How much Neanderthal are you?

I recently had my genome analysed through 23andme and one of the factoids that you get out of this is the estimated percentage of you DNA that is from Neanderthal ancestry.  In my case, 2.9% on the 76th percentile compared to the 2.5% average for European and Chinese. My wife was on the average… I knew she was not as “primitive” as me. But what does this all really mean?


Neanderthals are cousins to modern humans, and they were already living in Europe and the Middle East when modern humans began the migration out of Africa some 100,000 years ago. In 2010, a 60% complete draft genome of a Neanderthal genome was published. As  Neanderthals have been extinct for thousands of years this was a difficult undertaking. The results are based on sequences of DNA salvaged from a number of bones from a cave in Croatia that were dated at between 38,000 and 44,000 years old. Comparing the Neanderthal genome to a number of modern human genomes from people with different ancestry, confirmed that all humans are much more closely related to each other than to Neanderthals, with an estimated split about 800,000 years ago. Nothing too surprising there.

The big news was that 1-4% of the genes in some humans were more closely related to Neanderthals than other humans. The study concluded  that this is a result of interbreeding as modern humans were migrating out of Africa. The average fraction of Neanderthal genes in genomes of African people is only 0.4%, while Asian and European genomes have the 1-4% of genes more closely related to the Neanderthal. They time the interbreeding at some 45,000-80,000 years ago.

Some of the Neanderthal gene variants we have acquired appears even to have helped us adapt to the new environments we were encountering. Neanderthal genes  for keratin, a protein in skin and hair, have been found and these may have given some survival advantages to these people coming out of the warmer climate of Africa. However, A number of Neanderthal genes associated with disease, such as Crohn’s, type 2 diabetes and lupus, were also found.

How does 23andme measure my Neanderthal percent?

The data 23andme uses is not from a genome sequencer, but uses genotyping, which is based on the results of a Illumina OmniExpress Plus gene chip. These technologies select particular SNPs to test for and for the chip 23andme used over 100,000 SNPs selected from the more than 10 million SNPs that makes up the variation between individuals in the human genome. The process of selection means that the data you get is limited to the SNPs on the chip, so you cannot see new variants or SNP locations that you are not looking for. But, this trade off is made to keep costs down and this – amongst other things – means that the genotyping only costs US$100. But the prices are coming down fast and affordable full genome re-sequencing to get SNP profiles may not be too far off.

The limited selection of SNPs and the selection process itself can cause some biases when the comparisons to the Neanderthal genome is being made. The “direct way” to measure Neanderthal DNA is the genome would be to look for the characteristic  Neanderthal SNPs across the regions of the genome that are thought to come from Neanderthals, and based on this estimate how much of these areas of our genome that are the Neanderthal variants. However, there were only 180 SNPs in the 23andme oligotyping and these covered less than 1% of the parts of the genome believed to be of Neanderthal origin, so this was not enough to get a make maeningful estimate of the total amount of Neanderthal DNA.

Instead they use principal components analysis (PCA) of the SNPs from Neanderthals, Denisovans (other humanoid species that diverged from us even before we diverged from the Neanderthals) and Chimp as references and then project the human SNPs onto these components  to estimate the distance from Neanderthal.


I cannot asses the validity of this and they do not attempt to suggest possible margins of error for the results. They do mention “we do not completely trust between-populations comparison (ie. East Asian vs. European)”. This is due to a technicality with the correction for “ascertainment bias” which is caused by the non-random selection of the SNPs which will affect different populations to different degrees (not that this stops them presenting a comparison of my result with the average Asian and African result on the webpage).

So is the 2.9% vs. 2.7% difference between myself and my wife significant? We are both of European ancestry so it probably means I have somewhat more of the Neanderthal variants.

Does it tell me something about how “primitive” I am. Certainly not.

Can it tell really tell me something useful. Maybe…

The Neanderthal percentage is – like so many of the other things on 23andme – a tool that is still in its infancy. A funny number to talk about at parties, but without any real value. In truth, it is trivial if I got the gene from Neanderthals of if I got it down the line from the common ancestor of Neanderthals and humans, what we want to know is how do my SNPs make me different from others that do not have these SNPs. What would be interesting is to specifically name the genes which have the 180 putatively Neanderthal SNPs and suggest how these might have give us particular traits compared to other variants. Perhaps this will be available in future updates to 23andme (or when get around to downloading the data and looking at it myself).

Leave a comment

Filed under Personal Genomics

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s