Commercial DNA testing VS. Whole Genome Sequencing

[eastern]

Most of us on here have taken a basic, commercial DNA test from companies like MyHeritage, 23andme, AncestryDNA etc. There's also Whole Genome Sequencing, which sequences your entire genome, 100% (as opposed to companies like 23andme, MyHeritage, ancestry etc. which only showcase 0.1% of your entire genome). This means that your results when using calculators will be more consistent and accurate as many calculators will impute alleles that are missing from your Myheritage/23andme/ancestry/etc. raw DNA file, which can sometimes make your results less accurate.

So for the people that HAVE taken a WGS test (the Nebula 30x one is an example), what was different from your commercial DNA results and your WGS results, when it comes to ancestry and heritage? Were there clear differences, or did it align with your previous results?

And for the people that have not taken a WGS test, do you want to take one?

[HannibaltheGreat]

Good question. Subbed for this

[Zoro]

I’m also planning on taking it but based on what i read it’s not useful for some things but very useful for other things, For example you should get similar results at Gedmatch because those calculators use only 23andme or ancestry dna snps about 150k maybe. They don’t use all those millions WGS Snps. Same thing uploading to sites because those sites also use 23andme or ancestry dna Snps.

But if you want like medical information or if you know someone that can compare you to other populations that have WGS Snps using qpadm or dstats then your results should be different using WGS than the commercial companies

I have read different things about it. For example

To date there has been very limited analysis on Asian population histories using whole genome sequencing (WGS). Most population history genetics studies only utilize a few hundred thousand SNPs, which is small fraction of the genetic variation in humans. This small fraction of the genetics variation is ascertained using certain populations. An example are the datasets at the Reich Lab at Harvard or Max Plank which utilize the 1240K Affymetrix Axiom chipset SNPs. Another example is the use of the 2.3M SNP datasets ascertained with the Illumina Omni 2.5 chipset, such as various 1000 Genome Project datasets.

While using a small fraction of the human genetic variation made sense in the past due to the cost of sequencing whole genomes being prohibitively high, nowadays when sequencing a whole genome at an average depth of 30X can be done for under $500, it makes more sense to use whole genomes in population history genetics analysis since analysis results are less biased than using a small fraction of the variation in human DNA.

Genetics analysis based on a small fraction of the SNPs of humans, whether it be 500,000 or 1,000,000 SNPs, ascertained using certain populations introduces all sorts of biases and leads to false conclusions on population histories.

Through a data sharing agreement, the GenomeAsia 100K project1 provided us with WGS VCF files aligned using the GrCh37 Human Reference for 1163 primarily Asian individuals from around 170 populations . We merged this dataset with WGS genomic data sequenced to a depth of 30X from two Kurd individuals from the Kurdistan, Iraq region.

Here we present IBD and IBS analysis results using an intersecting 65 million SNPs across Asian populations.

[eastern]

I’m also planning on taking it but based on what i read it’s not useful for some things but very useful for other things, For example you should get similar results at Gedmatch because those calculators use only 23andme or ancestry dna snps about 150k maybe. They don’t use all those millions WGS Snps. Same thing uploading to sites because those sites also use 23andme or ancestry dna Snps.

But if you want like medical information or if you know someone that can compare you to other populations that have WGS Snps using qpadm or dstats then your results should be different using WGS than the commercial companies

I have read different things about it. For example

Interesting. I've been in contact with Dilawer from EurasianDNA who said that WGS would be useful and more accurate for ancestry calculators (like w eurasian vs e eurasian), but maybe that's only true for some calculators.

[Jingle Bell]

I never taken a WGS test but would be interesting compare to my actual results (Which i use abt 50k SNP)
Just asking, how much SNP g25 uses?

[eastern]

I never taken a WGS test but would be interesting compare to my actual results (Which i use abt 50k SNP)
Just asking, how much SNP g25 uses?

Hold up, 50k snp? Don't the major companies like Myheritage and 23andme use around 600-700k SNPs? What test did you take?

[Beowulf]

I never taken a WGS test but would be interesting compare to my actual results (Which i use abt 50k SNP)
Just asking, how much SNP g25 uses?

only 50k? that's quite bad for something good minumum is 75k wich is my raw data

[eastern]

only 50k? that's quite bad for something good minumum is 75k wich is my raw data

What tests have you guys taken...? 75000 snp? Myheritage, 23andme etc. all use 600-700k.

[Beowulf]

What tests have you guys taken...? 75000 snp? Myheritage, 23andme etc. all use 600-700k.

i have 600.000 snp but when i use k13 or k36 it uses 75.000

[eastern]

i have 600.000 snp but when i use k13 or k36 it uses 75.000

Now I get it. I thought you were saying your original raw dna file was 75k SNP.

Some calculators use 75-80k SNPs. Many also only use 30k, lol. That's no good...