[jefftk]

> that sort of thing must happen frequently in nature

Individual chiral molecules can happen naturally, sure, but not entire organisms. From the report: In a mirror bacterium, all of the chiral molecules of existing bacteria—proteins, nucleic acids, and metabolites—are replaced by their mirror images.

In the fitness landscape there is an absolutely enormous gulf between standard and mirror bacteria, large enough that no amount of incremental evolutionary pressure could flip the ~billion chiral bonds in a given bacterium simultaneously.

[roenxi]

Nature has built a lot of bacteria over the millennia. It probably has experimented with just building the whole thing backwards.

That sounds simpler to do than, say, evolving a cogwheel (which does happen occasionally) or actually developing some of these molecules in the first place. It'd be weird for evolution to struggle so much on trying such a simple concept somewhere. This is a process that naturally figured out solar power, a bunch of mechanical engineering properties, various chemical techniques, all sort of fluid dynamics and statics, radiation resistance, sensing and control systems, etc, etc. If it doesn't build things backwards, more likely than not it is because it doesn't work well. The process knows how to do quite complex engineering tasks.

[jefftk]

> It'd be weird for evolution to struggle so much on trying such a simple concept somewhere.

While "building it backwards" sounds simple to us, it's not simple at all in evolutionary space. Evolution operates over a fitness landscape, where every configuration of an organism has some fitness (essentially, probability of reproducing). It can make good progress as long as there are paths that don't require traversing areas of impractically low fitness. The problem in this case is that between "normal" and "mirror" is an area of absurdly low fitness, which evolution shouldn't be expected to be able to cross.

[roenxi]

We can't really say what is hard, because evolution tends to be more creative than anyone expects. But we can say it sounds easy - evolution would only need to come up with a "build this with reverse the chirality" mechanism and run everything through it. That'd even be what humans are doing in this research, I suspect. If there are advantages to reversing chirality then there'd be lots of paths where partially reversing chirality of parts of an organism would be interesting or creating clouds of reverse-chirality molecules would be evolutionary powerful.

It wouldn't be likely, but we're talking millions of years and a process that has overcome some remarkable engineering challenges. Finding one place where building backwards then extending that just wouldn't be such a big deal.