quote:There were a lot more populations included These 9 are those that have populations specific clusters in ADMXIUTRE. These 9 populations are extremely remote, highly isolated and very endogamous. The inclusion of this type of worldwide genetic diversity that is mainly not known to have contributed to Africa let alone the Nile Valley basically ate up a lot of otherwise meaningful substructure that could have better defined the affinities of the mummified remains. This again goes back to the motive of exactly how you are looking to analyze data. Repeating the example given previously, the type of populations substructure that characterized North East Africans in the studies above is useless in investigating the ancestry of a Ukrainian. It's likewise useless to include Arctic, Oceanic, East Asian, Amerindian, etc in an analysis of ancient North African mummies. Its actually worse than useless because these included populations distinguished 9 of the 16 components that could have been used to give the Middle East/North African region more resolutions. Was the goal to go deep into the analysis of the region or simply to obfuscate the region? You decide. What we can agree on is that the data was produced using a non optimal model for Africa. This is why this study has the Somali as nearly Half Yoruba+Half Natufian, a model we already know is insufficient. This is why the study has no North African component distinguishing Maghrebi populations from Levantines. How is the study purported to detect "African" ancestry from these North African mummies when it doesn't have a model that has a North African ancestral component in the first place?
Huh, I never would have considered that including all those far-flung populations in the ADMIXTURE analysis could skew the results like that. Though admittedly I am not enough of a genetics software wonk to know how that stuff works.
We've discussed this several times before when Nature's Abusir Mummy study first came out and again with the Moroccan Neolithic Study. The Abusir mummies which again hail from the Late Period of dynastic history, show a Levantine genomic profile identified with EEF. Yet we all know that said Levantine profile itself has strong African genetic influence mainly from North Africa via 'ANA' component as well as an unknown Sub-Saharan component associated with the Natufians. The fact that the Nature study not only used the YRI (Yoruba of Ibadan) as reference for all 'Sub-Saharans' is bad enough, but that they seemed to downplay any North African substructure is highly suspect.
And again just to show the inconsistency of their argument, recall the 2013 Minoan Genetic Study. According to that paper the Bronze Age Minoans showed NO ties to North Africa including Egypt but with Neolithic Southwest Asia, but then in a later study we are to believe the Egyptians show the same profile as those Minoans with no African affinities??! It's become laughable.
Posts: 26239 | From: Atlanta, Georgia, USA | Registered: Feb 2005
| IP: Logged |
We've discussed this several times before when Nature's Abusir Mummy study first came out and again with the Moroccan Neolithic Study. The Abusir mummies which again hail from the Late Period of dynastic history, show a Levantine genomic profile identified with EEF. Yet we all know that said Levantine profile itself has strong African genetic influence mainly from North Africa via 'ANA' component as well as an unknown Sub-Saharan component associated with the Natufians. The fact that the Nature study not only used the YRI (Yoruba of Ibadan) as reference for all 'Sub-Saharans' is bad enough, but that they seemed to downplay any North African substructure is highly suspect.
And again just to show the inconsistency of their argument, recall the 2013 Minoan Genetic Study. According to that paper the Bronze Age Minoans showed NO ties to North Africa including Egypt but with Neolithic Southwest Asia, but then in a later study we are to believe the Egyptians show the same profile as those Minoans with no African affinities??! It's become laughable.
LOL, yep same games being played. We saw it over 10 years ago -from dental to DNA:
and Schuenemann of course.. plus others after..
-------------------- Note: I am not an "Egyptologist" as claimed by some still bitter, defeated, trolls creating fake profiles and posts elsewhere. Hapless losers, you still fail. My output of hard data debunking racist nonsense has actually INCREASED since you began.. Posts: 5905 | From: The Hammer | Registered: Aug 2008
| IP: Logged |
posted
Just something I have repeatedly wrote about on different media. I wanted to lay it all out in detail so a laymen unfamiliar with the process could see the draw back of the study.
Posts: 2463 | From: New Jersey USA | Registered: Dec 2007
| IP: Logged |
posted
I am just using Powerpoint with the final slide saved as a jpg screenshot and trimmed as needed.
-------------------- Note: I am not an "Egyptologist" as claimed by some still bitter, defeated, trolls creating fake profiles and posts elsewhere. Hapless losers, you still fail. My output of hard data debunking racist nonsense has actually INCREASED since you began.. Posts: 5905 | From: The Hammer | Registered: Aug 2008
| IP: Logged |
quote:There were a lot more populations included These 9 are those that have populations specific clusters in ADMXIUTRE. These 9 populations are extremely remote, highly isolated and very endogamous. The inclusion of this type of worldwide genetic diversity that is mainly not known to have contributed to Africa let alone the Nile Valley basically ate up a lot of otherwise meaningful substructure that could have better defined the affinities of the mummified remains. This again goes back to the motive of exactly how you are looking to analyze data. Repeating the example given previously, the type of populations substructure that characterized North East Africans in the studies above is useless in investigating the ancestry of a Ukrainian. It's likewise useless to include Arctic, Oceanic, East Asian, Amerindian, etc in an analysis of ancient North African mummies. Its actually worse than useless because these included populations distinguished 9 of the 16 components that could have been used to give the Middle East/North African region more resolutions. Was the goal to go deep into the analysis of the region or simply to obfuscate the region? You decide. What we can agree on is that the data was produced using a non optimal model for Africa. This is why this study has the Somali as nearly Half Yoruba+Half Natufian, a model we already know is insufficient. This is why the study has no North African component distinguishing Maghrebi populations from Levantines. How is the study purported to detect "African" ancestry from these North African mummies when it doesn't have a model that has a North African ancestral component in the first place?
Huh, I never would have considered that including all those far-flung populations in the ADMIXTURE analysis could skew the results like that. Though admittedly I am not enough of a genetics software wonk to know how that stuff works.
It can and does skew the results, especially if the same algorithm you run excludes neighboring populations! This is what happened back in 2005 with the craniometric analysis of Nubians by Armelagos et al. using the FORDISC 2 program, with some crania classified as Spaniard, Polynesian, Ainu etc.
This is why when it comes to statistical studies using principle component analysis, it is always important to have a good number of samples and in particular equally differentiated samples that are spatially close to the subject sample. This is why I've been critical of Hanihara's cranial multiple discriminant analysis because he does the same thing generalizing Africans into a 'Sub-Saharan' (true negro) archetype while grouping both Egyptians and Nubians with Southwest Asian (caucasoid) archetypes. When samples aren't properly grouped in their respective regions, and samples from far afield are unnecessarily including it leads to stacking the deck i.e. obfuscation to get skewed results. Now the same thing is being done with genetic analysis!
Fortunately smart people (not just the experts) but laypeople who how to properly assess the data are pointing this out.
-------------------- Mahirap gisingin ang nagtutulog-tulugan. Posts: 26239 | From: Atlanta, Georgia, USA | Registered: Feb 2005
| IP: Logged |
Nice write-up. Just wondering, with all this new software that genetic bloggers are using by themselves and the wide availability of published genomes, I have never seen an official genetic study published by "non-experts" or scientists not associated with major universities or genetic labs. Why is that?
Genteic technology needs to be made widely available or more democratic and used by non-genetic experts or we will just keep getting more and more crackpot history, such as Philistines or Sea peoples giving birth to modern-day Ethiopians.
Posts: 288 | From: Asia | Registered: Mar 2016
| IP: Logged |
Nice write-up. Just wondering, with all this new software that genetic bloggers are using by themselves and the wide availability of published genomes, I have never seen an official genetic study published by "non-experts" or scientists not associated with major universities or genetic labs. Why is that?
Genteic technology needs to be made widely available or more democratic and used by non-genetic experts or we will just keep getting more and more crackpot history, such as Philistines or Sea peoples giving birth to modern-day Ethiopians.
I would think We probably dont see it because of the costs of getting into a "Reputable" journal. There is one good instance i can think of though. When THIS INDIVIDUAL published THIS ARTICLE which is actually a good read right now. Its old and only based on published data but it is what it is. THe other stuff that going on in the enthusiast community seems to be just posted to their personal blogs.
The issue i have with these new genetic models is that they are largely statistical and get to the point where they are not reinforced by any known interpretation of history or multidisciplinary evidence IE Sea Peoples associated with Sardinians migrating into the Horn of Africa.
Posts: 2463 | From: New Jersey USA | Registered: Dec 2007
| IP: Logged |
Abusir el-Meleq should probably be specified since it's a different site from Abusir
With such a wide range of halplotypes and not all late period how useful is an admixture analysis?
Posts: 42920 | From: , | Registered: Jan 2010
| IP: Logged |
Abusir el-Meleq should probably be specified since it's a different site from Abusir
With such a wide range of haplotypes and not all late period how useful is an admixture analysis?
When i say "Abusir" everyone knows what i am talking about. My analysis was written specifically for someone like you. Someone that doesn't know how ADMIXTURE works. What it does, what it doesn't do. How it can be used to clarify or obfuscate. How it came be used to produce a model in which Ethiopians are 81% European and 15% African. You need the know the HISTORY of the technique, how it has been used in the past and how they use it in the article.
How useful is admixture analysis in face of that maternal diversity? Are you really asking that question in 2021? Do the world a favor and read the passage. Is ad free.
Posts: 2463 | From: New Jersey USA | Registered: Dec 2007
| IP: Logged |
I was reading your insightful breakdown in the link, and thanks for it. My question is about this unique North African lineage that is found in modern Copts in Egypt and Sudan. Is this element found in any of the tested mummies so far?
posted
The genetic variation of the world being as great as it is does not fit into neat boxes. So forcing a computer to designate a portion of any given genome as belonging to or having a special affinity to any certain regions can and will produce questionable results. How useful is admixture analysis in face of the maternal diversity of these mummies? How informative is the ADMIXTURE program as compared to other genetic analysis? So you think it would be an excellent tool if it had more African substructure input? That seems hard to measure though
look, I'm more against ADMIXTURE then you are
Posts: 42920 | From: , | Registered: Jan 2010
| IP: Logged |
posted
@The Lioness - See this is why you should have spend years learning instead of trolling. What you wrote is incoherent gobbledegook. You have been here all this time, shame.
@ Big O - They didn't test for that type of component because the algorithm they used didn't generate any component specific to North Africa. The way they chose to include certain samples forced the clustering into this situation.
If you see my example with the Ethiopian genome on 23andMe circa 2008 its only 15% African. Lets pretend for a moment this is an Axumite Empire sample. Lets assume modern Ethiopians taking this same test, using this SAME failed model and are 30% African. The data could then be used to argue that Ethiopians "Became more African" After the collapse of the Axumite empire.(In the same way Egyptians "became more African since Roman times".)
A proper study that contains East Africans, and generates a model with an East African component would better explain the differences between the genomes as they (ancient and modern) could BOTH be majority East African in ancestry and what is being categorized as European or Asian in the old model was simply different types of African Ancestry. TO make matters worse lets assume you had 3 Ethiopians samples that had that same 85% "European" ancestry in that model. Lets say the Axumite empire had wide ranging contacts in the Maghreb and Egypt. After you run the proper model, each of these samples could be distinctly Egyptian, Maghrebi or East African if all these ancestries are dumped into a "European" wastebasket using the failed model.
This was the case with Abusir with a least one of these mummies being particularly more North African than the others.
Posts: 2463 | From: New Jersey USA | Registered: Dec 2007
| IP: Logged |
posted
Beyoku, think this is germane to your point:
"Abstract Principal Component Analysis (PCA) is a multivariate analysis that allows reduction of the complexity of datasets while preserving data's covariance and visualizing the information on colorful scatterplots, ideally with only a minimal loss of information. PCA applications are extensively used as the foremost analyses in population genetics and related fields (e.g., animal and plant or medical genetics), implemented in well-cited packages like EIGENSOFT and PLINK. PCA outcomes are used to shape study design, identify and characterize individuals and populations, and draw historical and ethnobiological conclusions on origins, evolution, whereabouts, and relatedness. The replicability crisis in science has prompted us to evaluate whether PCA results are reliable, robust, and replicable. We employed an intuitive color-based model alongside human population data for eleven common test cases. We demonstrate that PCA results are artifacts of the data and that they can be easily manipulated to generate desired outcomes. PCA results may not be reliable, robust, or replicable as the field assumes. Our findings raise concerns on the validity of results reported in the literature of population genetics and related fields that place a disproportionate reliance upon PCA outcomes and the insights derived from them. We conclude that PCA may have a biasing role in genetic investigations. An alternative mixed-admixture population genetic model is discussed."
quote:Originally posted by Mansamusa: Beyoku, think this is germane to your point:
"Abstract Principal Component Analysis (PCA) is a multivariate analysis that allows reduction of the complexity of datasets while preserving data's covariance and visualizing the information on colorful scatterplots, ideally with only a minimal loss of information. PCA applications are extensively used as the foremost analyses in population genetics and related fields (e.g., animal and plant or medical genetics), implemented in well-cited packages like EIGENSOFT and PLINK. PCA outcomes are used to shape study design, identify and characterize individuals and populations, and draw historical and ethnobiological conclusions on origins, evolution, whereabouts, and relatedness. The replicability crisis in science has prompted us to evaluate whether PCA results are reliable, robust, and replicable. We employed an intuitive color-based model alongside human population data for eleven common test cases. We demonstrate that PCA results are artifacts of the data and that they can be easily manipulated to generate desired outcomes. PCA results may not be reliable, robust, or replicable as the field assumes. Our findings raise concerns on the validity of results reported in the literature of population genetics and related fields that place a disproportionate reliance upon PCA outcomes and the insights derived from them. We conclude that PCA may have a biasing role in genetic investigations. An alternative mixed-admixture population genetic model is discussed."
UBBFriend: Email this page to someone!
![[Confused]](confused.gif)
Are you really asking that question in 2021? ![[Roll Eyes]](rolleyes.gif)
Printer-friendly view of this topic