[gravelc]

Sceptical of this. Where did that number come from? What's the uncertainty? Yes, it's an extremely unlikely event, but unlikely events happen all the time. There's just as good an argument that given the advantages conferred by the union, it's bound to be successful given enough iterations.

[vikramkr]

I mean, it happened again with chloroplasts and there's a bunch of other cases of endosymbiosis too it's not that rare

[bonzini]

Making it happen once is harder than doing it a second time. Plant cells have both mitochondria and chloroplasts.

[vikramkr]

I'm unfamiliar with research into the difficulty of the second endosymbiotic event that no humans were present to measure and analyze being reduced by the existence of a first endosymbiotic event that has also not been replicated in a lab yet, resulting in another n=1 data point

[favoritecolor]

I think this might count as endosymbiosis replicated in a lab: https://pubmed.ncbi.nlm.nih.gov/30373839/

[vikramkr]

It's a system to measure endosymbiosis in an organism that's already undergone endosymbiosis so not directly applicable for the entirety of gp's hypothesis, but there's definitely a ton of research into it. There'll always remain a question of if we engineer a system does that work the same way as the evolutionary path since we have the benefit of being non-random/guided by intelligence. It's like with abiogenesis, if we ever figure out a system that lets us reproducibly create life within the timespan of a couple years so we can replicate it and study it reasonably that system almost certainly isn't the original way it happened, but it'll give us good insights into the concepts hopefully

[bonzini]

In order to protect its genome from the mitochondrial genome, the host cell had to evolve the nucleus and possibly also sexual reproduction. Once that's done, it works just as well for the chloroplast.

[vikramkr]

The nucleus is also considered a potential candidate for endosymbiosis [1] and I'd really love to see some actual evidence for your claims at some point instead of just vague handwaving

(1) https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4571569

[kridsdale1]

I appreciate you for calling out the number. It’s ok to say “vanishingly minuscule”. A number may seem insignificant but what was presented was actually precise.