If Humans Are 99% Genetically Similar, How Do Paternity Tests Work?

And how racists may misrepresent these genetic comparisons

“Are you sure this is my child? I can talk, but he can’t talk at all.” Photo by Bethany Beck on Unsplash

In genetics, we sometimes talk about how so much of our DNA sequence is conserved; we share 96% of our DNA with chimpanzees and bonobos.

But we also talk about shared DNA when it comes to family relatedness; we share 50% of our DNA with siblings, 12.5% with first cousins, and about 0.1% with a random stranger.

What’s going on? Reading these two statements, it sounds like I’ve got more in common with a chimpanzee than with my sister! I do like bananas a lot more than her, but this is what happens when two different kinds of genetic comparisons get accidentally mixed together.

This sort of confusion is where misinformation thrives. It’s also sometimes intentionally exploited, like when certain people (cough, racists, cough) use it to claim that there are different subgroups of humans that are more different from each other than we are to apes.

DNA is like a giant cookbook, with very few recipes

The confusion starts with the fact that there’s different ways to compare DNA.

Let’s start with defining what we’re talking about, here.

DNA is a very long molecule, made of four building blocks — adenine, thymine, cytosine, and guanine. We usually refer to these by their first initial, A, T, C, and G. They’re called bases.

The full human genome — our entire DNA sequence — consists of ~3 billion bases. (And, if that wasn’t enough, most of the cells in our body maintain two working copies of the genome, for a total of 6 billion bases per cell!)

3 billion bases of A, C, T, and G molecules, chained together… but not all them are as important as others.

Certain stretches of our DNA encode the instructions for building the proteins that make up our bodies and let us survive; these areas of the genome are called genes. But only about 2% of our genome is genes.

What about the other 98%? It’s not junk; it serves other functions:

• Repetitive caps on the ends of our chromosomes, called telomeres, help keep the DNA from wearing away and shortening over time.
• Repetitive stretches in the middle of each chromosome, called centromeres, help the two copies stay near each other.
• Certain regions help turn nearby genes on or off, or boost or dampen their signal. These are called promoters and operators.
• And yes, some of it is just junk sequences. Evolution isn’t worried about efficiency; it just cares what works.

Think of your genome as a cookbook. It’s got 400 pages in it, but:

• The first and last 5 pages are just scratch paper, padding for the rest of the book.
• The middle ~20 pages are also scratch paper, slightly larger than the rest to help the books stack neatly.
• Even though the book is 400 pages, there are only 8 pages of actual recipes.
• Another 20 pages tell you when you should or shouldn’t make those recipes (serve for guests, but not after Labor Day!).
• The rest of the cookbook is mostly gibberish; there are some partial recipes in there that are just crossed out, since they no longer work.

It’s not the best cookbook in the world in terms of efficiency, but it’s good enough to get the job done in providing recipes, and that’s good enough for us!

Now we know what DNA is. How do we compare it?

Most of the genome is pretty conserved

When we talk about the amount of shared DNA between different species, we’re talking about all the DNA. Every page in that cookbook, including all the gibberish and the scratch pages.

When we do this level of comparison, it turns out that most cookbooks are pretty similar. Humans and chimpanzees both need telomere caps at the ends of their chromosomes. They both need centromeres to help their chromosomes pair up. And a lot of the junk sequences haven’t changed much over millions of years of evolution, because there’s no fitness advantage to changes in those areas.

Thus, 96% similar DNA (when looking at everything in the cookbook) between humans and chimps.

Even if we just focus on the actual 8 pages of recipes, there’s still a lot of similarity. Both humans and chimps have two arms, two legs, lungs, a stomach, hair and nails made of keratin, eyes, a brain, blood… the similarities go on and on.

In fact, when just looking at the genes themselves, we’re even more similar — 99.1% similar when looking at the genes that make proteins (the recipes).

If you look at the cookbooks of any two random people, they’re going to be about 99.9% similar.

When we start talking about differences between family members, or between different groups of people, we’re talking about everything in that final 0.1%.

From 3 billion to 3 million

Across all of humanity, there’s about 3 million bases that vary. This is where comparisons of different populations comes in.

When you get an ancestry test from a consumer sequencing company, they look at even fewer bases. AncestryDNA’s ancestry testing panel looks at 116,800 spots across the genome.

When we talk about being “50% similar” to a sibling, we’re talking about just that little 3 million base slice of the total genome.

“So does this mean that two siblings will differ by 1.5 million bases?”

Possibly even fewer than that. If the two parents have any of those 3 million bases in common, then the children will also be more similar.

Example: your parents are both from western Europe. They aren’t related to each other, but since they’re both from the same broad geographic region, they have an additional 10–15% similarity. This means your parents have about 450,000 additional base pairs in common with each other, and you and your sibling will differ by about 1.3 million bases.

But this is where some people, especially those who tend to espouse views such as “racial purity,” will protest.

“You’ve got more of your DNA in common with others from your own race!” they say. “Having children with someone from a different race will make them less similar to you!”

But there’s two big issues with this sort of assumption.

Skin color says very little about genetics

Take two people. One has dark skin, the other has light skin. How genetically different are they?

The answer: it’s basically impossible to tell; there’s not enough information. Of those up-to-3 million DNA bases that distinguish any human from another, 10–15% of them are linked with geographical ancestry. The other 85–90% of them are variations *across all populations.*

Ethnicities are not genetically isolated groups. They are sometimes genetically distinguishable (you can guess from a DNA test), but if you were to group people into ethnicities, each would contain basically the same broad genetic variation found in the total human population.

If you took two random strangers — one from the next town over from you, the other from halfway across the globe — and compared their DNA to yours, you may find that the person from the other side of the world has more in common with you, genetically.

It’s a bit like comparing wheeled vehicles. Some vehicles have four wheels. Some do not. But does this mean that any two vehicles with four wheels will be more similar to each other than to any vehicle with a different wheel count?

(My pickup truck has 4 wheels. My child’s training bike has 4 wheels. A cement mixer has 6 to 12 wheels. Is my pickup truck noticeably more similar to a training bike than to a cement mixer? Or is “number of wheels” not a good metric for comparing all wheeled vehicles?)

When we talk about DNA in common, it’s a different comparison between people versus between species. In the comparison of the entire genome — all the DNA — any two people will have over 99% of their DNA in common.

When we compare two people, however, we’re focusing only on the 0.1% of genome-wide variation that’s within our species. When we talk about two related people having 50% shared ancestry, we’re talking about similarities within that last 0.1% that varies human-to-human.

So the next time someone posts, “how can I be 60% similar to a banana when I’m only 50% similar to my own sibling,” you can now correct them — they’re talking about two different types of comparison!