tag:blogger.com,1999:blog-3023805782808412230.post4247883009936746162..comments2024-03-09T15:46:44.638+01:00Comments on For what they were... we are: Major Guanche genetic influence in Puerto Ricans (guest article by Thierno)Majuhttp://www.blogger.com/profile/12369840391933337204noreply@blogger.comBlogger59125tag:blogger.com,1999:blog-3023805782808412230.post-62907743973513782902024-03-08T18:30:16.529+01:002024-03-08T18:30:16.529+01:00Nassim Taleb et al. just published a new version o...Nassim Taleb et al. just published a new version of their paper called “Informational Rescaling of PCA Maps with Application to Genetic Distance” which aims to improve the underlying mathematical statistics of genetic distances. The distances between populations in conventional PCA are significantly changed with this entropy based method. Hundreds of thousands of published admixture results could potentially be revised under this new method. <br />Link: https://arxiv.org/pdf/2303.12654.pdf<br /><br />Here’s an interesting finding regarding Latin American samples from the 1000 Genomes Project: <br /><br />“For example, for Puerto Ricans (PUR) and Colombians (CLM) conventional PCA spreads them along the beginning of the African cline, whereas rescaling shows them in the vicinity of other Latin American populations (Mexicans and Peruvians). Iranians, Turkish, Palestinian, Druze, French, Iberian (IBS), British (GBR), Russian, Finnish, Puerto Rican, and the majority of Colombians all form a much tighter cluster in the rescaled PCA, indicating that these populations are not as far from each other as the conventional PCA suggests.”<br />Thiernohttps://www.blogger.com/profile/13452118643827292940noreply@blogger.comtag:blogger.com,1999:blog-3023805782808412230.post-24915922475395243882023-09-09T12:36:33.989+02:002023-09-09T12:36:33.989+02:0040 new ancient Canarian samples that are estimated...40 new ancient Canarian samples that are estimated to have lived from the 3rd to the 16th century were analyzed in this new paper from Serrano et al (2023), “The genomic history of the indigenous people of the Canary Islands.” As previously observed, the ancient Canarians are still very divergent from Paleolithic/Early Neolithic Moroccans, and cluster more closely with Late Neolithic Moroccans and present-day North African populations. <br /><br />However, all present-day Northwest Africans have less ancient North African, but a lot more West Asian & African ancestry compared to those ancient Canarian samples, and all of them are also very similar to one another (including Sahrawis). Sub-Saharan African ancestry was still detected in some of those ancient Canarian samples, suggesting the existence of trans-Saharan migrations in North Africa already before the Common Era. Moreover, those ancient Canarian samples seem to be divided in roughly 2 clusters. One located in the western islands that’s more closely related to Upper Paleolithic and Early Neolithic North Africans and the other in the eastern islands to European populations.<br /><br />https://www.nature.com/articles/s41467-023-40198-wThiernohttps://www.blogger.com/profile/13452118643827292940noreply@blogger.comtag:blogger.com,1999:blog-3023805782808412230.post-38071094121867547162023-07-01T05:34:13.169+02:002023-07-01T05:34:13.169+02:00This new paper, “Northwest African Neolithic initi...This new paper, “Northwest African Neolithic initiated by migrants from Iberia and Levant” from Simões et al. (2023), shows that the Guanches from the Canary Islands are much more similar to Late and Middle Neolithic North Africans (KEB), than they are to Iberomaurusian/Taforalt, Epipaleolithic and Early Neolithic (KTG & IAM) North Africans. PCA analysis (Fig.1) also shows that the Guanches are shifted towards ancient Levantine populations. Interestingly, the Guanches’ ancestral components are also very similar to those of modern-day Northwest African samples that were used in this analysis: Saharawi and Mozabite populations from the Simons Genome Diversity Project (SGDP) dataset.<br /><br />All this information suggests that the ancestry of the Guanches, Saharawi people and other geographically similar Northwest African populations is unlikely to be different from other modern-day coastal Northwest Africans. <br /><br />https://www.nature.com/articles/s41586-023-06166-6#MOESM1Thiernohttps://www.blogger.com/profile/13452118643827292940noreply@blogger.comtag:blogger.com,1999:blog-3023805782808412230.post-50102984210848461022023-04-16T17:05:27.261+02:002023-04-16T17:05:27.261+02:00The findings of this new paper called “Echoes from...The findings of this new paper called “Echoes from the last Green Sahara: whole genome analysis of Fulani, a key population to unveil the genetic evolutionary history of Africa,” contradict the hypothesis that the NW African component of Fulani is solely due to recent admixture between western African and northern African groups.<br />https://www.biorxiv.org/content/10.1101/2023.04.06.535569v1<br /><br />The so-called Muslim expansion south of the Sahara in historical times as the only explanation for the origin of Fulani people is not credible, especially considering these admixture, qpWave, qpAdm and older Y-DNA analyses. They reveal much more complex and ancient phenomena which may have been more widespread across the Sahara and Sahel than previously anticipated. The advantage of their analyses is that they used more than 1800 samples, including ancient DNA samples such as Iberomaurusians, from the relevant geographic regions and continents. This contrasts with other DNA studies about these populations, which were more limited both in scope and in sequencing depth & coverage. <br /><br />Some of the findings in this paper confirm what I had long suspected: that there could be some deeper & older underlying phenomena associated with the genetic origins of Fula people and similar groups across the Sahel that have been largely unexplored up to this point. This is in part why, in relation to the Puerto-Rican/Guanche connection with my data, I had limited myself by simply implying that it could suggest some complex genetic links with other Northwest African populations and didn’t mention recent historic Muslim expansion south of the Sahara as a hypothesis for disentangling the complex origins of Fulani people. At best, I considered it to be just anecdotal depending on each person’s ancestry.<br />Thiernohttps://www.blogger.com/profile/13452118643827292940noreply@blogger.comtag:blogger.com,1999:blog-3023805782808412230.post-74103649492902887382019-03-16T22:24:55.540+01:002019-03-16T22:24:55.540+01:00@Fonte
Thanks for the ideas that you shared with m...@Fonte<br />Thanks for the ideas that you shared with me privately, about exploring “matching patterns” between samples. I familiarized myself with some notions from various sources. PLINK v1.07 has an old section aimed for pairwise IBD estimation and the detection of shared segments. <br />I simply reused the samples from the analysis. So 100 Iberians, 70 Puerto Ricans, 17 Canarians, 17 Andalusians and 17 Galacians. Shared SNPs between Puerto Ricans and Iberians for the analysis are above 150K, while shared SNPs between Puerto Ricans and Canarians/Andalusians/Galacians is around 40K. I set a minimum threshold of 5 cM and allowed for all pairs to be reported. <br /><br />As expected, the counts of the number of pairs sharing cumulative IBD are the highest between Puerto Rican samples. In terms of frequency, it appears that single segments are prevalent between Puerto Ricans and the rest of the Iberian and Canarian samples.<br /><br />Probably due to the disparity between the number of Iberian and Canarian samples, only 4 Canarians match with 3 Puerto Ricans. In spite of this, intriguingly, one of those matching segments between 1 Puerto Rican and another Canarian, is 51cM. It the highest of the entire analysis. According to various sources, the larger the segment the closer the relationship, while high frequency shared segments are more likely to indicate distant sharing at the population level.Thiernohttps://www.blogger.com/profile/13452118643827292940noreply@blogger.comtag:blogger.com,1999:blog-3023805782808412230.post-7712063158911591182019-03-14T19:04:17.448+01:002019-03-14T19:04:17.448+01:00Article got a new update or "follow up" ...Article got a new update or "follow up" by Thierno.Majuhttps://www.blogger.com/profile/12369840391933337204noreply@blogger.comtag:blogger.com,1999:blog-3023805782808412230.post-22090671387786959552019-01-20T20:51:41.572+01:002019-01-20T20:51:41.572+01:00Sure. You are welcome.
My sample is in the exerc...Sure. You are welcome. <br /><br />My sample is in the exercises of the post, so issues in relation with the genetic origins of Fula people eventually needed to be addressed with clarifications.<br /><br />In the French speaking world, I realized that many people haven’t been exposed to this information. Eventually, one day maybe, I would like to share all this information somewhere, this time in French.<br /><br />Except for the presence of Fula communities in other non-West African countries and in some cases, some inter-mixing with other communities, depending on private anecdotal examples, unfortunately, many fictitious stories have been perpetuated on the so-called origins of the Fulbe people. Many historical accounts of so-called origins or migrations aren’t always true.Thiernohttps://www.blogger.com/profile/13452118643827292940noreply@blogger.comtag:blogger.com,1999:blog-3023805782808412230.post-41082690407812603202019-01-17T17:53:50.454+01:002019-01-17T17:53:50.454+01:00Thanks again for replying so thoroughly, Thierno.
...Thanks again for replying so thoroughly, Thierno.<br /><br />I must say that I'm very pleased that those "quickies" I had already forgotten about have been of some use to you. It's always a moral gratification when you know that your little contributions have some value for others.<br /><br />Cheers.Majuhttps://www.blogger.com/profile/12369840391933337204noreply@blogger.comtag:blogger.com,1999:blog-3023805782808412230.post-54912200983305098732019-01-17T17:30:21.891+01:002019-01-17T17:30:21.891+01:00*"On your own" (I don't think "...*"On your own" (I don't think "on yourself" is a valid English phrase, sry).Majuhttps://www.blogger.com/profile/12369840391933337204noreply@blogger.comtag:blogger.com,1999:blog-3023805782808412230.post-47955647401500579062019-01-17T17:29:29.175+01:002019-01-17T17:29:29.175+01:00Thanks for explaining that, Thierno. I must admit ...Thanks for explaining that, Thierno. I must admit I gave you that lead and then you figured out all the rest on yourself. So sure: better follow your references. Majuhttps://www.blogger.com/profile/12369840391933337204noreply@blogger.comtag:blogger.com,1999:blog-3023805782808412230.post-70153784739822031382019-01-15T19:31:42.600+01:002019-01-15T19:31:42.600+01:00Hello Fonte,
Thanks for your input. It’s good qu...Hello Fonte, <br /><br />Thanks for your input. It’s good question, but it’s really hard to say. This notion of “matching” with the data of someone else, is unclear.<br /> <br />What is the methodology behind the so-called science which helps to determine how testers match with each other? It’s an unknown field for me.<br /><br />For examples, I have a known relative who tested and who shares 32 centimorgans with my data. A decent number of other unknown testers (most likely Fulani), possibly from other countries, share higher values with my data. It’s confusing.<br /><br />Genetic studies, usually don’t sample as many people from 1 ethnic group or 1 geographic area. They would possibly use alder or malder in order to detect significant admixture signals, and ultimately deduce time estimates.<br />And possibly qpAdm and f4/D stats for admixture models. <br /><br />https://github.com/DReichLab/AdmixTools<br /><br />I think that there are really 2 different approaches to population genetic. And this explains why there is so much confusion, it seems.<br /><br />The methods and tools that are used by private commercial DNA companies and academic scientific studies are different, because the expectations aren’t the same.<br /><br />Commercial DNA deals with recent and private genealogy. It has to be more “palpable” for the customer base, so it won’t get in the depth of admixture analysis. Just enough to have discernible continentally and/or regionally separated categories. Of course, blog posts like yours add values to it because you collect results and look for patterns.<br /><br />On the other hand, academic studies that are produced by geneticists are not designed for this sort of individualized and recent genealogical background. They explore and analyze admixture more deeply with a much older time span.<br />Thiernohttps://www.blogger.com/profile/13452118643827292940noreply@blogger.comtag:blogger.com,1999:blog-3023805782808412230.post-13995918246152102142019-01-15T17:21:22.489+01:002019-01-15T17:21:22.489+01:00I remember when I first checked this link. Unfortu...I remember when I first checked this link. Unfortunately, many links of Khan's post had already expired. I couldn't even find the dataset for the populations.<br />So I skipped it, and went straight to use the manuals from ADMIXTURE, PLINK, and R.<br />I would suggest to check the links of the appendix section from this post.<br /><br />This site has many datasets (incl. from Lazaridis): https://reich.hms.harvard.edu/datasets<br /><br />For ADMIXTURE: http://software.genetics.ucla.edu/admixture/<br /><br />Thiernohttps://www.blogger.com/profile/13452118643827292940noreply@blogger.comtag:blogger.com,1999:blog-3023805782808412230.post-31936040229880444662019-01-15T08:43:41.070+01:002019-01-15T08:43:41.070+01:00…
The Y-DNA study from the paper of D’Atanasio 201...…<br />The Y-DNA study from the paper of D’Atanasio 2018, suggests a strong correlation between haplogroup E-M2_Z15939 and Fulani communities from various areas in the Sahel.<br /><br />My closest matches on the database of commercial DNA, are overwhelmingly Fulani people from various countries. Senegal, Nigeria, Sierra Leone, Mauritania and Mali. All the way to Saudi Arabia.<br />I had the opportunity to write emails with some of them. Typically they date the presence of their families (in most cases the Adamawa clan) in areas such as Northern Nigeria, or the Red Sea area, no earlier than 2 centuries ago. And it would correspond to the historical events which occurred in West Africa, 2 centuries earlier.<br /><br />https://en.wikipedia.org/wiki/Fula_jihads<br /><br />Considering the timescale in the field of genetic, that’s almost no time. It’s very recent. I just don’t see how in 2 centuries, various Fulani communities which abandoned a long time ago their nomadic lifestyle and inter-mixed with other local ethnic groups could have formed other so-called specific components. <br /><br />When I used alder and malder, I progressively modified the dataset and ended up eliminating sources for which I have no admixture with. Based on my admixture results, I only selected the following populations: Mandenka;Yoruba;Gambian;Esan;Mende;Moroccan;Saharawi;Tunisian;Algerian;Mozabite<br /><br />The only successful exercise with a 2-ref weighted LD significant curve was obtained with the Yoruba and the Moroccan samples. It failed for the other samples. <br />It estimated 18.76 +/- 9.07 generations for the signal.<br />Even using Maju’s correction of time estimate, by 1.5 or 2, it falls deep into the period of the Berber dynasties and presence in the region. <br /><br />https://en.wikipedia.org/wiki/History_of_Morocco#Berber_dynasties_(c._1060_–_1549)<br /><br /><br />“ how you ran ADMIXTURE and PLINK to do this, I did this many years ago, and need a refresher.”<br /><br />I downloaded my data ( around 690k of snp) from the DNA company. I converted it to the PLINK format. <br />Various website have scripts and tools which you can download and use.<br /><br />http://www.y-str.org/p/tools-utilities.html<br /><br />I installed Ubuntu. It makes thing easier.<br />I found the data from the published studies. Check the links in the appendix section.<br />Also check: https://reich.hms.harvard.edu/datasets<br />It has datasets from Lazaridis.<br /><br />You would have to become familiar with PLINK and its manual just like with the ones from ADMIXTURE and R. <br />The genetic and physical positions of your files and the data you want to merge have to match, depending on the type of genomic assembly. <br />The type of SNPs from the data you want to merge, has to correspond to the ones of the dataset you are interested in.<br />Ideally, you want to perform ADMIXTURE’s cross-validation procedure because it will help to know which K value has the highest CV errors.<br /><br />http://software.genetics.ucla.edu/admixture/<br /><br />I won’t lie to you, it’s a hassle and time-consuming lol. Running it using public datasets is not as bad, but merging private commercial data to them can be challenging, especially if you are trying to keep a decent number of SNPs.<br /><br />“Sampling strategy” is tricky.<br />It doesn’t seem like it at first, but it’s perhaps the most challenging part, in my opinion. <br />You have to be aware of previous studies and what has been established already for the populations you are interested in.<br />Thiernohttps://www.blogger.com/profile/13452118643827292940noreply@blogger.comtag:blogger.com,1999:blog-3023805782808412230.post-53441869070010325192019-01-15T08:35:43.024+01:002019-01-15T08:35:43.024+01:00…
“may there be multiple fulani-specific north-afr...…<br />“may there be multiple fulani-specific north-african components? and perhaps your clan has a different one from the folks tested? “<br /><br />It’s not impossible and it can’t be dismissed, considering that all existing Fulani communities haven’t been sampled, tested and compared to each other. <br />However, based on the accumulated and most recent serious evidence on the subject, there is systematically a common old trans-Saharan component. <br />If there are differences, I would say that it is mostly due to inter-mixing with local populations of the geographic area where they live.<br /><br />There are several studies other than the ones that I have already mentioned to you with links.<br /><br />This study, “Begona Dobon et al., The genetics of East African populations: a Nilo-Saharan component in the African genetic landscape. Nature – Scientific Reports 2015" <br /><br />Here is the link: http://www.nature.com/articles/srep09996 <br /><br />Maju made a post on it some years ago.<br /><br />http://forwhattheywereweare.blogspot.com/2015/08/sudan-and-south-sudan-autosomal-genetics.html<br /><br />If you observe the Fulani samples from Sudan, at K=3, they shared ancestry with local Sudanese ethnic groups. At k=5, however, they display an additional unique specific component of the color pink. The authors have suggested: <br />“A population that shows signals of recent admixture is the Fulani. Fulani are nomadic pastoralists who speak a Niger-Kordofanian (Niger-Congo) language and occupy a large area in Africa’s Sahel. Their origin is still controversial, as mitochondrial DNA indicates a West African and traces of North African origin23, whereas Y-chromosome studies showed shared ancestry with Afro-Asiatic and Nilo-Saharan Sudanese populations8. This shared ancestry with East African populations can be seen in Fig. 3 (k = 3), suggesting that they have admixed with local populations.”<br /><br />Secondly, the paper Gurdasani et al.2014 showed in figure 7 b. an exercise which included Fula from Gambia, dating an eurasian admixture around 100 generations ago and a second one btw 320-780 ago.<br /><br />Lastly, the paper Busby et al., 2015, Figure 6-figure supplement 1 shows that the penetration of IBS and TSI gene flow in Africa before 0CE was via Northeast Africa. <br />The authors stated that that they had identified some southern European ancestry in the Fulani which suggested that it might have entered West Africa via North East Africa. <br /><br />This is a passage from the paper: “The Fulani, a nomadic pastoralist group found across West Africa, were sampled in The Gambia, at the very western edge of their current range, and have previously reported genetic affinities with Niger-Congo speaking, Sudanic, Saharan, and Eurasian populations [Tishkoff et al., 2009; Henn et al., 2012], consistent with the results of our mixture model analysis (Figure 4A). Admixture in the Fulani differs from other populations from this region, with sources containing greater amounts of Eurasian and Afroasiatic ancestry, but appears to have occurred during roughly the same period (c. 0CE; Figure 5).” <br /><br />Check this link here: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4915815/<br /><br /> or this one https://elifesciences.org/articles/15266<br /><br />On this blog, Maju made a quick post on it.<br /><br />http://forwhattheywereweare.blogspot.com/2016/08/african-admixture-events.html<br /><br />This is likely the same source of admixture. So already, there are Fulani samples from Nigeria, Sudan, Gambia and my anecdotal data for what it’s worth, since my grand parents were from the Fouta Djallon region of present-day Guinea. <br /><br />...<br />Thiernohttps://www.blogger.com/profile/13452118643827292940noreply@blogger.comtag:blogger.com,1999:blog-3023805782808412230.post-70758713880877381852019-01-15T08:23:57.131+01:002019-01-15T08:23:57.131+01:00…
“But basically, the "north-african" s...…<br /><br />“But basically, the "north-african" segmetn as it shows on my grandpa on CHR 12, from position 11 to 21. Shows as "senegambian and guinean" for his mexican match. So it seems in my case the north-african I share with fulanis is fulani specific.”<br /><br />On gedmatch, what I referred to is in the Admixture utilities section, where it says “Paint differences between 2 kits, 1 chromosome”. Using Dodecad, for example, which has the North African category that’s created not just based on Mozabite samples, but also on Moroccan ones. The chromosome painting tool, is precisely what I used. I didn’t identify this restricted and shared common North African component, by just relying on chromosome positions. It doesn’t inform on the type of admixture that’s shared with other matches.<br />Based on the information that you provided, you have no evidence to claim a fulani-specific component. If you don’t properly compare your data using the unsupervised mode of ADMIXTURE with the right populations, there is no way to know.<br /><br />“I would also add, that you say your north-african component did not get swalloed by the fulani-specific component, “<br /><br />It’s what the results from the experiments with my data suggest. That’s the idea behind those investigations. To make hypotheses, based on the outputs.<br />...Thiernohttps://www.blogger.com/profile/13452118643827292940noreply@blogger.comtag:blogger.com,1999:blog-3023805782808412230.post-40187987545356030412019-01-15T08:19:29.895+01:002019-01-15T08:19:29.895+01:00…
“THe fulanis in most datasets are from Burkina ...…<br /><br />“THe fulanis in most datasets are from Burkina Faso to my understanding.”<br /><br />Please check the links that I shared with you. They are full of information. The Fula samples from the Henn’s study, from Maju’s exercises, and from my work are from Nigeria. Check the map of the study.<br />I am not aware of any study on the autosomal DNA of Fulani, from Burkina Faso.<br /><br />I wouldn’t want to throw the baby out with the bathwater. I understand that for many Afro-descendants, as well as other communities, commercial DNA tests allow them to get information about the origin of their enslaved ancestors and/or immigrant ancestors.<br />However, I think that excessive wishful thinking without credible evidence, may sometimes get in the way of improving our knowledge.<br /><br />In the conversations, I had initially with Maju, on the genetic origin of Fulani people, I shared with him narratives that are commonly found among different Fulani communities, that I've always heard without ever having the opportunity to verify them. <br />When he showed me scientific evidence (genetic studies, his exercises), which oriented and pointed to older and different sources of likely origins, and in the process kept mentioning ADMIXTURE, I was interested to learn on my own, how to use it, and verify my private autosomal DNA more accurately. <br /><br />It led me to compare my data with other Fulani samples. My understanding of the complex genetic origin of Fulani people has improved, as a result. <br /><br />In the field of genetic, the benefit of science, is also to have reliable standards for acknowledging some hypotheses or claims. Of course, new studies or new findings can always help modify, refine current knowledge or orient in different directions.<br /><br />After ADMIXTURE, I experimented with many types of software that are used in the recent studies on autosomal DNA by current geneticists. Naturally, each of them has its pros and cons, and ultimately its own limitations, depending on a myriad of factors and what’s expected from the user. For present-day modern populations, with those peculiarities, ADMIXTURE works fine, if “the sampling strategy” makes sense. <br />For ancient populations, qpAdm, f4/D stats might be better adapted than ADMIXTURE, just to give an example. Checking for standard errors and adjusting to it, is also the key.<br />...Thiernohttps://www.blogger.com/profile/13452118643827292940noreply@blogger.comtag:blogger.com,1999:blog-3023805782808412230.post-51236454413862130342019-01-15T07:59:26.398+01:002019-01-15T07:59:26.398+01:00“I am seeing this fulani-specific component in the...“I am seeing this fulani-specific component in the NEW 23andme being clustered under s enegambian. FOr example. my grandpa with all the Fulani matches , overlaps with 5 Mexicans, and these are large segments i'm talking aobut 15-21 cm. “<br /><br />Your description of “fulani-specific” component in the context of the so-called “Senegambian” category from 23andme, is very incorrect.<br /><br />https://customercare.23andme.com/hc/en-us/articles/212169298-Reference-Populations#Sub-Saharan_African<br /><br />The category called “Senegambian & Guinean”, according to their own description, groups people from Senegal, Gambia, Guinea-Bissau and Guinea. Needless to say, that it has many different ethnic groups that are found all across the region. In addition, the method used by commercial DNA companies produces imprecise results.<br /><br />This is what I wrote to Fonte on this issue: “the tested data, as well as the preselected samples, are isolated in categories ( such as Senegambian from 23andme) that have been "arbitrarily" selected. They don't let each individual sample form itself. So all the existing individual variations of the dataset are ignored. This element, I suspect, increases errors, and produces more bias issues.”<br /><br />The Fulani-specific component which was identified in the Henn’s study at k=10, by Maju, as well as in my work, is produced by using the unsupervised mode of ADMIXTURE, where individual members of specific “ethnic categories” are distinctively identified. <br />It’s completely different from the approach that consists of selecting people on the basis of recently established borders/countries or regions which aren’t reflecting the genetic or even the linguistic diversity. This is why, I suggested to you and others to become more familiar with the history of the region. <br />Commercial DNA test can differentiate between continentally and regionally (in some cases) separated populations. With endogamous or ancient populations, the true ancestral components won’t be detected and identified.<br /><br />...Thiernohttps://www.blogger.com/profile/13452118643827292940noreply@blogger.comtag:blogger.com,1999:blog-3023805782808412230.post-19946369157446369992019-01-14T16:36:49.119+01:002019-01-14T16:36:49.119+01:00Just for the record, the basic instructions on how...Just for the record, the basic instructions on how to use Admixture were explained by Razib Khan in this 2011 entry: http://blogs.discovermagazine.com/gnxp/2011/03/analyzing-ancestry-with-admixture-step-by-step/Majuhttps://www.blogger.com/profile/12369840391933337204noreply@blogger.comtag:blogger.com,1999:blog-3023805782808412230.post-59802922976993685732019-01-14T16:34:11.660+01:002019-01-14T16:34:11.660+01:00You can replicate the exercise at home, all the so...You can replicate the exercise at home, all the software and samples are public domain and gratis. It just has a mild learning curve that should be no major issue for any interested person with a bit of patience. <br /><br />Of course the seal of approval of formal peer-reviewed research would add some clout of greater credibility but this is as scientific as it gets, including the possibility of replication by anyone. The more the merrier anyhow, so I do hope for more people independently researching this issue in the future, as it has some interest for Puerto Ricans and other populations (Canarians, Berbers, other Caribbean peoples that might have similar results, etc.)Majuhttps://www.blogger.com/profile/12369840391933337204noreply@blogger.comtag:blogger.com,1999:blog-3023805782808412230.post-65776909022488092862019-01-14T16:27:29.320+01:002019-01-14T16:27:29.320+01:00In general the problem with commercial autosomal D...In general the problem with commercial autosomal DNA analyses is that it is too generic: they pre-select "artificial" model reference components (what I call "zombies", because they are not even actual populations but "distilled" components that are most concentrated in this or that area but often shared among many). So it may be an OK-ish preliminary approach but one that cannot replace properly designed unsupervised runs with an optimal "sampling strategy". <br /><br />In this case I "tutored" Thierno on how to design the ideal sampling strategy: we need of course some Native Americans and all the rest we need are samples from West Africa, NW Africa and SW Europe (Iberia). All the rest, say Swedes or Chinese or peninsular Arabs or Indians... are in excess because we know already Puerto Ricans don't have any meaningful inflow from those areas, they are not part of the question itself and can clutter the analysis in various ways, causing noise, etc. <br /><br />This is particularly true of "endogamic" populations (i.e. populations that have suffered severe bottlenecks, that clearly have low internal genetic diversity). For example many DNA testing companies use Ashkenazi Jews and people go literally nuts because they get some X% (usually small and meaningless) of that affinity, when the problem is that such endogamous popuplations tend to create their own artificial clusters or components (especially if oversampled) and attract others to their group in what is clearly an artifact, a bug and not a desirable feature of the algorithm. Some of that can happen with Mozabites and it definitely happens with the 1000Genomes Tunisian sample, which is from some endogamic population of Southern Tunisia and always gives weird results. It happens with Finns in European analyses, etc. <br /><br />Anyhow, finding this result was relatively hard. It is clearly there and is notable but it took several trials with different sampling strategies to have a clear result. The simplified straightforward sampling strategy I proposed: some NW Africans, some "controls" (Iberians, West Africans and Native Americans), worked best in the end. Majuhttps://www.blogger.com/profile/12369840391933337204noreply@blogger.comtag:blogger.com,1999:blog-3023805782808412230.post-38892143718815500162019-01-12T23:56:50.863+01:002019-01-12T23:56:50.863+01:00" Independent research just published HERE sh..." Independent research just published HERE shows that about half of the ancestry of Puerto Ricans is Berber (North African) rather than Iberian and thus that the Guanche or Canario ancestry hypothesis posited by Charles is almost certainly correct. "<br /><br />I wonder how any future research into autosomal DNA matching patterns might corroborate or refine these findings. Research conditions might not be ideal yet. But let's assume a customer database exists with 1000 Canarian DNA testers (preferably with increased Guanche lineage) as well as 1000 Spanish DNA testers (preferably from Spanish regions historically known to have been most involved in the colonization of the Hispanic Carribean). <br /><br />When compared with a representative sample group of 1000 Puerto Rican DNA testers would one then indeed observe a nearly equal number of DNA matches between Puerto Ricans and Canarians on the one hand and between Puerto Ricans and Spaniards on the other? And would the average size of shared DNA indeed also be nearly equal for the matches between Canarians and Puerto Ricans when compared with the matches between Puerto Ricans and Spaniards? <br /><br />Or will additional and more complex ancestral scenarios complicate any straightforward comparison?<br /><br />Also I'm wondering if maybe the time framing of absorption of either Canarian or Spanish lineage among Puerto Ricans would be a major factor? Manifesting itself also in such DNA matching patterns. A parallel to be drawn perhaps with the fascinating time framing applied in Moreno-Estrada et al.(2013).Fonte Felipehttps://www.blogger.com/profile/14718819493327107246noreply@blogger.comtag:blogger.com,1999:blog-3023805782808412230.post-60049520036769014232019-01-12T23:46:06.617+01:002019-01-12T23:46:06.617+01:00This comment has been removed by the author.Fonte Felipehttps://www.blogger.com/profile/14718819493327107246noreply@blogger.comtag:blogger.com,1999:blog-3023805782808412230.post-76699970894963524742019-01-12T23:27:49.454+01:002019-01-12T23:27:49.454+01:00"I have also been asked to replace Yoruba wit..."I have also been asked to replace Yoruba with Senegambian Mandinka samples to check for potential differences. This is something that I had already checked, but I didn't notice any difference in either the Berber percentage in Puerto Ricans or in their homogenization, which indicated a recent founder effect."<br /><br />Thanks for letting me know! I was wondering if perhaps such Senegambian samples would make for a more suitable proxy for Tropical African DNA among North Africans as well as Puerto Ricans with minor SSA. But it doesn't seem to make much difference indeed, going by the adjusted results you showed me.<br /><br />In my survey findings based on AncestryDNA results for Puerto Ricans I actually also tried to account for likely substructure by subdividing my sample group in persons with minor SSA (<25%) and more pronounced SSA (>35%). Aside from the pronounced North African scores I also found it very interesting to see that for Puerto Ricans with a greater degree of SSA lineage other African regions tend to be more important, such as "Nigeria" and "Cameroon/Congo".<br /><br />While for Puerto Ricans with <25% SSA the regional scores for "Senegal" and "Mali" were much more predominant. I have hypothesized that this could indicate a Upper Guinean founding effect. Which would actually also be in line with the findings of Moreno-Estrada et al.(2013)!<br /><br />I first blogged about this in 2015 but I intend to delve into this topic again this year. Also extending my analysis to other Hispanic results, especially for Mexicans and Central Americans, among whom such an Upper Guinean founding effect also seems to be very apparent. See also:<br /><br />https://tracingafricanroots.files.wordpress.com/2015/10/pr35-vs-pr25b.jpg<br /><br />https://tracingafricanroots.wordpress.com/puerto-rican-results/<br /><br />https://tracingafricanroots.wordpress.com/2015/06/09/african-origins-of-hispanic-caribbeans-according-to-dna-studies/Fonte Felipehttps://www.blogger.com/profile/14718819493327107246noreply@blogger.comtag:blogger.com,1999:blog-3023805782808412230.post-75188142036364934252019-01-12T23:15:43.624+01:002019-01-12T23:15:43.624+01:00Intriguing demonstration of how unsupervised ADMIX...Intriguing demonstration of how unsupervised ADMIXTURE runs can contribute to our understanding of the significance of Canarian/Guanche genetic legacy among Puerto Ricans! And quite likely also for other Hispanic Caribbeans.<br /><br /><br />"This finding contrasts with the typically much lower North African scores of Hispanic Caribbeans that are reported in commercial autosomal DNA tests. I suspect that the use of Mozabite samples as proxies for North African may conflate their Berber ancestral component with the Iberian ancestral side of their complex genetic makeup."<br /><br />Indeed, I find it striking how your choice of Moroccan & Sahrawi samples results in a much sharper delineation! Even though given both ancient & more recent genetic connections between the Iberian Peninsula and North Africa such overlap is not surprising in itself.<br /><br /><br />Despite several shortcomings I do find it useful to compare commercial autosomal DNA test results across a wide and relevant selection of various nationalities. Firstly to establish where a certain ancestral component might be most or least prevalent. And secondly to contrast these findings with historical plausibility and/or known genealogy. In order to find out how much correlation might already be found, even if obviously imperfect ;-)<br /><br />I have performed such an exercise for various Afro-descended people, incl. Puerto Ricans as well as native Africans. Based on a survey of over a thousand AncestryDNA results. The focus being on the African breakdown incl. Ancestry's own version of a "Africa North" region. <br /><br />Like you rightfully point out the North African scores on commercial DNA testing platforms such as Ancestry appear to be underestimations. Given also how North Africans themselves rarely scored above 60 % "Africa North" (before the recent update of September 2018). However I still find it fascinating how this pronounced North African connection for Puerto Ricans you found actually also came to light during my survey of AncestryDNA results among the Afro-Diaspora. In my most recent summary of these findings (featuring 8 nationalities and 860 samples) the group average for 110 Puerto Ricans was 2.9% "Africa North" with a maximum individual score of 10%. And this level (again even if quite a likely an underestimation) turned out to be the highest when compared with the other nationalities, safe for Cape Verdeans (n=90) who showed the exact same group average of 2.9% "Africa North".<br /><br />See also this table:<br /><br />https://tracingafricanroots.files.wordpress.com/2018/06/continental-breakdown.png<br /><br /><br />And this link for all the references:<br />https://tracingafricanroots.wordpress.com/2018/06/16/update-afro-diaporan-ancestrydna-survey-part-2/Fonte Felipehttps://www.blogger.com/profile/14718819493327107246noreply@blogger.comtag:blogger.com,1999:blog-3023805782808412230.post-79639189900292508732019-01-12T06:11:37.948+01:002019-01-12T06:11:37.948+01:00I did a blog post some time ago on my fulani match...I did a blog post some time ago on my fulani matches.<br /><br />https://dominicanroots.blogspot.com/2018/05/confirming-african-matches-abuelos.html?fbclid=IwAR1T1WD0lcpeO1ztz15QeRT-T9fCK5pKhVs0XHk0qnkw8AfrGDxmPsDFJdsKirindiambohttps://www.blogger.com/profile/10735919780056272087noreply@blogger.com