Friday, January 07, 2011

How Similar Are Humans and Chimpanzees?

When it comes to comparing DNA sequences of individual genes, the human and chimp versions are almost identical in sequence. They differ by only 1-2%. That result gave rise to the oft-quoted similarity of 98-99%.

But that's not the whole story. Outside of the genes there's a large amount of DNA that's less similar. We know this because we now have the sequences of both the human and chimp genomes. Furthermore, there are sequences present in the human genome that are absent in the chimp genome and vice versa. If you look at the whole genomes, the overall similarity is about 95% or so depending on how you do the calculation.

Creationists make a big deal about this. They claim that the newest data proves that evolutionists are wrong and chimps aren't necessarily our cousins. The latest debate is between Fazale Rana on the Reasons to Believe (RTB) website and Dennis Venema on the BioLogos website. The important scientific point is about the actual similarity and how is it calculated?

Fortunately for us, Todd Wood of the Center for Origins Research at Bryan College in Dayton, Tennessee, USA is on the case. Todd belongs to a young-Earth creation study group [BSG] but don't let that fool you. He's doing a pretty good job of sorting out the facts in the case.

RTB and the chimp genome Part 1
RTB and the chimp genome Part 2
RTB and the chimp genome Part 3
RTB and the chimp genome Part 4
RTB and the chimp genome Part 5

Photo Credit:


  1. It's rare to see such honest and careful scholarship and reasoned analysis from a creationist. It's a shame he doesn't apply the same amount of reason and intellectual honesty to the question of creationism itself.

  2. I can never understand how it is that creationists fail to see that using the 'counting' method of Britten et al. increases the differences between ALL taxa, not just humans and chimps - even taxa they accept being related via descent (from some 'kind').

  3. After reading the articles, my cognitive dissonance is exceeded only by Todd Wood's.

    I had a look at the BSG site and they seem to be full out batshit crazy YECers.

    It reminds me of stories I've heard of petroleum geologists who by day locate oil using all the scientific tools that current geology provides but by night bow down to a tribal god that created the earth 6000 years ago.

  4. How about this:
    "Humans and chimps are 80% different"?

    That's true. If one looks at the identity at protein level, 80% of human and chimp proteins differ at one or more amino acid residues.

  5. A good discussion that helps describe the issues related to "similarity" "identity" and "commonality", as those terms are used comparatively.

    Although much easier to read them all by clicking on his home page and just reading the posts up from the bottom. (

    He has another post put up today that continues the thread.

  6. So: humans share 20% of their proteins with chimps. How much with another randomly-chosen human.

  7. Except that the differences between chimps and humans cannot be simply made from a comparison of the their respective genomes.

    This is where both creationists and evolutionists make a fatal mistake.

  8. Those posts are indeed excellent, and helped me prepare tomorrow's genetics lecture.

  9. Tim,

    Note that 80% number in itself is a strong indication that a lot of mutations are under strong purifying selection. Without selection, the ~2% difference in protein-coding DNA would translate into almost all proteins to be different.

    This, coupled with the very small average DNA difference between humans (I don't even know what the number is), implies to me that difference in proteins would be rather small. If I had to guess, ~5%. Of course, there is enough info now to get the real experimental answer. I wonder if anyone bothered?

  10. I think the Hapmap project is reporting the number on SNPs as they go along, and at least in one of their nature paper classed them as synonymous and non-synonymous.

    I can't find the paper or remember the number, though.