PDA

View Full Version : Possible new crowdfunding test



Chad Rohlfsen
07-07-2015, 06:24 PM
I am working on a couple of runs, which I will post here and run by everyone. I am waiting to hear David's ideas about a crowdfunding run. It should be pretty informative and possibly lead off into other directions, with Oracles and R files. What does everyone think? If you're interested in possibly being involved, give this a like. I'm curious about a head count. Thanks!

J Man
07-07-2015, 09:27 PM
I am working on a couple of runs, which I will post here and run by everyone. I am waiting to hear David's ideas about a crowdfunding run. It should be pretty informative and possibly lead off into other directions, with Oracles and R files. What does everyone think? If you're interested in possibly being involved, give this a like. I'm curious about a head count. Thanks!

I gave it a like as you can see and I am definitely very interested! :D

Moderator
07-08-2015, 07:09 AM
[MOD] Thread moved from auDNA to Crowdfunding Initiatives section, 1 day redirect and OP informed - Thanks all for your cooperation

Varun R
07-09-2015, 06:46 AM
Please do it Chad. Will it convince you more if I say "pretty please with a cherry on top?" :)

Chad Rohlfsen
07-11-2015, 12:28 AM
Please do it Chad. Will it convince you more if I say "pretty please with a cherry on top?" :)

It's in progress. I've basically got my computer running Admixture 24/7. There's a real possibility that I could have it ready in the next few days.

J Man
07-11-2015, 01:34 PM
It's in progress. I've basically got my computer running Admixture 24/7. There's a real possibility that I could have it ready in the next few days.

Amazing stuff Chad!

J Man
07-14-2015, 12:29 PM
It's in progress. I've basically got my computer running Admixture 24/7. There's a real possibility that I could have it ready in the next few days.

Any updates on this at all Chad?

Chad Rohlfsen
07-14-2015, 04:19 PM
I'm just trying to get the most separation of EEF and WHG, as I can. Everything else is ready.

John Doe
07-14-2015, 04:38 PM
I'm just trying to get the most separation of EEF and WHG, as I can. Everything else is ready.

Sweet, will it be available for everyone who donated to Eurogenes?

Chad Rohlfsen
07-14-2015, 08:38 PM
This is a whole new deal, from myself. David will be assisting, as he is the only one that knows how to make the calculator at this time. I believe that he is going to show me how to run 23andme and FTDNA files, so genomes can be sent to both of us. So, there will be a new charge, which I am splitting with David. I may come up with a separate one for SC and S Asians, with components from Afanasievo, Sintashta, and Andronovo. Let me finish this one here, I can post the results of the samples I have and go from there.

J Man
07-14-2015, 09:29 PM
This is a whole new deal, from myself. David will be assisting, as he is the only one that knows how to make the calculator at this time. I believe that he is going to show me how to run 23andme and FTDNA files, so genomes can be sent to both of us. So, there will be a new charge, which I am splitting with David. I may come up with a separate one for SC and S Asians, with components from Afanasievo, Sintashta, and Andronovo. Let me finish this one here, I can post the results of the samples I have and go from there.

Who are ''SC"? South Central Asians?

Chad Rohlfsen
07-14-2015, 11:29 PM
Here is the one without WHG. It made a component that is actually 8% more WHG than Spain MN, and it looks about 44% WHG, 56% Stuttgart. It appears to cover all WHG in the Baltic folks and works pretty well for several Bronze Age samples. I am working on adding a WHG component, and EEF. Spain MN looks about 36% WHG, 64% EEF. It looks pretty solid, just separating WHG and EEF from the MN and Near East, is all that remains. Unless, folks are more interested in testing on this run.

https://docs.google.com/spreadsheets/d/17_z7KMnlz2dTTqp5G1Z900Tg-9x8rgM9WzR7jpzaUgs/edit?usp=sharing

Chad Rohlfsen
07-14-2015, 11:29 PM
Who are ''SC"? South Central Asians?

Yes, that's what I mean.

J Man
07-15-2015, 01:07 AM
Here is the one without WHG. It made a component that is actually 8% more WHG than Spain MN, and it looks about 44% WHG, 56% Stuttgart. It appears to cover all WHG in the Baltic folks and works pretty well for several Bronze Age samples. I am working on adding a WHG component, and EEF. Spain MN looks about 36% WHG, 64% EEF. It looks pretty solid, just separating WHG and EEF from the MN and Near East, is all that remains. Unless, folks are more interested in testing on this run.

https://docs.google.com/spreadsheets/d/17_z7KMnlz2dTTqp5G1Z900Tg-9x8rgM9WzR7jpzaUgs/edit?usp=sharing

I like everyone else most likely am/are most interested in testing on the most accurate run possible. So if it takes a bit longer to tweak things to make the most accurate test possible that is fine by me. :)

J Man
07-16-2015, 04:17 PM
Will you post here Chad once the test is ready?

Arbogan
07-21-2015, 06:23 AM
Im all for it and willing to pay a reasonable sum if you include the indo iranian samples and the RISE armenians.

Chad Rohlfsen
07-21-2015, 01:58 PM
Yes, there will be some Andronovo, Sintashta, and Armenians in it, to use with R, or Oracle. I'm just trying to maximize the WHG component. Each run tends to take half a day, that's the reason for the delay. Once I feel like I can't do anymore, I'll post results and work with David on setting up the test.

Chad Rohlfsen
07-21-2015, 02:02 PM
I may set one up to be specific for SC Asians, with an Afanasievo, Andronovo, and Sintashta components.

J Man
07-21-2015, 05:27 PM
Yes, there will be some Andronovo, Sintashta, and Armenians in it, to use with R, or Oracle. I'm just trying to maximize the WHG component. Each run tends to take half a day, that's the reason for the delay. Once I feel like I can't do anymore, I'll post results and work with David on setting up the test.

Chad are you trying to maximize actual true WHG type ancestry that comes from Mesolithic European hunter-gatherers that is not part of the WHG like ancestry that is present in the EEF gene pool since we know that the EEF population and even Anatolian Neolithic farmers had/have a good amount of WHG like ancestry in their genomes?

Chad Rohlfsen
07-21-2015, 05:48 PM
Chad are you trying to maximize actual true WHG type ancestry that comes from Mesolithic European hunter-gatherers that is not part of the WHG like ancestry that is present in the EEF gene pool since we know that the EEF population and even Anatolian Neolithic farmers had/have a good amount of WHG like ancestry in their genomes?

European specific, above EEF.

Chad Rohlfsen
07-22-2015, 09:07 PM
It's getting pretty close. I might be on the final run. As soon as I see it, I'll give an update. It should be done before the end of the weekend, at the latest.

Chad Rohlfsen
07-23-2015, 06:27 PM
Still sorting. EEF is the only tight cluster left.

Chad Rohlfsen
07-25-2015, 04:16 PM
Okay. I'm doing what may be the final run. Hopefully, it's done in the next 12 hours.

Chad Rohlfsen
07-25-2015, 04:42 PM
Just keep your fingers crossed. There can't be more than 1-3 runs left. It should be done by Sunday night. Hopefully, there can be a test offered sometime by the middle of next week.

Don't be alarmed by the low WHG. This makes sense when looking at how little changed from EN to MN. There are a couple MN samples that just about overlap with the EN. Then, we have to consider the dip from MN, when adding in Yamnaya and Near East ancestry.

J Man
07-26-2015, 06:22 PM
Just keep your fingers crossed. There can't be more than 1-3 runs left. It should be done by Sunday night. Hopefully, there can be a test offered sometime by the middle of next week.

Don't be alarmed by the low WHG. This makes sense when looking at how little changed from EN to MN. There are a couple MN samples that just about overlap with the EN. Then, we have to consider the dip from MN, when adding in Yamnaya and Near East ancestry.

So the WHG component in this test will basically be extra European specific WHG that is is excess of the MN component?

Chad Rohlfsen
07-26-2015, 08:17 PM
So the WHG component in this test will basically be extra European specific WHG that is is excess of the MN component?

In excess of EEF. EEF is fairly variable, so putting together the right group is what I'm working on. I have one finishing in about 3 hours. I'm hoping it's done after that.

Chad Rohlfsen
07-26-2015, 08:19 PM
But yes, some came after the MN. That's what I gather with modern pops at about the same number as the MN.

J Man
07-27-2015, 01:52 PM
In excess of EEF. EEF is fairly variable, so putting together the right group is what I'm working on. I have one finishing in about 3 hours. I'm hoping it's done after that.

How did the last run go? Pretty much there?

Chad Rohlfsen
07-27-2015, 02:28 PM
Yeah, very close. It may have to be a couple percent short on whg. The more near eastern ones aren't quite numerous enough to cluster. I'll have to add one or two more with some WHG.

J Man
07-28-2015, 06:02 PM
Yeah, very close. It may have to be a couple percent short on whg. The more near eastern ones aren't quite numerous enough to cluster. I'll have to add one or two more with some WHG.

Can't wait! :)

Chad Rohlfsen
08-02-2015, 02:11 AM
Sorry, for the delay. As soon as I get a good EEF cluster, I end up with a Baltic cluster. After removing some Baltic pops, I get a MN cluster instead of EEF. It's being a pain in the ass. Just giving an update. I'm not giving up here.

J Man
08-04-2015, 02:07 AM
Sorry, for the delay. As soon as I get a good EEF cluster, I end up with a Baltic cluster. After removing some Baltic pops, I get a MN cluster instead of EEF. It's being a pain in the ass. Just giving an update. I'm not giving up here.

Almost ready?

Chad Rohlfsen
08-04-2015, 03:11 AM
I hope so. I'll check this one in the morning.

Chad Rohlfsen
08-05-2015, 03:42 PM
Alright, this is coming together. Now, another question for everyone. Whatever is most wanted, I'll try my best to do. I believe that I can split Yamnaya between EHG and a West Asian type. If everyone likes, I can split Yamnaya, or leave it whole. What does everyone want?

J Man
08-05-2015, 03:45 PM
Alright, this is coming together. Now, another question for everyone. Whatever is most wanted, I'll try my best to do. I believe that I can split Yamnaya between EHG and a West Asian type. If everyone likes, I can split Yamnaya, or leave it whole. What does everyone want?

Great news! Either or is good with me.

John Doe
08-05-2015, 03:59 PM
Alright, this is coming together. Now, another question for everyone. Whatever is most wanted, I'll try my best to do. I believe that I can split Yamnaya between EHG and a West Asian type. If everyone likes, I can split Yamnaya, or leave it whole. What does everyone want?

Are we supposed to send you our raw data?

Chad Rohlfsen
08-05-2015, 05:16 PM
Are we supposed to send you our raw data?

Yes. When it's ready.

John Doe
08-05-2015, 05:52 PM
Yes. When it's ready.

And when is it supposed to be ready?

Chad Rohlfsen
08-05-2015, 05:59 PM
That depends on how this test turns out tonight.

Varun R
08-05-2015, 06:20 PM
Chad, I think it'd be more useful for most members to keep Yamnaya intact.

John Doe
08-05-2015, 06:32 PM
Yes. When it's ready.

What's your email address?

Chad Rohlfsen
08-05-2015, 06:49 PM
What's your email address?

[email protected]

John Doe
08-05-2015, 07:51 PM
[email protected]
Thanks, will it cost anything?

Chad Rohlfsen
08-05-2015, 08:04 PM
I think we're going for $20. This project is to help fund the purchase of new programs to analyze ancient genomes and their admixture. I may end up taking all information at another e-mail that's not as busy. Let's go with [email protected]

J Man
08-05-2015, 08:05 PM
Chad, I think it'd be more useful for most members to keep Yamnaya intact.

Thinking about this a bit more I agree. I would also like to see Yamnaya intact. Perhaps after the first test is created later on Chad can make another test with the EHG and West Asian type split.

ZephyrousMandaru
08-05-2015, 08:10 PM
I'm interested, but I'd like to see a sample test. Would that be possible to run?

Chad Rohlfsen
08-05-2015, 08:30 PM
I'll post a run with a few hundred samples and then post my results.

John Doe
08-06-2015, 07:29 AM
I think we're going for $20. This project is to help fund the purchase of new programs to analyze ancient genomes and their admixture. I may end up taking all information at another e-mail that's not as busy. Let's go with [email protected]

And what if I already donated 20$ to David? Will I have to donate again? And if so at the same place (via paypal)?

J Man
08-06-2015, 05:05 PM
I'll post a run with a few hundred samples and then post my results.

I suppose that the BedouinB Yamnaya issue will need to be resolved first before the test can be offered.

Chad Rohlfsen
08-06-2015, 05:30 PM
And what if I already donated 20$ to David? Will I have to donate again? And if so at the same place (via paypal)?

This is separate from Eurogenes. I'll post my paypal info when I'm ready.

Chad Rohlfsen
08-06-2015, 05:31 PM
I suppose that the BedouinB Yamnaya issue will need to be resolved first before the test can be offered.

We'll see.

Chad Rohlfsen
08-11-2015, 05:51 PM
I'm very close. Thanks, for your patience.

J Man
08-12-2015, 03:36 PM
I'm very close. Thanks, for your patience.

I am looking forward to seeing your trial/test run(s).

Chad Rohlfsen
08-12-2015, 10:44 PM
David is testing out the SNPs, for coverage with 23andme. That'll be the deciding factor on what happens next.

Chad Rohlfsen
08-13-2015, 03:36 AM
SNPs look good. I have two sets that I am trying out. I'm hoping that this is up and running by the weekend.

Ignis90
08-13-2015, 07:57 PM
Any news on the actual crowdfunding part?

J Man
08-13-2015, 08:24 PM
SNPs look good. I have two sets that I am trying out. I'm hoping that this is up and running by the weekend.

Good news! Can you give us a lost of the components that will be part of this test? That would be awesome to see.

Chad Rohlfsen
08-15-2015, 05:25 AM
Sure, I will try to get around to posting some stuff, very soon. I am in the process of moving into a new house this weekend. I'm very busy, but I will try.

J Man
08-15-2015, 10:09 AM
List*.

Chad Rohlfsen
08-21-2015, 10:43 PM
Just so everyone sees where I'm at. I am very close to done. It wont be long now.

https://docs.google.com/spreadsheets/d/1S81m-HdWpnRYtPhF7uubuoETSQSCnAbe2KEuJRNG_yU/edit?usp=sharing

MatAust21
08-23-2015, 07:06 PM
Just so everyone sees where I'm at. I am very close to done. It wont be long now.

https://docs.google.com/spreadsheets/d/1S81m-HdWpnRYtPhF7uubuoETSQSCnAbe2KEuJRNG_yU/edit?usp=sharing

Seems very promising.

I like the idea of having EHG and West Asian separated, instead of a single Yamnaya percentage. The more detailed, the better!

J Man
08-24-2015, 04:11 PM
Seems very promising.

I like the idea of having EHG and West Asian separated, instead of a single Yamnaya percentage. The more detailed, the better!

I would be happy with it either way. It is interesting to see a EHG component for sure though.

Huijbregts
08-26-2015, 03:13 PM
After studying the datasheet, I realized that I had overnight morphed from a WHG into an EHG.
The rule seems to be that every HG which has a Yamnaya-like admixture is defined as EHG, which leaves Loschbour, La Brana and HungaryGamba_HG as WHG.
Isn't it preferable to rename these labels? For instance palaeoHG and neoHG for WHG and EHG.
P.S. I prefer to split Yamnaya between EHG and West Asian.

Dorkymon
08-26-2015, 03:37 PM
Just so everyone sees where I'm at. I am very close to done. It wont be long now.

https://docs.google.com/spreadsheets/d/1S81m-HdWpnRYtPhF7uubuoETSQSCnAbe2KEuJRNG_yU/edit?usp=sharing

Impressive, but is it possible to get a Romanian sample? It's kind of ridiculous seeing for example Slovakians and Slovenians, while a way larger European population is missing.

J Man
08-26-2015, 03:45 PM
After studying the datasheet, I realized that I had overnight morphed from a WHG into an EHG.
The rule seems to be that every HG which has a Yamnaya-like admixture is defined as EHG, which leaves Loschbour, La Brana and HungaryGamba_HG as WHG.
Isn't it preferable to rename these labels? For instance palaeoHG and neoHG for WHG and EHG.
P.S. I prefer to split Yamnaya between EHG and West Asian.

It makes sense that Europeans overall have more EHG than WHG ancestry since the Indo-Europeans who expanded into Europe from the steppe had a lot of EHG ancestry in their genomes. EHG is probably some sort of ANE like and WHG mix though.

dp
08-26-2015, 08:44 PM
Just so everyone sees where I'm at. I am very close to done. It wont be long now.

https://docs.google.com/spreadsheets/d/1S81m-HdWpnRYtPhF7uubuoETSQSCnAbe2KEuJRNG_yU/edit?usp=sharing
One of the first calculator to split Amerind into N. & S.
dp :-)

Chad Rohlfsen
08-27-2015, 03:40 AM
It makes sense that Europeans overall have more EHG than WHG ancestry since the Indo-Europeans who expanded into Europe from the steppe had a lot of EHG ancestry in their genomes. EHG is probably some sort of ANE like and WHG mix though.

A WHG/ANE mix is very unlikely. We're talking about the Mal'ta sample, which is over 20kyo. It isn't much more relevant to the conversation than Kostenki. It's best to stick to samples that fall within the same general timeframe, like WHG, EEF, EHG.

Chad Rohlfsen
08-27-2015, 03:41 AM
Impressive, but is it possible to get a Romanian sample? It's kind of ridiculous seeing for example Slovakians and Slovenians, while a way larger European population is missing.

I don't have any Romanians. You could be the first, if you like? I may start testing it this weekend.

Chad Rohlfsen
08-27-2015, 03:42 AM
After studying the datasheet, I realized that I had overnight morphed from a WHG into an EHG.
The rule seems to be that every HG which has a Yamnaya-like admixture is defined as EHG, which leaves Loschbour, La Brana and HungaryGamba_HG as WHG.
Isn't it preferable to rename these labels? For instance palaeoHG and neoHG for WHG and EHG.
P.S. I prefer to split Yamnaya between EHG and West Asian.

WHG and EHG are from close to the same timeframe. Renaming them would be more confusing. WHG and EHG are the standard names that they're known by.

ZephyrousMandaru
08-27-2015, 03:48 AM
Chad, it's best to use the Assyrian volunteers from David, the "Assyrian_WGA" sample may not be representative of Assyrians. In some tests such as MDLP K23b, they appear almost indistinguishable from Armenians and I clustered fairly distantly from them and Armenians.

Chad Rohlfsen
08-27-2015, 04:01 AM
Seems very promising.

I like the idea of having EHG and West Asian separated, instead of a single Yamnaya percentage. The more detailed, the better!

Thanks! I think it truly shows the difference between Corded, Beaker, and Yamnaya. Corded looks more like 50% Yamnaya-like, plus a lot of EHG and EEF, rather than Yamnaya plus MN, which doesn't make sense considering the difference in genesis. Also, the Karelia HG shows WHG and Amerind admixture, which makes more sense. Archaeologically speaking, Swiderians (WHG), were the first settlers in the East Baltic, and the Karelian shows Siberian admixture, plus there is at least one Mongoloid like skull in this area. I think it is much more likely that the Karelian really does have some WHG and Amerind like admixture. One looks admixed with ENAs.56895690

Chad Rohlfsen
08-27-2015, 04:05 AM
Chad, it's best to use the Assyrian volunteers from David, the "Assyrian_WGA" sample may not be representative of Assyrians. In some tests such as MDLP K23b, they appear almost indistinguishable from Armenians and I clustered fairly distantly from them and Armenians.

Okay. I'll see what I can do.

J Man
08-27-2015, 11:32 AM
Thanks! I think it truly shows the difference between Corded, Beaker, and Yamnaya. Corded looks more like 50% Yamnaya-like, plus a lot of EHG and EEF, rather than Yamnaya plus MN, which doesn't make sense considering the difference in genesis. Also, the Karelia HG shows WHG and Amerind admixture, which makes more sense. Archaeologically speaking, Swiderians (WHG), were the first settlers in the East Baltic, and the Karelian shows Siberian admixture, plus there is at least one Mongoloid like skull in this area. I think it is much more likely that the Karelian really does have some WHG and Amerind like admixture. One looks admixed with ENAs.56895690

Which archaeological groups do you think the EHG component comes from when seen in modern Europeans ultimately? Mesolithic forest-steppe hunter-gatherers?

MatAust21
08-29-2015, 03:13 AM
Do you think it will be much longer before we can send our donations and raw datas?

Chad Rohlfsen
08-30-2015, 12:45 AM
It's coming along. I've made a change to it. I'll go over it, after I've posted it. I'm getting it to agree with alternate forms of testing. Let's just say that WHG is higher now.

J Man
08-30-2015, 12:50 AM
It's coming along. I've made a change to it. I'll go over it, after I've posted it. I'm getting it to agree with alternate forms of testing. Let's just say that WHG is higher now.

Which archaeological groups do you think the EHG component comes from when seen in modern Europeans ultimately? Mesolithic forest-steppe hunter-gatherers?

Chad Rohlfsen
08-30-2015, 03:02 AM
I think it's almost all Samara-like. The excess ENA is lacking really, outside of Balts.

J Man
08-30-2015, 03:35 AM
I think it's almost all Samara-like. The excess ENA is lacking really, outside of Balts.

So the EHG component when seen in modern people comes from ancestors who were Samara like East European Mesolithic hunter-gatherers?

Chad Rohlfsen
08-30-2015, 05:10 AM
It's from a Yamnaya related pop, for the most part. Is that what you're asking?

J Man
08-30-2015, 12:25 PM
It's from a Yamnaya related pop, for the most part. Is that what you're asking?

No what I am asking is this. Is the EHG component from this test made up of Mesolithic East European hunter-gatherer alleles? Or is it made up of Near Eastern/West Asian Neolithic farmer alleles? Or both? I am thinking that it is probably made up.of Mesolithic East European hunter-gatherer alleles since it is called "EHG" but I just wanted to check for sure.

DMXX
08-30-2015, 12:45 PM
EHG looks very distinct from anything emanating out of West or Southwest Asia, so it has to be derived from East Euro HG's.

It's that "other side" of Yamnaya (the "teal" component) which looks West Asian related. From the posts, looks like Chad has already largely resolved that into his West Asian cluster.

Chad Rohlfsen
08-30-2015, 08:11 PM
It's EHG. The Samara hunter is 100%. The Karelian has some WHG and ENA affinities, which is expected.

J Man
08-30-2015, 08:46 PM
It's EHG. The Samara hunter is 100%. The Karelian has some WHG and ENA affinities, which is expected.

So then the EHG component when seen in modern day people certainly does ultimately come from Mesolithic East European hunter-gatherer ancestors. Very cool! :)

Chad Rohlfsen
08-30-2015, 08:58 PM
Yes, but I think the last run was off. EHG is going to be lower and WHG higher. I think having WHG 25% higher and EHG lower is more accurate. I'm trying to fix it towards qpAdm numbers.

J Man
08-30-2015, 09:04 PM
Yes, but I think the last run was off. EHG is going to be lower and WHG higher. I think having WHG 25% higher and EHG lower is more accurate. I'm trying to fix it towards qpAdm numbers.

Ahhh I see...Well I hope that the test is ready soon! I am excited about it.

Chad Rohlfsen
08-30-2015, 09:54 PM
Yeah, qpAdm is not showing the same thing as Haak and this last run. It looks like the greatest variation is N-S, rather than E-W.

J Man
08-30-2015, 11:06 PM
Yeah, qpAdm is not showing the same thing as Haak and this last run. It looks like the greatest variation is N-S, rather than E-W.

I hope it goes smooth.

Chad Rohlfsen
08-30-2015, 11:36 PM
EHG looks very distinct from anything emanating out of West or Southwest Asia, so it has to be derived from East Euro HG's.

It's that "other side" of Yamnaya (the "teal" component) which looks West Asian related. From the posts, looks like Chad has already largely resolved that into his West Asian cluster.

I think this is better than the teal component. The opposite of the teal was a Euro HG, which was a mix of WHG and EHG, so the Teal was probably not a real population. This, on the other hand, would be that other population, if EHG is in fact the population that is ancestral to Yamnaya.

Edit: I think the hard part now is trying to piece it all together with Europeans and everyone else. Trying to keep EHG and the amount of WHG, is pretty tough.

J Man
08-31-2015, 01:05 AM
I think this is better than the teal component. The opposite of the teal was a Euro HG, which was a mix of WHG and EHG, so the Teal was probably not a real population. This, on the other hand, would be that other population, if EHG is in fact the population that is ancestral to Yamnaya.

Edit: I think the hard part now is trying to piece it all together with Europeans and everyone else. Trying to keep EHG and the amount of WHG, is pretty tough.

At some point still do you think that you will offer this as a test for us to purchase/use?

Chad Rohlfsen
08-31-2015, 02:51 AM
For sure. I'm just trying to do a little bit of doctoring here.

J Man
09-02-2015, 02:58 PM
For sure. I'm just trying to do a little bit of doctoring here.

Are you very close to releasing it by now?

Chad Rohlfsen
09-02-2015, 03:15 PM
Pretty close, yes. I've got mid-terms right now, so I will try to get something going before the weekend is up.

Chad Rohlfsen
09-06-2015, 03:17 PM
Okay. I'm trying one to two more stabs at getting this as close to qpAdm as possible, then I'm calling it good. I'd like to be able to start testing folks by Monday.

J Man
09-06-2015, 04:45 PM
Okay. I'm trying one to two more stabs at getting this as close to qpAdm as possible, then I'm calling it good. I'd like to be able to start testing folks by Monday.

Awesome! Which email address should we send payment to?

Chad Rohlfsen
09-07-2015, 03:41 AM
Awesome! Which email address should we send payment to?

I'll post the paypal and email I'm using once it's ready. I'm going to do a test run on myself first.

MatAust21
09-07-2015, 03:47 AM
How much will it be again? 20 USD, right?

Chad Rohlfsen
09-07-2015, 06:58 AM
Yes, 20 USD.

Chad Rohlfsen
09-08-2015, 02:56 AM
Alright, I may give up on making EEF more Near Eastern. I apologize for the delay. I have a run that should end in the morning, and the next by tomorrow night. I should know what I am going to go with after that. Thank you all, for your patience. I will be testing this week.

J Man
09-08-2015, 04:23 PM
Alright, I may give up on making EEF more Near Eastern. I apologize for the delay. I have a run that should end in the morning, and the next by tomorrow night. I should know what I am going to go with after that. Thank you all, for your patience. I will be testing this week.

Sounds good! Can you post a list of components that will be part of this test? Would be interesting to see.

Chad Rohlfsen
09-08-2015, 09:49 PM
I'll put it up after the run. I'm trying one to two more runs using KO2, as it sits where Barcin was. If these two attempts fail, I'll go back towards the original run. Then, testing starts.

ZephyrousMandaru
09-08-2015, 10:02 PM
Are you going to upload it to GED Match or are you going to upload the calculator files, or both?

Generalissimo
09-09-2015, 01:12 AM
Are you going to upload it to GED Match or are you going to upload the calculator files, or both?

How is he going to charge for it if he puts it up at GEDmatch or releases the calculator files?

Chad Rohlfsen
09-09-2015, 03:12 AM
I'll do the test for folks, and I'm sure someone will make a calculator/oracle in 4-mix R, or what-not. There's plenty of people that like making those oracles. I'll have most of the ancient samples included in the spreadsheet.

ZephyrousMandaru
09-09-2015, 03:37 AM
How is he going to charge for it if he puts it up at GEDmatch or releases the calculator files?

He said he was going to put it up, and I wasn't sure what he meant by that.

DMXX
09-09-2015, 03:45 AM
I'll put it up after the run. I'm trying one to two more runs using KO2, as it sits where Barcin was. If these two attempts fail, I'll go back towards the original run. Then, testing starts.

Looking forward to it.

As the enthused check-ins from our regulars show, we're dealing with a forum-wide case of calculator withdrawal... And you have the medicine at hand. :D

ZephyrousMandaru
09-09-2015, 03:48 AM
Chad, if you have the current population averages, I can create a 4Mix and Oracle-4 datasheet file for the standalone 4-Ancestors Oracle. So if you want to post the link to the final population averages I can create them.

Chad Rohlfsen
09-09-2015, 04:38 AM
Chad, if you have the current population averages, I can create a 4Mix and Oracle-4 datasheet file for the standalone 4-Ancestors Oracle. So if you want to post the link to the final population averages I can create them.

Sounds like a plan!

Chad Rohlfsen
09-09-2015, 10:14 PM
Okay, I've started my last little tweak to the test. It should be done after midnight. I'll check it in the morning, and post an update. I've added a Motala and Sweden NHG to see if that helps things out. There's an outside chance I could run a max of one more after that. Either way, I'm running tests starting after work, either tomorrow or Friday. Everyone get ready!

Chad Rohlfsen
09-09-2015, 11:53 PM
BTW, as soon as those Iberians are available, I'll run them with this test. I took a look at downloading and reformatting them, but they're too big. I only get 100GB a month, and with that first sample being 12GB, I had to pass on it. I need my net for school. If I have time, maybe I can go to the library and do it. We shall see. Please, if you know someone has them, alert David or myself. I have the tool to reformat the data if it hasn't been done. Thanks!

Chad Rohlfsen
09-10-2015, 11:38 AM
I'll have to check the test later. It was wrapping up as I left.

Chad Rohlfsen
09-10-2015, 02:00 PM
Alright, I got the rest of the day off. I am running what should be the last test, prior to running everyone's genomes. This should be done running this evening and I will go over the testing method with David. Once I am all set up, I will run myself and post the results. David will also be doing a PCA for folks. I will try to learn that part as well. If this last test does what I want, then I will post the results. If not, I will go back to the original run. I'll post an update as soon as this run finishes tonight.

MatAust21
09-10-2015, 03:58 PM
Since you seem to be very close to a conclusion, could you give us the email address, please?

I'm sure some of us are already very interested, even before seeing the final spreadsheet.

Chad Rohlfsen
09-10-2015, 11:11 PM
The e-mail that I want the 23andme files sent to is [email protected]

My paypal address is [email protected] Please advise me who you are, when sending the genome, so I can verify payment.

I would appreciate it if everyone holds off on sending me any genomes or money until I say that I'm ready. I can only run 3-5 people at a time, so patience is needed. If I get bombarded, some people may be waiting up to 2-3 days for results. David will also be assisting with the testing. I am going to learn how to do the PCA for everyone, as well. I am not able to do FTDNA files yet. I have to download and run a program to reformat them. I'll try to have that figured out in the next day or two.

Also, I would appreciate it if everyone named their 23andme file the same as their e-mail, just to make it easier to keep them straight. Thanks!

I will be posting my results as soon as they're done and give a heads up when I'm ready.

Kaido
09-11-2015, 12:20 AM
Not sure if you've already mentioned it but is this test aimed primarily at those of European descent?

Chad Rohlfsen
09-11-2015, 04:38 AM
It's for anyone. I've had another hiccup trying to learn the test set-up. I'll get this going ASAP.

J Man
09-11-2015, 12:49 PM
It's for anyone. I've had another hiccup trying to learn the test set-up. I'll get this going ASAP.

Can't wait!

J Man
09-12-2015, 08:04 PM
It's for anyone. I've had another hiccup trying to learn the test set-up. I'll get this going ASAP.

How is it looking now?

Chad Rohlfsen
09-12-2015, 09:23 PM
How is it looking now?

Not too bad. David is making the test files. I may be able to run my first test tonight or tomorrow.

J Man
09-14-2015, 03:38 PM
Not too bad. David is making the test files. I may be able to run my first test tonight or tomorrow.

Any updates?

Chad Rohlfsen
09-14-2015, 06:39 PM
I'm testing myself and ATP3, right now. It may not be done until morning. I'm really trying to get this up and rolling. I'm going through finals right now too, so I'll do as much as I can.

J Man
09-19-2015, 01:09 AM
I'm testing myself and ATP3, right now. It may not be done until morning. I'm really trying to get this up and rolling. I'm going through finals right now too, so I'll do as much as I can.

Coming along well?

Chad Rohlfsen
09-19-2015, 11:37 PM
Alright. I was messing around with it and I think it's possible to split WHG into a LaBrana and KO1 heavy split. I just tweaked EHG, and I'm trying another run.

Arbogan
09-21-2015, 06:50 AM
I'll be sending you money soon. As soon as I can transfer money to my paypal account. How is the funding proceeding?

Chad Rohlfsen
09-21-2015, 02:33 PM
I'm not waiting for any goal, just for people to pay as they send their data. I'll post here when I'm ready.

Arbogan
09-22-2015, 08:05 PM
I'm not waiting for any goal, just for people to pay as they send their data. I'll post here when I'm ready.

Is your official paypal mail still?: [email protected]

Chad Rohlfsen
09-22-2015, 08:09 PM
That's the email for genomes. PayPal is [email protected]

J Man
09-23-2015, 12:22 AM
That's the email for genomes. PayPal is [email protected]

Still testing a few things out eh?

Dimanto
09-23-2015, 02:23 AM
There are no Dutch samples listed in the spreadsheet. If possible would you add them too ?

https://docs.google.com/spreadsheets/d/1S81m-HdWpnRYtPhF7uubuoETSQSCnAbe2KEuJRNG_yU/edit#gid=163828793

Chad Rohlfsen
09-23-2015, 02:36 AM
There's no Dutch samples in my dataset.

Dimanto
09-23-2015, 02:39 AM
There's no Dutch samples in my dataset.

What is the reason that you didn't add Dutch samples with it ?

I didn't see any southern Italian either except Sicilians and Sardinians if I'm not mistaken.

I'd like to send my genome + the additional 20 $ but without these 2 populations tested, I don't know if it would be interesting for me to participate.

Chad Rohlfsen
09-23-2015, 02:50 AM
What is the reason that you didn't add Dutch samples wit it ?

I didn't see any southern Italian either except Sicilians and Sardinians if I'm not mistaken.

I'd like to send my genome + the additional 20 $ but without these 2 populations tested, I don't know if it would be interesting for me to participate.

There is one south Italian in the set. It's rather limited in that area. I'd love to have some Germans, but I don't have them either.

Dimanto
09-23-2015, 03:04 AM
There is one south Italian in the set. It's rather limited in that area. I'd love to have some Germans, but I don't have them either.

I'll try and ask some people if they could contribute to the southern Italian reference.

J Man
09-23-2015, 03:27 AM
There is one south Italian in the set. It's rather limited in that area. I'd love to have some Germans, but I don't have them either.

Are you accepting payments now?

crossover
09-23-2015, 05:46 AM
what type of caculator will this be

Arbogan
09-23-2015, 08:09 AM
That's the email for genomes. PayPal is [email protected]

Ok donation sent. I'll send the genomes later.

J Man
09-26-2015, 06:43 PM
Bump.

Chad Rohlfsen
09-27-2015, 12:18 AM
I'm going to practice the PCA program this evening, to see if I can get that figured out. I've got to finish an assignment first.

Arbogan
10-01-2015, 05:02 AM
When can we expect any of our genomes to be processed?

J Man
10-02-2015, 07:12 PM
When can we expect any of our genomes to be processed?

A good question.

Varun R
10-02-2015, 08:16 PM
Im confused. Have people already sent in their genomes and money?

Sangarius
10-02-2015, 08:27 PM
Im confused. Have people already sent in their genomes and money?

It seems like Arbogan did, but people aren't supposed to do that yet.

DMXX
10-02-2015, 11:27 PM
The most appropriate move for prospective participants is to hold tight until Chad's good and ready to receive payments for his testing services. I too am eagerly expecting his work to be made available, but as he's indicated before, there are either time or technical constraints in the way.

J Man
10-03-2015, 12:44 AM
The most appropriate move for prospective participants is to hold tight until Chad's good and ready to receive payments for his testing services. I too am eagerly expecting his work to be made available, but as he's indicated before, there are either time or technical constraints in the way.

Well said.

J Man
10-15-2015, 07:19 PM
Bump...Is this test still happening/being made or has it been scrapped? Not trying to be rude or anything I am just genuinely curious.

Chad Rohlfsen
10-16-2015, 11:58 PM
It's still happening. I'm splitting South Asian up into West and East Eurasian components.

J Man
10-17-2015, 02:01 AM
It's still happening. I'm splitting South Asian up into West and East Eurasian components.

Excellent and that is a good idea.

Chad Rohlfsen
10-17-2015, 11:43 PM
Okay. I'm taking a little poll here. People please chime in with your answer and reasoning. Input from South and South Central Asians is greatly appreciated. Thanks!

How much ENA does everyone think should be in the Paniya and Pulliyar?

Varun R
10-18-2015, 04:36 AM
Okay. I'm taking a little poll here. People please chime in with your answer and reasoning. Input from South and South Central Asians is greatly appreciated. Thanks!

How much ENA does everyone think should be in the Paniya and Pulliyar?

Errr...going by previously posted admixture runs, I would assume ENA in the 50-55 percent range for Paniya. Pulliyar appear to be a non-tribal population, and so ENA should be somewhat lower than in Paniya (maybe ~45%)? Hopefully someone who has worked with admixture will chime in.

J Man
10-19-2015, 07:01 PM
Okay. I'm taking a little poll here. People please chime in with your answer and reasoning. Input from South and South Central Asians is greatly appreciated. Thanks!

How much ENA does everyone think should be in the Paniya and Pulliyar?

How much longer do you think until you start testing people? Sorry to be a bother.

Chad Rohlfsen
10-20-2015, 04:55 PM
As soon as the break in ANI and ASI looks appropriate. I'm at it again tonight. I had a final that if been working on.

Dr_McNinja
10-21-2015, 08:22 PM
Okay. I'm taking a little poll here. People please chime in with your answer and reasoning. Input from South and South Central Asians is greatly appreciated. Thanks!

How much ENA does everyone think should be in the Paniya and Pulliyar?At least 30-40% would be my guess. That would be a mix of ASI/ASE, East Asian, and Oceanian.

bored
10-21-2015, 08:49 PM
At least 30-40% would be my guess. That would be a mix of ASI/ASE, East Asian, and Oceanian.

Probably much more. I know the correlation between phenotype and genotype is not straightforward but Paniya definitely do not look 30 % ASI.

Dr_McNinja
10-21-2015, 09:20 PM
Probably much more. I know the correlation between phenotype and genotype is not straightforward but Paniya definitely do not look 30 % ASI.
I'd still put an upper bound on their ENA at around 60%. They come out like ~20% ANE equivalent in calculators.

J Man
10-21-2015, 09:31 PM
As soon as the break in ANI and ASI looks appropriate. I'm at it again tonight. I had a final that if been working on.

I am also curious about whether Europeans have more WHG or EHG ancestry overall based on the tests that you have done so far?

Chad Rohlfsen
10-23-2015, 02:44 PM
It looks like EHG for most, when using EEF.

Chad Rohlfsen
10-25-2015, 02:08 AM
98 samples to tickle everyone's fancy, and to see that I'm still working here. I'm still perfecting it. I'm trying to dump the NE Asian and keep the non Central Asian part of South Asians as just Austronesian and Papuan.

https://docs.google.com/spreadsheets/d/12mbaouZeCmNx4jzZuJQUC_DfprF28fi7kN_ofLhPLC4/edit?usp=sharing

BalkanKiwi
10-25-2015, 02:47 AM
My fancy has been tickled.

J Man
10-25-2015, 11:35 PM
98 samples to tickle everyone's fancy, and to see that I'm still working here. I'm still perfecting it. I'm trying to dump the NE Asian and keep the non Central Asian part of South Asians as just Austronesian and Papuan.

https://docs.google.com/spreadsheets/d/12mbaouZeCmNx4jzZuJQUC_DfprF28fi7kN_ofLhPLC4/edit?usp=sharing

Now for the sake of comparing is David's West Eurasia K8 test still the best test out there for accurately estimating levels of the ANE component in our genomes would you say?

http://eurogenes.blogspot.com.au/2014/12/ane-is-primary-cause-of-west-to-east.html

Dr_McNinja
10-26-2015, 01:55 AM
Looking good, I like how the Pulliyar are breaking down.

Dr_McNinja
10-26-2015, 02:08 AM
98 samples to tickle everyone's fancy, and to see that I'm still working here. I'm still perfecting it. I'm trying to dump the NE Asian and keep the non Central Asian part of South Asians as just Austronesian and Papuan.

https://docs.google.com/spreadsheets/d/12mbaouZeCmNx4jzZuJQUC_DfprF28fi7kN_ofLhPLC4/edit?usp=sharing
There is some legit NE-Asian in South Asia though, especially in the North and Northeast. SE-Asian as well, both representing East Asian. I suppose Austronesian would pick up for SE-Asian.

As long as NE-Asian is peaking in like Japanese, I don't see why you'd want to remove it?

Btw, have you run Balochis/etc through this calculator yet? In whom does Central Asian peak and at what amount?

Chad Rohlfsen
10-26-2015, 03:47 AM
I have about 1600-1700 samples in this. I'd have to check, but I think it peaked in Gujaratis at about 60-65% I'll post some more tomorrow.

Helgenes50
10-26-2015, 08:52 AM
98 samples to tickle everyone's fancy, and to see that I'm still working here. I'm still perfecting it. I'm trying to dump the NE Asian and keep the non Central Asian part of South Asians as just Austronesian and Papuan.

https://docs.google.com/spreadsheets/d/12mbaouZeCmNx4jzZuJQUC_DfprF28fi7kN_ofLhPLC4/edit?usp=sharing

Is most of our (W)HG ancestry included in the EEF ?

Chad Rohlfsen
10-26-2015, 09:53 AM
I think so.

Edit: I've noticed the LN/EBA have less WHG than modern folks. I'd guess that the mixing continued a little while longer.

crossover
10-26-2015, 01:35 PM
are these samples from someone else or did you actually get dna samples from people

Chad Rohlfsen
10-26-2015, 02:24 PM
They're datasets available for download.

Chad Rohlfsen
10-27-2015, 12:30 AM
Okay. Here are about 400 more samples from EN, MN, Europe, West, East, and South Asia. What is everyone's opinion?

https://docs.google.com/spreadsheets/d/1s8JFyZVSGeIPHrSFaC2qiIuE6XpceWp75GXN2pOUsj4/edit?usp=sharing

DMXX
10-27-2015, 01:02 AM
Looks very good. Did you have any success with isolating ANE independent of EHG? I'm just wondering how Amerindian pops will look.

Chad Rohlfsen
10-27-2015, 01:07 AM
The only way of doing that is to eliminate the Siberian component and double up on MA1. I think the age of MA1 (looking like he has SC Asian Neolithic ancestry) causes some issues. It ends up taking out the South Central Asian stuff and giving pretty low Near Eastern to Yamnaya and SC Asians. Hopefully, someone will get some Keltiminar samples. I think those will be closer to MA1 than the Samaran and Karelian. The numbers are about the same, 36-41% in the Karitiana, but again, it robs a lot of Neolithic ancestry specific to SC Asia, the Caucasus, and Yamnaya.

J Man
10-27-2015, 03:51 AM
The only way of doing that is to eliminate the Siberian component and double up on MA1. I think the age of MA1 (looking like he has SC Asian Neolithic ancestry) causes some issues. It ends up taking out the South Central Asian stuff and giving pretty low Near Eastern to Yamnaya and SC Asians. Hopefully, someone will get some Keltiminar samples. I think those will be closer to MA1 than the Samaran and Karelian. The numbers are about the same, 36-41% in the Karitiana, but again, it robs a lot of Neolithic ancestry specific to SC Asia, the Caucasus, and Yamnaya.

Sounds like you are getting very close! :D

Arbogan
10-27-2015, 05:55 PM
Have you had any success separating BE out of EEF

Chad Rohlfsen
10-27-2015, 06:36 PM
Have you had any success separating BE out of EEF

I don't think anyone will be able to do that. At lower k-values, I can get something about 75% BE.

Sangarius
10-27-2015, 06:38 PM
I don't think anyone will be able to do that. At lower k-values, I can get something about 75% BE.

Why is that the case? Seems like nobody is able to create a synthetic BE.

Chad Rohlfsen
10-27-2015, 06:41 PM
We have no ancient or modern pure samples. Admixture won't latch onto it.

Sangarius
10-27-2015, 06:46 PM
We have no ancient or modern pure samples. Admixture won't latch onto it.

I see. But if I remember correctly David created his ANE component from Karitiana samples?

Chad Rohlfsen
10-27-2015, 06:51 PM
Yes, in a way. He knows how to do things that I don't. I still don't think it's possible at this time.

Arbogan
10-27-2015, 08:37 PM
Yes, in a way. He knows how to do things that I don't. I still don't think it's possible at this time.

You could always consult him. It would make much more sense, than having non-Europeans have false "EEF".

Chad Rohlfsen
10-27-2015, 09:58 PM
Who are you claiming doesn't have EEF? West Asians, central, and South Asians were mixed with EEF bearing pops.

Generalissimo
10-27-2015, 11:09 PM
You could always consult him. It would make much more sense, than having non-Europeans have false "EEF".

Karitiana are a straight two-way mix of ANE and ENA, while populations that carry a lot of Basal Eurasian are really complex mixtures with stuff that hasn't yet been identified properly from West Eurasia and even East Africa. So it's basically impossible to isolate Basal Eurasian like ANE. It might never be possible.

Also, Near Easterners and Central Asians do carry EEF. Those Neolithic Anatolians, for instance, were basically EEF, with actual European admixture, which can be seen in their autosomal DNA as well as Y-DNA (I2 and C1). Modern Anatolians and other West Asians have ancestry from them, while Central Asians have ancestry from steppe groups that were partly EEF.

Arbogan
10-28-2015, 12:06 AM
Who are you claiming doesn't have EEF? West Asians, central, and South Asians were mixed with EEF bearing pops.

I don't think it's EEF. Atleast not in the case of non-Anatolian/non-Caucasus groups. It's rather something similar to UHG or WHG from steppe groups and ENF, Combining. If we take your assumption that Iranians have 23% EFF. Similar to Stuttgart, the ratio being 1:1 or even 0.6:1 WHG/ENF, that would mean that Iranians on average have 11.5 or 9.2 WHG. Which doesn't show up in any test. In lower K tests, the maximum WHG is like 3-4%, and it's steppe related.

Also another thing i'd like to note. Your inclusion of EHG, basically makes almost all WHG, EHG related. Since there is no proxy for ANE.

Chad Rohlfsen
10-28-2015, 01:07 AM
Two things;

EEF is about identical to Anatolian farmers. EEF is in LBA groups.

EHG is significantly closer to WHG than ANE is. For all we know, MA1 is a dead end and not ancestral. He might not represent the West Eurasian in Native Americans. EHG is just as close to Native Americans, while being closer to WHG.

Why are you comparing my run with ENF? I have an EEF and EHG cluster, so admixture won't create the same stuff. This has actual pops and not possibilities. It's like saying horses aren't real and unicorns are.

Edit: I have three West Eurasian pops that nearly exist at the exact same time. This fixation with MA1, ENF, and Teal is not helping us move forward. MA1 is too old and no one knows what his place is. ENF is a possibility, but not a sure thing. Teal, may be a modern cluster, but is not the other half of Yamnaya. It's not a real ancient population. Let's stick to solid representations like EEF, WHG, and EHG, and by force, Bedouins, for West Eurasians. I will use Anatolians, once they become available.

Arbogan
10-31-2015, 01:45 PM
Two things;

EEF is about identical to Anatolian farmers. EEF is in LBA groups.

EHG is significantly closer to WHG than ANE is. For all we know, MA1 is a dead end and not ancestral. He might not represent the West Eurasian in Native Americans. EHG is just as close to Native Americans, while being closer to WHG.

Why are you comparing my run with ENF? I have an EEF and EHG cluster, so admixture won't create the same stuff. This has actual pops and not possibilities. It's like saying horses aren't real and unicorns are.

Edit: I have three West Eurasian pops that nearly exist at the exact same time. This fixation with MA1, ENF, and Teal is not helping us move forward. MA1 is too old and no one knows what his place is. ENF is a possibility, but not a sure thing. Teal, may be a modern cluster, but is not the other half of Yamnaya. It's not a real ancient population. Let's stick to solid representations like EEF, WHG, and EHG, and by force, Bedouins, for West Eurasians. I will use Anatolians, once they become available.
Those low coverage barcin samples and the sample Stuttgart don't represent more than the the neolithics in the europe and western anatolia. So the ENF construct is neccesary unless we're going to assume there was a population replacement. Which is much more remote than the hypothetical ENF.

Chad Rohlfsen
10-31-2015, 03:10 PM
I'm not using Barcin or Stuttgart. Im using the LBK and Spanish samples. Did you read the Anatolian abstract? They said there's been a large replacement in West Asia. You can see that in the Armenian BA and IA samples too. I'm not continuing this conversation, because you obviously aren't looking at my samples or what the papers say.

J Man
10-31-2015, 08:01 PM
I'm not using Barcin or Stuttgart. Im using the LBK and Spanish samples. Did you read the Anatolian abstract? They said there's been a large replacement in West Asia. You can see that in the Armenian BA and IA samples too. I'm not continuing this conversation, because you obviously aren't looking at my samples or what the papers say.

How is the work on the ANI and ASI clusters coming along?

Arbogan
11-01-2015, 04:24 AM
I'm not using Barcin or Stuttgart. Im using the LBK and Spanish samples. Did you read the Anatolian abstract? They said there's been a large replacement in West Asia. You can see that in the Armenian BA and IA samples too. I'm not continuing this conversation, because you obviously aren't looking at my samples or what the papers say.

I looked at the paper's abstract:


1) Department of Genetics, Harvard Medical School, Boston, MA USA; 2) Broad Institute of MIT and Harvard, Cambridge, MA USA; 3) Conway Institute of Biomolecular and Biomedical Research, University College Dublin, Dublin, Ireland; 4) Howard Hughes Medical Institute, Harvard Medical School, Boston, MA USA; 5) Independent physical anthropologist, Netherlands.

It has hitherto been difficult to obtain genome-wide data from the Near East. By targeting the inner ear region of the petrous bone for extraction [Pinhasi et al., PLoS One 2015] and using a genome-wide capture technology [Haak et al., Nature, 2015] we achieved unprecedented success in obtaining genome-wide data on more than 1.2 million single nucleotide polymorphism targets from 34 Neolithic individuals from Northwestern Anatolia (~6,300 years BCE), including 18 at greater than 1 coverage. Our analysis reveals a homogeneous population that is genetically a plausible source for the first farmers of Europe in the sense of (i) having a high frequency of Y-chromosome haplogroup G2a, and (ii) low Fst distances from early farmers of Germany (0.004 0.0004) and Spain (0.014 0.0009). Model-free principal components and model-based admixture analyses confirm a strong genetic relationship between Anatolian and European farmers. We model early European farmers as mixtures of Neolithic Anatolians and Mesolithic European hunter-gatherers, revealing very limited admixture with indigenous hunter-gatherers during the initial spread of Neolithic farmers into Europe. Our results therefore provide an overwhelming support to the migration of Near Eastern/Anatolian farmers into southeast and Central Europe around 7,000-6,500 BCE [Ammerman & Cavalli Sforza, 1984, Pinhasi et al., PLoS Biology, 2005]. Our results also show differences between early Anatolians and all present-day populations from the Near East, Anatolia, and Caucasus, showing that the early Anatolian farmers, just as their European relatives, were later demographically replaced to a substantial degree.

They samples are listed to be derived from barcin. I'd wait for samples from the rest of the region before I'd jump to this conclusion. I highly doubt that there was such a large population replacement, not until they get samples from Iran, the levant, and Mesopotamia, that would confirm this. The group in north-western Anatolia, might have been a sub-group, that was already differentiated by the time it mixed with WHGs, since the earliest Neolithic site in the area was in the natufian 12k. And if these samples were really representative of the entire region, they'd have a much stronger influence, since their dated immigration 6.3-6k b.c. Which is only a few hundred years far off from the establishment of the first major "settlements" in the area, it would mean that their genetic input would have been much more prevalent. It's probably more complex than this. But we won't know until they have samples from the entire region. And the fact that they have European hunter-gatherer ancestry, makes them even less viable as representative samples for the region. So it makes no sense to just jump to a drastic conclusion, and if this large replacement did happen, what evidence do they have, yet, to support this explanation model, which postulates this? It's just too early to jump to that conclusion, with the limited research in the area. I wonder if the ain ghazal is consistent with these barcin samples. And the fact that west-Asians score some EFF in your preliminary runs, could mean a lot of things.

anyway, i'm looking forward to the test being released. And future samples.

Chad Rohlfsen
11-09-2015, 05:16 PM
I think I've taken this as far as it can go. If people are open to waiting, the Anatolians should be released in the next 3-4 weeks. That could really help in breaking down Yamnaya and other LN/EBA groups.

MatAust21
11-16-2015, 06:37 PM
Will the new paper on CHG (Caucasus Hunter-Gatherers) be of any use to make the test more reliable?

After the Anatolian farmers are released, it seems like we will have samples from all the four ingredients that make the Europeans: EHG, WHG, CHG and Neolithic Farmers.

Chad Rohlfsen
11-17-2015, 01:02 AM
Yeah, I will have the Anatolians, and CHG samples in the test. I'm just waiting for them to be released, and for David to get them ready for me. As soon as I have them, I will work my way on whittling down the SNP count.

Chad Rohlfsen
11-28-2015, 02:31 AM
Just keeping everyone posted. I'm making some progress in pruning the samples. I'm hoping that I can have something posted before the weekend is up.

J Man
11-30-2015, 03:46 PM
Just keeping everyone posted. I'm making some progress in pruning the samples. I'm hoping that I can have something posted before the weekend is up.

How is it coming along?

Chad Rohlfsen
11-30-2015, 08:47 PM
I'm still trying to get EHG to come out right. Each run takes about 12 hours. I've got 122k SNPs and 3100 samples. It's going 24/7 and I won't give up.

Chad Rohlfsen
12-03-2015, 10:34 PM
I'm getting very close here!

J Man
12-04-2015, 12:00 AM
I'm getting very close here!

I have a very good feeling about this test! :D

Generalissimo
12-04-2015, 10:41 AM
We're sort of aiming for the New Year, but accuracy is more important than a date.

J Man
12-10-2015, 09:28 PM
What kind of components are going to be included in this test? EHG?, WHG?, EEF?

Chad Rohlfsen
12-11-2015, 05:06 AM
What kind of components are going to be included in this test? EHG?, WHG?, EEF?

I will have up a spreadsheet in the next couple of days. Possibly tomorrow. I'll go over it then. I'm just doing some tweaks.

Chad Rohlfsen
12-12-2015, 07:14 AM
First share, here. I am still working on it. My computer froze when I was at work, so I'm a day behind. Another day or so, and I should be pretty comfortable with it.

https://docs.google.com/spreadsheets/d/1cihUQZAykzy2kFEr9-mv6RRPxry2PHWDjkgpVDdOJ_Q/edit?usp=sharing

Varun R
12-12-2015, 08:02 AM
Chad,

What do you make of the (seemingly) high Hadza scores in S Asians?

Chad Rohlfsen
12-12-2015, 08:05 AM
Slave trade? I'm not sure. East African stuff does pop up like this, from time to time. It's still a work in progress. Most of it could end up disappearing.

Varun R
12-12-2015, 08:12 AM
Ok. I'm also curious about the CHG in Australian samples...

Ultimately I'm still confused about the nature of ASI/ whether it's purely E Eurasian or as some have been suggesting incorporates unique Basal Eurasian ancestry that doesn't fall into either category...I'm increasingly thinking that we'll require paleolithic hunter-gatherer samples from the southern part of the subcontinent to verify.

Chad Rohlfsen
12-12-2015, 08:17 AM
Admixture from South Asians? It could disappear. I'm still working on it.

Ignis90
12-12-2015, 12:39 PM
The North African component is based on the high endogamy of Mozabites, it's a useless component (not even good to measure North African influence in Europe and the Middle East). I asked David to do something about it a couple of years ago, and only the most outbred ones were kept (and outliers were excluded of course), which is only 4-5 Mozabites. As a result, no North African component ever appeared again since then at Eurogenes.

Same for the Hadza component, it's useless as it is (it hides their Eurasian ancestry).

And the Southwest Asian component is often problematic when it hides the known unambiguous African ancestry of Bedouins_B, which is around 7-8% IIRC.

Chad Rohlfsen
12-12-2015, 02:43 PM
SW Asian captures extra BE, and is the least SSA of any cluster that could form from pops in or near Arabia. The North African captures the recent flow into Europe, from North Africa. It is still in work. I am not done, and may try to remove the Anatolian component, looking for something more Basal that allows me to drop SW Asian and N African. This one was done as supervised.

Chad Rohlfsen
12-12-2015, 07:44 PM
Okay. Here is a bare bones, K8. This has no North African, no SW Asian. It has something about 75% BE, causing Anatolians to look WHG admixed 20-25 percent, which I think makes sense considering their location. It looks pretty clean! Take a look!

https://docs.google.com/spreadsheets/d/1wuFChh1HgwTc5ydgjcLETAP8vYixEQD13gGj9enAtTA/edit?usp=sharing

Chad Rohlfsen
12-13-2015, 01:38 AM
The K8 looks like the way we're going to go. I'm going to do a little playing around with it, just to see if it can be fine-tuned a little more. Testing should begin soon.

Dr_McNinja
12-13-2015, 02:51 AM
The K8 looks like the way we're going to go. I'm going to do a little playing around with it, just to see if it can be fine-tuned a little more. Testing should begin soon.
Shouldn't there be an "NE-Asian" type component in addition to Austronesian and Papuan?

What happened to that ASI component you had in the previous versions? Couldn't get it to stay?

Chad Rohlfsen
12-13-2015, 03:38 AM
Shouldn't there be an "NE-Asian" type component in addition to Austronesian and Papuan?

What happened to that ASI component you had in the previous versions? Couldn't get it to stay?

I can go up in K's in order to find a NE Asian, but I run the risk of breaking ENF into SW Asian and Anatolian. There is also the possibility of eating up some of the EHG. The ASI component created before, involved the Onge. I can't use them in a crowd-funding test. This does have the Paniya at 54% ENA. The Kharia are at 64%. Which is what I had them at in qpAdm, with the Onge involved. I think these two are covering it.

Chad Rohlfsen
12-13-2015, 04:01 AM
Also, if you'll notice, the Karitiana is listed here as 61.8% Austro, 36.8% EHG, 1.4% Papuan. Clovis is 60.8% Austro, 38.7% EHG, 0.5% Papuan. I think it's okay to not try and force something more NE.

Generalissimo
12-13-2015, 04:10 AM
NE Asian is unlikely to be a really ancient pop. It just seems to be a reflection of strong recent drift in Siberia and nearby areas. The Japanese, Amerindians, etc. all show very strong southeast Asian influence.

No one knows what ASI is yet. Without Mesolithic DNA from north and south India it'll remain a mystery, and there's probably little point in trying to recreate it blind, unless it helps the other clusters that are based on ancient DNA...but this isn't the case here.

It's frustrating but I'd say next year will be the year that the mystery of South Asian origins is finally solved.

Chad Rohlfsen
12-13-2015, 04:25 AM
I forgot to mention, the Nganasan is at 80.4% Austro and 19.6% EHG. There is a Karasuk individual that looks like Native Americans. They're about 70% Austro, 29% EHG, and 1% Papuan.

J Man
12-13-2015, 03:49 PM
I forgot to mention, the Nganasan is at 80.4% Austro and 19.6% EHG. There is a Karasuk individual that looks like Native Americans. They're about 70% Austro, 29% EHG, and 1% Papuan.

Nganasans and Amerindians being a mix of Austro and EHG components mainly just this tests way of saying that they are a mix of ENA and ANE ancestries?

Chad Rohlfsen
12-13-2015, 04:10 PM
Yes, and I also included them as a verification that this is working right. Nganasans typically carry the Siberian component, maxing it out. This hides their EHG/MA1 ancestry. I figured if the Karitiana were 60% ENA and 40% EHG, then the test is finding what it should.

J Man
12-14-2015, 09:53 PM
It is interesting to see that the Anatolian Neolithic and LBK EN samples score significant amounts of WHG but not EHG which makes good sense.

Chad Rohlfsen
12-15-2015, 12:41 AM
Okay. I am going to add the NE Asian component. Dstats between non-Papuans in East Asia are very significant. I think the components will be SSA, ENF, CHG, WHG, EHG, Papuan, NE Asian, and SE Asian. I'll post up the spreadsheet tonight, or tomorrow. It depends when this run finishes.

Edit: I plan on beginning tests later this week. David and I have been talking, and if you just want your admixture output, the cost will be $20 USD. If you would also like a PCA, the cost will be $30 USD.

J Man
12-15-2015, 12:52 AM
Okay. I am going to add the NE Asian component. Dstats between non-Papuans in East Asia are very significant. I think the components will be SSA, ENF, CHG, WHG, EHG, Papuan, NE Asian, and SE Asian. I'll post up the spreadsheet tonight, or tomorrow. It depends when this run finishes.

Edit: I plan on beginning tests later this week. David and I have been talking, and if you just want your admixture output, the cost will be $20 USD. If you would also like a PCA, the cost will be $30 USD.

I want both please! :D

J Man
12-17-2015, 12:44 AM
When do you think you will be ready to roll Chad?

Chad Rohlfsen
12-17-2015, 03:32 AM
Within a couple of days. I'm running qpAdm on a few groups to see how it sits.

khanabadoshi
12-17-2015, 03:51 AM
Okay. I am going to add the NE Asian component. Dstats between non-Papuans in East Asia are very significant. I think the components will be SSA, ENF, CHG, WHG, EHG, Papuan, NE Asian, and SE Asian. I'll post up the spreadsheet tonight, or tomorrow. It depends when this run finishes.

Edit: I plan on beginning tests later this week. David and I have been talking, and if you just want your admixture output, the cost will be $20 USD. If you would also like a PCA, the cost will be $30 USD.

Let's us know when and where to send the funds. :)

Chad Rohlfsen
12-17-2015, 04:20 AM
Let's us know when and where to send the funds. :)

Will do!

Krefter
12-17-2015, 05:19 AM
@Chad,

What's the difference between supervised and unsupervised? Is it that in supervised you chose what samples create the components? And how do you determine that certain runs aren't accurate?

Chad Rohlfsen
12-17-2015, 03:37 PM
@Chad,

What's the difference between supervised and unsupervised? Is it that in supervised you chose what samples create the components? And how do you determine that certain runs aren't accurate?

That's basically the difference. I tend to compare my results with Dstats and qpAdm.

Helgenes50
12-18-2015, 08:04 AM
Will do!

Chad,

How many SNPs did you use in your last run ?

Chad Rohlfsen
12-18-2015, 11:01 AM
122 k

Chad Rohlfsen
12-19-2015, 04:19 PM
I'm doing my first test run right now. I'll start taking more people tonight.

J Man
12-19-2015, 11:38 PM
I'm doing my first test run right now. I'll start taking more people tonight.

How is it looking?

Chad Rohlfsen
12-19-2015, 11:48 PM
I think we're good. I'm running myself and Helgenes, right now. I will start taking new people late tonight. I won't be able to do the PCA part, that will probably have to be done by David. I haven't had time to learn it yet. I will be sharing the files with David later tonight, as well. Zeph will be making a 4-Ancestor Oracle too. I'll start getting people pumped out as fast as possible. I will need 23andme files. FTDNA is more complicated and will probably have to be done by David. I'll post here, when I'm ready for the next set of people. I'll probably do something like take the first 4 or 5 people that like the post.

Arbogan
12-20-2015, 12:20 PM
I think I sent my files 4 months/money prematurely.

Chad Rohlfsen
12-20-2015, 02:41 PM
If you sent it, I'll get you in. I think I'll be able to take ftdna files, but I won't be able to do the PCA. I've got a pretty busy day, but ill keep running through what I have here and keep taking new people. I should be able to get 4-8 people through a day.

Arbogan
12-20-2015, 04:13 PM
If you sent it, I'll get you in. I think I'll be able to take ftdna files, but I won't be able to do the PCA. I've got a pretty busy day, but ill keep running through what I have here and keep taking new people. I should be able to get 4-8 people through a day.

I did, in September 23rd. Both files I sent were 23andme.

J Man
12-20-2015, 04:41 PM
If you sent it, I'll get you in. I think I'll be able to take ftdna files, but I won't be able to do the PCA. I've got a pretty busy day, but ill keep running through what I have here and keep taking new people. I should be able to get 4-8 people through a day.

I can send you my 23andme raw data file once I know exactly where to send it. Also are you accepting payments through PayPal or?

khanabadoshi
12-20-2015, 09:17 PM
I can send you my 23andme raw data file once I know exactly where to send it. Also are you accepting payments through PayPal or?

I'm waiting for this as well!

Chad Rohlfsen
12-20-2015, 10:06 PM
I did, in September 23rd. Both files I sent were 23andme.

Okay. I see that you sent two genomes, but $20. Could you send another $20, and I'll get you on the next run?

ZephyrousMandaru
12-20-2015, 10:08 PM
Okay. I see that you sent two genomes, but $20. Could you send another $20, and I'll get you on the next run?

Do you have a link to the spreadsheet? If so, send it to me as an e-mail and I'll create the Oracle-4 file. I sent you an e-mail.

Chad Rohlfsen
12-20-2015, 10:18 PM
Do you have a link to the spreadsheet? If so, send it to me as an e-mail and I'll create the Oracle-4 file. I sent you an e-mail.

I'll send you that here in about 4-5 hours and throw you into the next run.

J Man
12-21-2015, 01:27 AM
Where do we send our 23andme files and payments to?

Chad Rohlfsen
12-21-2015, 03:38 AM
I'll put that up here, tomorrow. David is working on something for me, then we'll be in full swing.

J Man
12-21-2015, 03:56 AM
I'll put that up here, tomorrow. David is working on something for me, then we'll be in full swing.

I know that they are basically self explanatory and maybe this is a dumb question but would you be able to give us a breakdown on the origins of each component in this test and exactly what each one represents?

randwulf
12-21-2015, 03:56 AM
Oops....I sent to the address that you published earlier in the thread, including payment, before I saw this note. I hope that will work OK. Please let me know otherwise.

Chad Rohlfsen
12-21-2015, 04:22 AM
Rand,
If it was gmail, it's probably okay. I'll double check tomorrow.

Jman,
I'll explain them all.

randwulf
12-21-2015, 04:26 AM
It was gmail.....

Chad Rohlfsen
12-21-2015, 04:29 AM
It was gmail.....

And.. [email protected] for PayPal? If so, you're good. Please, no one else send anything just yet. I'll ask for more, when I'm ready.

Chad Rohlfsen
12-21-2015, 04:30 AM
Zeph,

That spreadsheet will have to be tomorrow. I apologize for the delay, everyone. I just want this to be solid as can be.

randwulf
12-21-2015, 04:44 AM
I will send a PM....

Arbogan
12-21-2015, 03:15 PM
Okay. I see that you sent two genomes, but $20. Could you send another $20, and I'll get you on the next run?

Ok.... process one of the files. For xxxSH, and send it to me. I'll send you another 20$ later on this week for the other file.

MatAust21
12-21-2015, 05:06 PM
Chad,

Please take a look at the private message I just sent you.

I had sent you another one last week, but got no reply.

Thanks.

Chad Rohlfsen
12-22-2015, 01:38 AM
I'll resume testing later tonight or in the morning. I'm getting the files ready for David.

Chad Rohlfsen
12-23-2015, 01:28 AM
Okay. In full-swing now. I'm running through who I have, then I'll ask for more.

Chad Rohlfsen
12-23-2015, 02:18 AM
Ok.... process one of the files. For xxxSH, and send it to me. I'll send you another 20$ later on this week for the other file.

I'm running Abbas. I hope that is the one you want.

Radboud
12-23-2015, 04:17 PM
And.. for PayPal? If so, you're good. Please, no one else send anything just yet. I'll ask for more, when I'm ready.

Nevermind.

J Man
12-23-2015, 06:29 PM
Chad I am ready to rock now...Should I PM you for payment instructions or?

Chad Rohlfsen
12-23-2015, 09:18 PM
Chad I am ready to rock now...Should I PM you for payment instructions or?

Keep your eyes on the test thread for my request.