 and my Geno results (that I now post) are from the Geno 2.0 Next Gen.
I transferred the Geno 2.0 Next Gen kit as well to FTDNA, but I can't remember the password or kit number...

jortita
01-07-2017, 05:48 AM
I did the same, my FTDNA results are transfer from Geno 2.0, I tested with Genographic

khanabadoshi
01-07-2017, 05:53 AM
I did the same, my FTDNA results are transfer from Geno 2.0, I tested with Genographic

So as I understand it, our MyOrigins might be different if we had just tested with FTDNA alone, and it might be more accurate?

jortita
01-07-2017, 09:31 AM
So as I understand it, our MyOrigins might be different if we had just tested with FTDNA alone, and it might be more accurate?

I tried ordering a kit, but they did not let me till recently. Given that they have severed their connections with Genographic just a few weeks back, I hope it would be possible to order a kit from them

kingjohn
01-07-2017, 02:03 PM
I transferred my Geno 2.0 to FTDNA. So my "FTDNA" results are from Geno 2.0 and my Geno results (that I now post) are from the Geno 2.0 Next Gen.
I transferred the Geno 2.0 Next Gen kit as well to FTDNA, but I can't remember the password or kit number...

didn't knew it dear khan ,
the serbian guy in the link did 23and me test
and transfered his raw data to ftdna my origins
and than did geno2 next generation test
and his results are very similar
maybe his case on his own ......
best regards
adam

Larth
01-07-2017, 02:34 PM
read is case
http://www.theapricity.com/forum/showthread.php?199562-NatGeo-FamilyTreeDNA
i agree that each case is individual
khandboshi from this forum
also got very similar autosomal components % in both ftdna and geno2 .
regards
adam

He is mixed, his first reference population is Iberian (Spanish and Portuguese) because there are no Romani or Serb populations on Geno 2.0.

A Norfolk L-M20
01-07-2017, 02:35 PM
Speaking as an outsider, a non Geno customer. Are you sure that any auDNA tests go further back than a few centuries? Regardless of business claims, are they not all subject to genetic recombination? Although I agree that older population admixture will show up regardless, I'm skeptical of claims by any of the businesses that they can reliably go back to say, "500 years ago". I hear all of these claims that one auDNA test for ancestry "focuses" on one period or another, like a telephoto lens with a set depth of field.

jortita
01-07-2017, 02:41 PM
Genographic Geno 2.0 is a deep ancestry test and also shows migration path from africa. My results on genographic are definitely deep ancestry, even if you look at the reference populations, the averages are deep ancestry as a northern han Chinese on a modern audna test such as 23andme will show as 100% east asian while genographic reference results includes 9% Finland and Northern Siberia which is more ancient and ANE related

kingjohn
01-07-2017, 02:54 PM
He is mixed, his first reference population is Iberian (Spanish and Portuguese) because there are no Romani or Serb populations on Geno 2.0.

agree that he is mixed
but that doesn't contradict the fact that his
autosomal components % are so close ...... ;)
here have a look :
his ftdna my origins results :
https://s24.postimg.org/8vvz95bqt/Ftdnamyorigins.jpg

his geno2 next generation results :
http://s29.postimg.org/ba2n49wl3/Natgeo_Regional.png
regards
adam

p.s

by they way it is not surprising because both geno2 and ftdna are partners
and geno2 probably uses ftdna labratory to test there clients .....

Missouri1455
01-07-2017, 02:58 PM
So as I understand it, our MyOrigins might be different if we had just tested with FTDNA alone, and it might be more accurate?

You should try it out; because the ethnic percentages will vary significantly from your geno 2.0 results.

The ethnic percentage you receive from FTDNA will be also very close to what you'll get in 23ndme, and ancestry dna; because these three companies tend to use very similar calculators

jortita
01-07-2017, 03:01 PM
This is the link with reference population for geno 2.0 next gen, Chinese are shown having 8% central Asia https://genographic.nationalgeographic.com/reference-populations-next-gen/,

jortita
01-07-2017, 03:02 PM
Thank you hman, I was under the impression ancestry DNA is deep ancestry and not recent ancestry

Larth
01-07-2017, 03:07 PM
agree that he is mixed
but that doesn't contradict the fact that his
autosomal components % are so close ...... ;)
here have a look :
his ftdna my origins results :
https://s24.postimg.org/8vvz95bqt/Ftdnamyorigins.jpg

his geno2 next generation results :
http://s29.postimg.org/ba2n49wl3/Natgeo_Regional.png
regards
adam .

What do you mean "so close"?

Missouri1455
01-07-2017, 03:11 PM
I think it will be identical, only FTDNA will give less accurate maps for what each region means and no background reference info.

I doubt it will be identical because I noticed ftdna/23ndme tend to calculate result similarly to gedmatch.

For example in Geno 2.0 i had 20% east african yet when i transferred my results to gedmatch; ill barley have 1 african cluster out of 20; which got me thinking that my minority admixture might have been significantly inflated.

So when i get my ftdna results ill definitely transfer to gedmatch; i wouldn't rely on ftdna alone.

Missouri1455
01-07-2017, 03:37 PM
Speaking as an outsider, a non Geno customer. Are you sure that any auDNA tests go further back than a few centuries? Regardless of business claims, are they not all subject to genetic recombination? Although I agree that older population admixture will show up regardless, I'm skeptical of claims by any of the businesses that they can reliably go back to say, "500 years ago". I hear all of these claims that one auDNA test for ancestry "focuses" on one period or another, like a telephoto lens with a set depth of field.
To be honest i think it all goes back to how they ''calculate'' their results; geno 2.0 tend to use a different calculation method than ftdna, ancestry, and 23ndme; thats why its results differ.

kingjohn
01-07-2017, 03:50 PM
What do you mean "so close"?

the numbers .... look at the numbers
the only diffrence is that he score 3% south east asia in geno2
and 1% north east asia in ftdna
but the other components numbers are almost the same like they copy it ......
regards
adam

wombatofthenorth
01-07-2017, 07:35 PM
Speaking as an outsider, a non Geno customer. Are you sure that any auDNA tests go further back than a few centuries? Regardless of business claims, are they not all subject to genetic recombination? Although I agree that older population admixture will show up regardless, I'm skeptical of claims by any of the businesses that they can reliably go back to say, "500 years ago". I hear all of these claims that one auDNA test for ancestry "focuses" on one period or another, like a telephoto lens with a set depth of field.

They mostly pick up only what you are carrying from a few gens back since stuff gets washed away by recombination and random inheritance. But they can use difference things to compare to and can look at various ages.
Look at Geno 2.0 (original) and it clearly looks way, way beyond 500 years for the basic component results, it even calls them basic ancient components. People even from the Baltics get lots of basic ancient Mediterranean and basic ancient Southwest Asia. But it makes sense since the farmers spread all across Europe and carried their genes with them. You can then compare the ratios of the basic ancient components to what various modern reference populations got and you'll see that say Scandinavians get a lot more Northern European basic ancient component than Mediterranean basic ancient component. It's like lots of the stuff on the GEDMATCH, the component scores, not talking Oracles, are hardly just based on the last 500 years, some of them appear to look and compare to stuff from thousand upon thousand of years ago, lots of GEDMATCH tests also seem to do a tricky thing of using very mixed components by region and age all at once and sometimes not complete basis either so I think it makes the basic component scores a bit unstable. The Oracles then sort of compare the component scores to how modern populations get in terms of ratios and try fitting in single choice or 2 ,3 ,4 choices at once mixes.

23 is a bit different in that they try to get away from the need to compare ratios of components to modern reference populations by using long blocks that somewhat filter results to gene patterns of just the last 500 years so you are supposed to be able to just look at the results and read off directly and compare to a modern family tree. It works out to varying degrees. Clearly it doesn't work out 100% since the chip tests only so many SNPS, they have onyl so many refernece samples, and many parts of the world are too mish-mashed. One thing you particularly encountered was the ratio spread for British & Irish to French & German and Broadly Northwestern European where it seems the more Irish you are the more the test works out directly, otherwise you sort of have to read it as a gradient and, as well, be left totally uncertain if you have no idea as to whether you might also have real reent French or German and such in addition to English and so on or not. You still, especially in some cases, have to look at 23 and try to find enough results people get from a region and compare your ratios, like how it seems no Italians ever get more than 70% so you have to realize that you can be 100% family tree Italian for the last 6 gens and still get maybe only 30%-70% Italian score, etc.

wombatofthenorth
01-07-2017, 07:38 PM
You should try it out; because the ethnic percentages will vary significantly from your geno 2.0 results.

The ethnic percentage you receive from FTDNA will be also very close to what you'll get in 23ndme, and ancestry dna; because these three companies tend to use very similar calculators

Hmm FTDNA actually uses the same calculator system that Geno 2.0 NG does, since they had partnered together when they first made the Geno 2.0 NG test not just with having FTDNA carryout the physical test but with Geno taking on FTDNA's calculator code, and it's very different from the one 23 uses.

I wonder if you are confusing Geno 2.0 with Geno 2.0 NG.

wombatofthenorth
01-07-2017, 07:40 PM
I have a feeling that this thread has become a confused mess because of the similar names between Geno 2.0 and Geno 2.0 NG which are two entirely different tests but it seems like in this thread they are being treated as if they were both the same test at times.

wombatofthenorth
01-07-2017, 07:43 PM
Geno 2.0 is actually extremely accurate, however it uses fewer and more ancient components than the other commercial tests so the results might seem puzzling if you don't compare to their modern reference population samples or realize that they are looking at the input you got from thousands of years ago. It uses an extensive private database, just a few stable and complete basic ancient components.

wombatofthenorth
01-07-2017, 07:48 PM
For the time being FTDNA MyOrigins and Geno 2.0 NG appear to be often identical with results although sometimes the labels have a slightly different name and the maps show each component covering a different area (from what I have seen the maps on Geno 2.0 NG show much more accurately the region that the component covers than the maps on FTDNA, one case in point is that the Eastern European/Eastern Europe component is shows on FTDNA to just manage to edge onto the Baltics while the Geno 2.0 NG maps shows the Baltics solidly shaded by that component and in reality the component actually peaks in the Baltics so the weak, fringe shading FTDNA shows is less accurate; another example is how Finnish and Siberian/Finland and Northern Siberia is shown on FTDNA to sharply peak right on Finland and that is about it while Geno 2.0 NG maps shows it cover Finland plus the northern Baltics plus all the across all of Siberia and northern Russia which also seems to be a lot more accurate as to what it actually picks up from). Geno 2.0 NG also gives you tons of modern reference populations ratios so you have some context within which to try to put the results.

Larth
01-07-2017, 08:43 PM
the numbers .... look at the numbers
the only diffrence is that he score 3% south east asia in geno2
and 1% north east asia in ftdna
but the other components numbers are almost the same like they copy it ......
regards
adam

Got it. Yes, I agree with you.

Missouri1455
01-07-2017, 09:45 PM
For the time being FTDNA MyOrigins and Geno 2.0 NG appear to be often identical with results although sometimes the labels have a slightly different name and the maps show each component covering a different area (from what I have seen the maps on Geno 2.0 NG show much more accurately the region that the component covers than the maps on FTDNA, one case in point is that the Eastern European/Eastern Europe component is shows on FTDNA to just manage to edge onto the Baltics while the Geno 2.0 NG maps shows the Baltics solidly shaded by that component and in reality the component actually peaks in the Baltics so the weak, fringe shading FTDNA shows is less accurate; another example is how Finnish and Siberian/Finland and Northern Siberia is shown on FTDNA to sharply peak right on Finland and that is about it while Geno 2.0 NG maps shows it cover Finland plus the northern Baltics plus all the across all of Siberia and northern Russia which also seems to be a lot more accurate as to what it actually picks up from). Geno 2.0 NG also gives you tons of modern reference populations ratios so you have some context within which to try to put the results.

im not burshing off geno 2.0 but they clearly inflated my east african side, and all im looking for is an accurate result rather than an inflated one.

in many of my gedmatch pop. distances i only end up having one or none african cluster out of 20; unless its an african focused admixture like puntdnal then the african cluster will increase. if want examples i can show you what i mean.

wombatofthenorth
01-08-2017, 01:49 AM
im not burshing off geno 2.0 but they clearly inflated my east african side, and all im looking for is an accurate result rather than an inflated one.

in many of my gedmatch pop. distances i only end up having one or none african cluster out of 20; unless its an african focused admixture like puntdnal then the african cluster will increase. if want examples i can show you what i mean.

To be clear are you talking about Geno 2.0 or Geno 2.0 NG?
And keep in mind that Geno 2.0 would be looking at the East African input of your region many, many thousands of years ago.

Missouri1455
01-08-2017, 01:59 AM
To be clear are you talking about Geno 2.0 or Geno 2.0 NG?
And keep in mind that Geno 2.0 would be looking at the East African input of your region many, many thousands of years ago.

Im looking at geno 2.0 NG because thats where i got my results from; but my results don't match up with the population distance i was given by gedmatch; maybe east african is based on many years ago but then again i was told by a person who works there, these results barley go back more than 100-200 years; at times they might pick up an admix from the last 1000 years but their "ancient" focus as the guy told me was actulley the y-dna and mt-dna subclades.

wombatofthenorth
01-08-2017, 06:49 AM
Im looking at geno 2.0 NG because thats where i got my results from; but my results don't match up with the population distance i was given by gedmatch; maybe east african is based on many years ago but then again i was told by a person who works there, these results barley go back more than 100-200 years; at times they might pick up an admix from the last 1000 years but their "ancient" focus as the guy told me was actulley the y-dna and mt-dna subclades.

hmm so it is NG then
I'm guessing it is just picking up mostly from your ancestors of the last 300 years, but that many of them maybe have ancient population wide signals of East Africa in that region??
I forget, do they have a reference sample for Yemen or a close by region? I wonder if they tend to avg a lot of East Africa or not.

Missouri1455
01-08-2017, 04:08 PM
hmm so it is NG then
I'm guessing it is just picking up mostly from your ancestors of the last 300 years, but that many of them maybe have ancient population wide signals of East Africa in that region??
I forget, do they have a reference sample for Yemen or a close by region? I wonder if they tend to avg a lot of East Africa or not.

There definitly is an ancestry there but its been inflated; geno 2.0 refrenced me with iranian as 1st reference and lebanese as 2nd reference; and iranians and lebanese avg 0%-4% SSA.

Overall Geno 2.0 reference populations tend to be more accurate with their european populations because they do give an insight of where you might come from but in my case they usually lump up iran with the levant and arabian peninsula so the reference population i was given might signal that my admixture is mostly from arabian peninsula and levant.

These are population distances from gedmatch
Keep in mind that geno 2.0 gave me 20% East African in my results; so here's some examples for my SSA clusters out of 20:

MLDP k23b: 0(SSA)
MLDP world 22: 1(SSA)
Harrapaworld(it serves best with people who have east african ethnicity): 0(SSA)
Deocad V3: 0(SSA)
Deocad k12b: 0(SSA)
PuntDnal k10: 2(SSA)


Also these african clusters i have tend to be listed at 21.00-35.00 which is very distant so if i had any east african admixture it will most likely be low.

I also read somewhere that the population distance loses it relevancy once it passes 16.00; not sure how accurate that may be but there you have it.

At the moment they dont have yemen as a reference population; but in yemen there's SSA mostly in the coastal areas, once you move inland that number dwindles down; if i have any east african i would deem it to be distant in the same way i was shown on gedmatch population distance and geno 2.0 reference populations; and any west african i was given by geno 2.0 will most likely not be there because it was absent in almost all of my population distances except in puntdnal k8 and also it is also absent in my 2 reference population.

raspberry
01-08-2017, 05:16 PM
There definitly is an ancestry there but its been inflated; geno 2.0 refrenced me with iranian as 1st reference and lebanese as 2nd reference; and iranians and lebanese avg 0%-4% SSA.

Yes, but somehow "Iranian" in Geno 2.0 means: 56% Arabian (biggest joke (or irony?) ever), 4% Eastern Africa, 2% Northern Africa, 4% Central Asia,6% Asia minor, 2% Southern Europe and 24% Southern Asia. I think your "Arabian" might match it the most with this "population" and thus they show this as your population. Probably they took an ahwazi and labeled him as a "Iranian" :D.
https://genographic.nationalgeographic.com/reference-populations-next-gen/

Missouri1455
01-08-2017, 06:33 PM
Yes, but somehow "Iranian" in Geno 2.0 means: 56% Arabian (biggest joke (or irony?) ever), 4% Eastern Africa, 2% Northern Africa, 4% Central Asia,6% Asia minor, 2% Southern Europe and 24% Southern Asia. I think your "Arabian" might match it the most with this "population" and thus they show this as your population. Probably they took an ahwazi and labeled him as a "Iranian" :D.
https://genographic.nationalgeographic.com/reference-populations-next-gen/

thats true, maybe thats what they meant when they listed "iranian" as my 1st reference population; geno 2.0 tend to generalize when it comes to middle eastern populations rather than being specific; so us middle easterners have to do extra work to get to the bottom of things;)

kingjohn
01-08-2017, 07:25 PM
dear hman92
do you score north east african
or sub sharan in eurogenes k13 calculator how much ?
best regards
adam

p.s
i think it is a good reliable calculator :)

Missouri1455
01-08-2017, 08:01 PM
dear hman92
do you score north east african
or sub sharan in eurogenes k13 calculator how much ?
best regards
adam

p.s
i think it is a good reliable calculator :)

yes i do but the ethnic percentage resembles geno 2.0 results because the results are pretty much a "transfer" and won't mean much unless i redo the test with another dna company. the only way i managed to detect inflation of my results was by three ways:

1. geno 2.0 calculates results differently from other dna companies.

2. gedmatch population distance failed to detect any african clusters and if it detected any they range between 21.00-35.00 which is considered very distant from my actual ethnic result; and if any east african pop. was detected it would barley be more than one pop. unless the admix was "african" focused like puntdnal.

3. Geno 2.0 usually lumps up iran with the levant and arabian peninsula so the reference population i was given most likely meant kuwait and Lebanon rather than iran and lebanon so my admixture would most likely be composed from arabian peninsula and levant.

anyways here's my Eurogene K13 results:

1 East_Med 35.64
2 Red_Sea 25.46
3 Northeast_African 15.49
4 Sub-Saharan 11.38
5 West_Asian 4.13
6 North_Atlantic 2.53
7 West_Med 1.81
8 South_Asian 1.62
9 Siberian 1.3
10 Oceanian 0.38
11 East_Asian 0.26

Even though I have SSA of approx. 27%
The population distance below failed to detect any SSA pop.

# Population (source) Distance
1 Egyptian 13.91
2 Bedouin 18.99
3 Saudi 21.53
4 Yemenite_Jewish 22.64
5 Jordanian 23.12
6 Tunisian 23.45
7 Palestinian 23.79
8 Algerian 23.79
9 Mozabite_Berber 25.41
10 Moroccan 25.56
11 Syrian 27.66
12 Libyan_Jewish 28.2
13 Tunisian_Jewish 29.29
14 Samaritan 29.76
15 Lebanese_Muslim 30.65
16 Lebanese_Christian 31.96
17 Lebanese_Druze 32.44
18 Algerian_Jewish 32.83
19 Sephardic_Jewish 32.96
20 Italian_Jewish 34.12

As you can see i have 0 SSA again and this have happend repeatedly which pushed me to apply for ftdna kit.

ArmandoR1b
01-08-2017, 08:17 PM
Nevermind

ArmandoR1b
01-08-2017, 08:27 PM
1. geno 2.0 calculates results differently from other dna companies.
All of the companies calculate results differently from each other except that FTDNA and Geno 2.0 NextGen in the U.S. used to use the same calculator. They all test different autosomal SNPs too. Who else have you tested with?



anyways here's my Eurogene K13 results:

1 East_Med 35.64
2 Red_Sea 25.46
3 Northeast_African 15.49
4 Sub-Saharan 11.38
5 West_Asian 4.13
6 North_Atlantic 2.53
7 West_Med 1.81
8 South_Asian 1.62
9 Siberian 1.3
10 Oceanian 0.38
11 East_Asian 0.26

Even though I have SSA of approx. 27%
The population distance below failed to detect any SSA pop.
# Population (source) Distance
1 Egyptian 13.91
2 Bedouin 18.99
3 Saudi 21.53
4 Yemenite_Jewish 22.64
5 Jordanian 23.12
6 Tunisian 23.45
7 Palestinian 23.79
8 Algerian 23.79
9 Mozabite_Berber 25.41
10 Moroccan 25.56
11 Syrian 27.66
12 Libyan_Jewish 28.2
13 Tunisian_Jewish 29.29
14 Samaritan 29.76
15 Lebanese_Muslim 30.65
16 Lebanese_Christian 31.96
17 Lebanese_Druze 32.44
18 Algerian_Jewish 32.83
19 Sephardic_Jewish 32.96
20 Italian_Jewish 34.12

What do you get with Oracle-4 results in Eurogenes K13? Do you know how to find them?


As you can see i have 0 SSA again and this have happend repeatedly which pushed me to apply for ftdna kit.
It's always good to have results from multiple companies.

Missouri1455
01-08-2017, 10:14 PM
All of the companies calculate results differently from each other except that FTDNA and Geno 2.0 NextGen in the U.S. used to use the same calculator. They all test different autosomal SNPs too. Who else have you tested with?

Only geno; i think geno and ftdna use the same calculator but test different autosomal snps; because ftdna results tend to be close 23ndme at least thats what i witnessed.


What do you get with Oracle-4 results in Eurogenes K13? Do you know how to find them?

1 East_Med 35.64
2 Red_Sea 25.46
3 Northeast_African 15.49
4 Sub-Saharan 11.38
5 West_Asian 4.13
6 North_Atlantic 2.53
7 West_Med 1.81
8 South_Asian 1.62
9 Siberian 1.30


SSA at approx. 27%
1 SSA detected at 34.77
34.77 is too distant to make any significant impact.

1 Egyptian @ 14.650436
2 Bedouin @ 19.327372
3 Saudi @ 21.025148
4 Yemenite_Jewish @ 23.236980
5 Jordanian @ 24.185453
6 Palestinian @ 24.514965
7 Tunisian @ 25.612926
8 Algerian @ 26.223650
9 Mozabite_Berber @ 28.058939
10 Moroccan @ 28.327612
11 Syrian @ 28.906326
12 Libyan_Jewish @ 29.343403
13 Tunisian_Jewish @ 30.448544
14 Samaritan @ 30.906591
15 Lebanese_Muslim @ 32.125290
16 Lebanese_Christian @ 33.357506
17 Lebanese_Druze @ 33.921894
18 Sephardic_Jewish @ 34.316078
19 Algerian_Jewish @ 34.376305
20 Ethiopian_Tigray @ 34.772373



It's always good to have results from multiple companies.

I agree

ArmandoR1b
01-08-2017, 10:25 PM
Only geno; i think geno and ftdna use the same calculator but test different autosomal snps; because ftdna results tend to be close 23ndme thats what i witnessed.
Most people get significantly different results between FTDNA and 23andme. How many people's 23andme results have you seen?

As far as the Oracle-4 results in Eurogenes K13, I forgot to add that I was looking for the 3 populations approximation and the 4 populations approximation. Will you post those?

Missouri1455
01-08-2017, 10:44 PM
Most people get significantly different results between FTDNA and 23andme. How many people's 23andme results have you seen?

As far as the Oracle-4 results in Eurogenes K13, I forgot to add that I was looking for the 3 populations approximation and the 4 populations approximation. Will you post those?

I've seen quite a few can you show me the differences you witnessed between the two

3 pop. approximation:
1 50% Saudi +25% Sudanese +25% Yemenite_Jewish @ 5.391012

4 pop. approximation:
1 Sandawe + Saudi + Saudi + Yemenite_Jewish @ 5.048526
2 Saudi + Saudi + Sudanese + Yemenite_Jewish @ 5.391012
3 Saudi + Sudanese + Yemenite_Jewish + Yemenite_Jewish @ 5.750894
4 Sandawe + Saudi + Yemenite_Jewish + Yemenite_Jewish @ 5.756707

These SSA approximations are directly tied to the ethnic percentage; because it doesnt match with gedmatch distance results nor the reference populations.

khanabadoshi
01-15-2017, 03:00 PM
agree that he is mixed
but that doesn't contradict the fact that his
autosomal components % are so close ...... ;)
here have a look :
his ftdna my origins results :
https://s24.postimg.org/8vvz95bqt/Ftdnamyorigins.jpg

his geno2 next generation results :
http://s29.postimg.org/ba2n49wl3/Natgeo_Regional.png
regards
adam

p.s

by they way it is not surprising because both geno2 and ftdna are partners
and geno2 probably uses ftdna labratory to test there clients .....


Genographic Geno 2.0 is a deep ancestry test and also shows migration path from africa. My results on genographic are definitely deep ancestry, even if you look at the reference populations, the averages are deep ancestry as a northern han Chinese on a modern audna test such as 23andme will show as 100% east asian while genographic reference results includes 9% Finland and Northern Siberia which is more ancient and ANE related

I compared the 23andMe V4, with FTDNA converted results of Geno 2.0 and Geno 2.0 NextGen on a few calculators to see if there is any substantial difference. I am looking for my Geno 2.0 raw data to compare it with the FTDNA converted data, can't seem to find it though. I'll update it when/if I do. The converted results seem to align well against 23andMe and each other:



HARRAPAWORLD
23andMe V4

FTDNA - Geno 2.0
FTDNA - Geno 2.0 NextGen


GEDROSIA ANE K6
23andMe V4

FTDNA - Geno 2.0
FTDNA - Geno 2.0 NextGen


S-Indian
28.14
27.81
27.79

Ancestral_North_Eurasian
42.41
42.08
42.06


Baloch
38.06
38.11
38.11

Ancestral_South_Eurasian
17.89
17.60
17.36


Caucasian
16.17
16.42
16.44

East_Asian
9.37
8.91
9.08


NE-Euro
8.15
7.93
7.95

West_European_Hunter_Gartherer
0.00
0.00
0.00


SE-Asian
0.00
0.60
0.62

Natufian
30.18
31.03
31.12


Siberian
4.04
3.88
3.90

Sub_Saharan
0.15
0.38
0.38


NE-Asian
1.01
0.94
0.93







Papuan
1.17
1.22
1.13







American
0.00
0.00
0.00







Beringian
0.00
0.00
0.00







Mediterranean
0.78
0.61
0.64







SW-Asian
2.12
2.29
2.32







San
0.07
0.20
0.18







E-African
0.00
0.00
0.00







Pygmy
0.00
0.00
0.00







W-African
0.24
0.00
0.00


















PUNTDNAL K10

23andMe V4
FTDNA - Geno 2.0
FTDNA - Geno 2.0 NextGen

MDLP K11

23andMe V4
FTDNA - Geno 2.0
FTDNA - Geno 2.0 NextGen


ASI
36.28
37.75
37.60

African
0.00%
0.19%
0.19%


Sub-Saharan
0.29
0.76
0.81

Amerindian
1.43%
0.99%
1.02%


Oceanian
1.54
1.84
1.86

ASI
38.04%
35.96%
35.93%


Beringian
0.15
0.3
0.38

Basal
9.95%
9.99%
10.05%


ENF
11.14
11.69
11.73

Iran-Mesolithic
6.60%
7.00%
6.93%


CHG
35.41
33.7
33.66

Neolithic
1.07%
1.56%
1.67%


Siberian
4.81
3.29
3.33

Oceanic
0.14%
0.53%
0.46%


E_Asian
0.59
1.96
1.88

EHG
32.52%
32.98%
32.93%


WHG
8.17
7.55
7.57

SEA
1.88%
3.16%
3.24%


Amerindian
1.63
1.17
1.17

Siberian
3.11%
2.68%
2.67%







WHG
5.23%
4.95%
4.90%


MDLP K16

23andMe V4
FTDNA - Geno 2.0
FTDNA - Geno 2.0 NextGen







Amerindian
1.25%
0.62%
0.65%







Ancestor
0.15%
0.00%
0.00%







Steppe
10.71%
10.87%
10.65%







Indian
47.92%
48.18%
48.21%







Arctic
0.06%
0.09%
0.08%







Australian
0.92%
0.00%
0.01%







Caucasian
25.91%
24.87%
24.90%







EastAfrican
0.00%
0.35%
0.33%







NorthEastEuropean
2.60%
2.92%
2.91%







NearEast
1.58%
2.19%
2.27%







Neolithic
0.44%
0.62%
0.67%







NorthAfrican
0.00%
0.00%
0.00%







Oceanic
1.02%
2.35%
2.31%







Siberian
3.32%
2.99%
2.95%







SouthEastAsian
4.11%
3.95%
4.06%







Subsaharian
0.00%
0.00%
0.00%

wombatofthenorth
01-16-2017, 05:41 AM
How did you get Geno 2.0 to upload to GEDMATCH? I know Geno 2.0 NG you can transfer to FTDNA and get FTDNA format data to upload to GEDMATCH but Geno 2.0 you couldn't do that with.
Do you create a custom converter to put the Geno 2.0 RAW data into FTDNA format?

khanabadoshi
01-16-2017, 09:07 AM
How did you get Geno 2.0 to upload to GEDMATCH? I know Geno 2.0 NG you can transfer to FTDNA and get FTDNA format data to upload to GEDMATCH but Geno 2.0 you couldn't do that with.
Do you create a custom converter to put the Geno 2.0 RAW data into FTDNA format?

Way back, I actually paid the 35 dollars for the autosomal transfer because I didn't know what I was doing. However, since then I've figured out that you can convert the Geno 2.0 raw data to FTDNA format using the Aconv tool (http://www.y-str.org/2013/06/autosomal-dna-converter.html) (converting to 23andMe format "works", but fails in R). The file is tiny and won't work on Gedmatch. However, it works perfectly fine for DIY calculators on R. If you aren't familiar with the Aconv tool, I can convert the file for you and try the MDLP calcs if you like?

I found my Geno 2.0 raw data. Here is the comparison. Honestly, if you have only Geno 2.0 it's totally worth it to convert the file so you can use it on R. The results aren't wildly off.




MDLP K11
23andMe V4
FTDNA - Geno 2.0
FTDNA - Geno 2.0 NextGen
GENO 2.0 RAW DATA


African
0.00%
0.19%
0.19%
0.20%



Amerindian
1.43%
0.99%
1.02%
0.33%


ASI
38.04%

35.96%
35.93%
38.19%



Basal
9.95%
9.99%
10.05%
10.75%


Iran-Mesolithic
6.60%
7.00%
6.93%
7.63%



Neolithic
1.07%
1.56%
1.67%
0.55%


Oceanic
0.14%
0.53%
0.46%
0.00%


EHG
32.52%

32.98%
32.93%
31.96%



SEA
1.88%
3.16%
3.24%
4.10%



Siberian
3.11%
2.68%
2.67%
2.86%


WHG
5.23%

4.95%
4.90%
3.42%







MDLP K16
23andMe V4
FTDNA - Geno 2.0
FTDNA - Geno 2.0 NextGen
GENO 2.0 RAW DATA



Amerindian
1.25%
0.62%
0.65%
0.51%



Ancestor
0.15%
0.00%
0.00%
0.00%


Steppe
10.71%
10.87%
10.65%
10.58%


Indian
47.92%

48.18%
48.21%
48.35%


Arctic
0.06%
0.09%
0.08%
0.00%


Australian
0.92%
0.00%
0.01%
0.62%


Caucasian
25.91%

24.87%
24.90%
27.91%



EastAfrican
0.00%
0.35%
0.33%
0.00%


NorthEastEuropean
2.60%

2.92%
2.91%

0.89%


NearEast
1.58%

2.19%
2.27%
0.61%


Neolithic
0.44%
0.62%
0.67%
0.03%


NorthAfrican
0.00%
0.00%
0.00%
0.00%


Oceanic
1.02%
2.35%
2.31%
1.48%


Siberian
3.32%

2.99%

2.95%
3.57%


SouthEastAsian
4.11%

3.95%
4.06%
5.44%


Subsaharian
0.00%
0.00%
0.00%
0.00%

Amerijoe
01-16-2017, 12:59 PM
Having used the four major testing companies, here is how I would list their results.

23andMe
FTDNA
Ancestrydna
Geno 2 NG

Geno is last due to two points, YDNA haplogroup should have been closer, mtdna was wrong. Repeated efforts to correct with established defining snps results, we're right, end of story. Graphics are nice.

wombatofthenorth
01-17-2017, 12:09 AM
Having used the four major testing companies, here is how I would list their results.

23andMe
FTDNA
Ancestrydna
Geno 2 NG

Geno is last due to two points, YDNA haplogroup should have been closer, mtdna was wrong. Repeated efforts to correct with established defining snps results, we're right, end of story. Graphics are nice.

For me:
1. Geno 2.0 + Geno 2.0 NG
2. 23
3. FTDNA (although special extras you can do on FTDNA are good, like if you eventually want to say do full mtDNA that is good as a special test)

wombatofthenorth
01-17-2017, 12:15 AM
Way back, I actually paid the 35 dollars for the autosomal transfer because I didn't know what I was doing. However, since then I've figured out that you can convert the Geno 2.0 raw data to FTDNA format using the Aconv tool (http://www.y-str.org/2013/06/autosomal-dna-converter.html) (converting to 23andMe format "works", but fails in R). The file is tiny and won't work on Gedmatch. However, it works perfectly fine for DIY calculators on R. If you aren't familiar with the Aconv tool, I can convert the file for you and try the MDLP calcs if you like?

I found my Geno 2.0 raw data. Here is the comparison. Honestly, if you have only Geno 2.0 it's totally worth it to convert the file so you can use it on R. The results aren't wildly off.



Thanks I already have Geno 2.0 NG though so I don't need to convert the Geno 2.0.
It is interesting that it produces pretty similar results despite 7x less total SNPs and probably 3x less ancestry SNPs.

I guess you Geno 2.0 FTDNA transfer was probably them just running a complete FF test for you (at a discount) and probably didn't use the Geno 2.0 results at all so your Geno 2.0 FTDNA table I bet could really be called a FTDNA FF table.

It seems FTDNA and the Geno 2.0 NG data produce pretty similar results even more similar to each other than to 23. That probably makes sense since they are likely based on the same chip with the Geno 2.0 NG with just some number of custom extras and the 23 v4 chip perhaps have less total ancestry SNPs although still having many in common. It is insteresting though that the probable pure FF test raw data does seem to be different than Geno 2.0 NG so I guess that does confirm the Geno 2.0 NG chip did add some number of custom autosomal SNPs.