4 ms·
This was my immediate reaction as well, although I'm not sure I would dismiss the article as "false". Since there seems to be a significant amount of misunderst
by StefanKarpinski 11y ago
This was my immediate reaction as well, although I'm not sure I would dismiss the article as "false". Since there seems to be a significant amount of misunderstanding of what PepeGomez is saying in the rest of this subthread, I'll take a crack at an explanation.
One way to understand is to look backwards at a single individual's ancestors. In an asexually reproducing species, the number of ancestors is O(n) in the number of generations you look back – and in any prior generation the individual has exactly one ancestor. In a sexually reproducing species, the number of ancestors is O(2^n) until you reach the point (surprisingly rapidly) where the whole species either is an ancestor or doesn't have any surviving descendants.
Why does this matter? Because multiple beneficial mutations are far more likely to have arisen in the O(2^n) ancestors in the sexual scenario than in the O(n) ancestors in the asexual scenario. In other words, sexual reproduction allows beneficial mutations to occur independently in any pair of unrelated individuals in a population and eventually both mutations will spread to the whole population – the whole species ends up with all the "winning" mutations. With asexual reproduction, unless two mutations just happen to fall on the same lineage – which is very unlikely – they will compete instead of merging, because there is no mechanism for merging lineages.
Bottom line: in asexual reproduction, independent mutations compete; in sexual reproduction, they can cooperate. Sexual species literally evolve exponentially faster than asexual ones.
- PepeGomez 11y agoExactly. You described it much better than I did.
- jballanc 11y agoExcept that's not how evolution works. You can't just randomly decide to "count backward" and you definitely can't ignore the fact that ancestors overlap. You might look up "The 7 Daughters of Eve". In short: all humans alive today descended from just 7 women, but those 7 women were not the only women alive at the time. All the other women living at the same time as the 7 had their genetic information lost to the sands of time.
- StefanKarpinski 11y agoThe Seven Daughters of Eve is about mitochondrial genetics. Mitochondrial DNA is inherited only from mothers, so the inheritance pattern of mtDNA is exactly like that of an asexually reproducing species. So yes, you can trace the mitochondrial ancestry of each person alive today back to exactly one of the "seven daughters of eve" and the mtDNA of all the other women alive at that time died out. But that doesn't mean that the rest of their genetic material didn't survive. Any woman who had only sons – no matter how many sons she had, or how reproductively successful they were – appears to have died out from the mitochondrial perspective. For that matter, from the mitochondrial view, every man who has ever lived failed to reproduce. Obviously that's not the case when we look at the full picture of all of our genetic material. There's a lot of focus on the mtDNA and Y chromosome lineages because they're linear and therefore relatively static (i.e. evolve at a slow, predictable rate), but it's important to keep in mind that this isn't how most of our genetic material behaves. In particular, you cannot understand sexual genetics by studying mtDNA and Y chromosome inheritance because those lineages are asexual, which is precisely why they are easier to study.