View Full Version : Autosomal DNA Testing to find out about your ancestral origins?

17-09-11, 08:13
This thread is a platform for the basic technical aspects of Autosomal testing and its applications within geographic origin tests.

Most interested individuals have a basic understanding of haplogroups and their subclades however some comments of late have made it increasingly clear that there exists a general need for information regarding Autosomal DNA tests used for geographic origin analyzes.


The methods used in biogeographical ancestry projects include the Autosomal STR and Autosomal SNP analyzes. Most use the latter because the former is more work and takes more time.

1. Autosomal STR Testing
2. Autosomal SNP Testing
3. 'Genome-wide' Autosomal Testing [see 2]

1. Autosomal STR Testing
What is tested: The alleles' respective frequency of occurrence and is similar to what we find in conventional Y-DNA STR testing.

Why is this tested: Allele frequencies develop over generations with equal input from the two parents, since for each location one value is taken from the mother and one from the father. The configuration of scores reflects inherited changes from all previous generations in all ancestral lines, and can predict an individual's unique probable ethnic matches based on the profile's frequency or rarity in different populations.

Advantage: Unlike SNP admixture tests, this analysis is based on objectively identified world regions and does not depend on any system of presumed biogeographic classifications.

Disadvantage: Not enough markers are tested to be of any genealogical or real ancestral use. Most STR analyzes examine markers chosen for their high intra-group variation, the utility of these particular STR markers to access inter-group relationships may be greatly diminished. In short, it cannot tell you how much North European you have, it can only say where your STR signature is found on a map.

2. Autosomal SNP Testing
What is tested: The raw data of SNPs are used.

What happens: You go to a company like FTDNA or 23and Me and buy a family finder (FTDNA) or relative finder (23andMe) test. The raw data of your Autosomal DNA is interpreted using ordinary statistical procedure. As more customers from a specific country are typed so the country becomes better defined in terms of those customers. Your profile is compared to other customer profiles and a comparison based on the customer database is projected.

Advantage: Testing your Autosomal DNA now will help develop further the tech and pay for future R & D.

Disadvantages: This test is based on the assumption that people are static beings, who remain fixed in one geographic position. It is based on the arbitrary declaration of customers who claim to have ancestry from this that and the other country. This is a significant weakness because studies base their allocations on what customers declare as their country of origin and most customers buy the test to find out what their origins are. A real case for post facto fallacy exists here as individuals deemed as outliers become relegated and others included to construct ancestral boundaries along popular beliefs currently in fashion. The costs involved make it very difficult to access for most working-class folk.

Claim: These tests claim to identify ancestral contributions to one's genome.

Counterclaim: These tests only test SNPs and do not look at the entire genome or the raw data embedded within Autosomal STRs. Ancestry is estimated with few controls. What specific data points mean is a controversial subject. Autosomal SNPs can not be considered ancestry-informative markers at the present time (yet).

3. 'Genome-wide' Autosomal Testing
What is it: No such test exists it is the same as the 2nd point above. Caution as this is the cheaper and lower resolution form of Autosomal testing available. Most projects use this form and has nothing to do with the finer detail embedded within the raw Autosomal STR data. In reality all genome-wide refers to is the method used to scan the SNPs and it comes directly from the 'genome-wide SNP scan'. This caught on as a marketing tool and has been used to refer to the SNP-based test.

Commercial DNA testing companies sometimes use the phrase genome-wide but this is NOT at all the same thing as mapping your entire genome. Do not be fooled, it's a marketing ploy.


Important: The nature or appearance of human genetic clustering (grouping) is a function of how populations are sampled, of how criteria for boundaries between clusters are set, and of the level of resolution used. All of these factors bias the results — People should be very cautious about relating genetic lineages or clusters to their own sense of identity (Charles Rotimi, of Howard University's National Human Genome Center).

Regarding FTDNA's Family Finder & 23andMe's Relative Finder results
FTDNA have chosen to only report estimated relations up to 5th cousins, here is why. If two people have a match in either of these two products that suggest that two individuals are 5th cousins, but the common link cannot be found, which is more than often the case, the reason is explained as a result of crossover or meiosis. The 'blame' is never placed on the technology or the company but on the ancestors, for passing down each of their parents' DNA in such a random way. The technology doesn't work well and the estimates are wrong. The further beyond 5th cousin one goes back in time the less accurate interpretations and estimates will become as all crossover functions are compounded. The companies know this!

Food for thought: The fact that autosomal results suggest that 25% of the markers tested are for example of European origin does NOT mean that 25% of that person’s ancestors were European. And this is the most confusing aspect of autosomal testing.

17-09-11, 08:25
What are 'Ancestry-Informative Markers'?

This refers to a set of polymorphisms for a single locus and refers to the Autosomal STR analysis. The frequencies of these marker values are used to determine the probable geographic location of other individuals who share or match a particular set of STR values.

Autosomal STR markers are used in much the same way as sex-linked Y-DNA STR markers have been used to match individuals and determine whether individuals are recently related. The more markers are tested the more certain one can be of a match, however as SNP 'ancestry informative markers' do not exist for Autosomal DNA we have nothing to confirm Autosomal STR matches. As a result, more pilot programs are underway to find geographic commonalities, there is still a long way to go before Autosomal SNPs are in anyway shape or form conclusively linked to any specific geographic or ethnic groups.

Mzungu mchagga
17-09-11, 11:27
Oh wow, thanks for all that information!
Frankly, after long consideration I am still undecided for testing. Maybe I should still wait a little and combine it with the purchase of an i-genomeberry (in 3D!).

17-09-11, 11:58
Very thank you Dorianfinder!

17-09-11, 12:54
Comparison between different Autosomal tests available and their respective companies:
1. 23andMe - Half a million Autosomal SNPs tested - Cost: $500
2. FTDNA - Half a million Autosomal SNPs tested - Cost $289
3. DecodeME - One million Autosonal SNPs tested - Cost: $2000
4. Ancestry by DNA - Hand picked Autosomal STR markers tested - Cost: $295
5. DNA Tribes - 15-27 Autosomal STR markers tested - Cost: $150-$310

Only the first two companies provide genealogical information. For genealogy purposes FTDNA are the best value for money. FTDNA's chromosome browser is also the best for comparing matching segments - it combines the best features of both of its main competitors.

If you are interested in "Racial" admixtures, three companies provide this type of product. DecodeMe, 23andMe and 'Ancestry by DNA'. Most people find DecodeMe results confusing and 'Ancestry by DNA' not as scientifically robust. The only alternative is 23andMe's 'ancestry painting'.

Regarding 23andMe's 'ancestry painting' admixture results
What can somebody expect? 23andMe finds most Europeans to be 100% European. For some people this is news and helps to settle certain insecurities they might have.

Some people of predominantly East African ancestry might find that the test says that they have some "European" and some "Asian" ancestry. People with Asiatic Indian ancestry also tend to find that their ancestry painting indicates that they have a higher percentage of "European" than they have "Asian".

These unexpected results most likely reflect the fact that human 'racial' genetic variation doesn't in reality fit into 3 discrete racial categories - instead there are in reality, gradients of variation. It also might reflect the fact that Europeans and Asians were most likely derived from East Africans, so some of the gene variants that are more typical of people of 'Asian' or 'European' ancestry may have originated in their original Eastern African ancestors. Likewise, the large European scores that Asiatic Indians have, might reflect that many of the 'European' gene variants originated in the common ancestors of Europeans and Asiatic Indians, and perhaps a better label would be Eurasian.

Thus there will be some people that do not have admixed ancestries, yet they will appear as a mix of two or all three admixture categories. This isn't because there is something wrong with the algorithm used - this merely reflects that variation between 'racial' groups is variation along continuums, and all that the algorithms do is artificially 'force' the gradual variation into discrete categories.

Sometimes the underlying real biological variation does not correlate with racial identity and/or human social perception.