Hacker News Reader

Comment by s1artibartfast | Hacker News Reader

One thing that I always struggle with in these arciles is how the percent of common DNA is defined.

>We still think that most – anywhere between about 92% and 98.5% – of the ancestry in people not living in Africa today does indeed derive from the out-of-Africa expansion.

>Interestingly, they suggest that 6%-7% of the genomes of West Africans is archaic in origin. But this archaic ancestry wasn’t Neanderthal or Denisovan.

How are these numbers reconciled with other statements like 99% of the human genome is shared with Bonobos [1]

https://www.sciencemag.org/news/2012/06/bonobos-join-chimps-...

View on HN

Re: definitions of shared DNA:

Some papers will use different definitions, and it can be confusing.

Between species (e.g. humans and chimps), it's most often considering "non-synonymous mutations," i.e. how many genetic differences there are in protein-coding genes (these are relevant to structure and function of the protein)

For distant relatives (e.g. humans and bananas), it's similar, looking specially at those genes which we can identify across many species (particularly those related to cell upkeep, DNA replication, structure, etc.).

For within-species, there are a few ways of doing it; you could model how much of the genome you expect came from each source (as this study did) (see admixture analysis, coalescence), and you could look for overall differences on the genome (of all existing variation in the species, how much is consistently different between populations) (see F-statistics for example)

Thank you. This is very helpful!

> How are these numbers reconciled with other statements like 99% of the human genome is shared with Bonobos [1]

99% of any hominid genome - whether Neanderthal, Denisovan, or the "archaic" ancestor refererenced - is also shared with bonobos. The branching point with bonobos is at a much earlier stage of evolution, so all descendant branches have roughly the same affinity to them.

My point exactly.

How can a homo sapien in Europe share 99% of their genome with a Bonobo, but only 92% with another homo sapien in Africa

99% is comparing the entire genome, introns and exons. 92% is comparing known genes gathered from the human genome project and subsequent research. We don't know the exact genes expressed by our ancestors since RNA is far less likely to survive in enough different types of tissue to get a clear picture of the entire genome but 92% of the genes we recognize in ourselves can be found in early hominids using the latter's full genome sequence. Between epigenetics and embryology, we know that the whole intron/exon dichotomy is fatally flawed and evolutionary genetics has long turned to more complex methods to trace relationships, which are very nonlinear even in our near family tree.

Both numbers are ridiculous and meaningless so don't pay them any mind.

> Both numbers are ridiculous and meaningless so don't pay them any mind

I agree that they have little meaning for the purpose of individual and group identity. It's been pretty disheartening, for example, to see some people extrapolate from the discovery of Neanderthal DNA in non-African populations to the current economic disparities between non-African and African societies. Ironicallly, that is a reversal of the previous false stereotypes associated with Neanderthal influence in humans. I guess once it was proven, it had to be turned into a "good" thing.

But the percentages are meaningful for building a picture of ancient human evolution and migration.

Reply on news.ycombinator.com