[nwatson]

Even if the organisms were shown to swap out phosphorus for arsenic in some instances in their DNA, would that be surprising? Arsenic comes from the same column in the periodic table as phosphorus, and exhibits many of the same macro chemical properties as phosphorus.

It would be like a software engineer swapping out their optimized, say, C++ string library, for one that was less optimal but perhaps took less space or ran better in a constrained embedded environment. Arsenic might not be optimal but it's what's available, what can be used.

In the software case you'd probably be OK but run more slowly than otherwise, or else maybe you'd stumble when for some strange reason the sub-optimal string library didn't guarantee thread-safe read operations. The arsenic could similarly trip up the organism in some cases.

[synparb]

I think your analogy is not quite right. Although seemingly similar, inserting Arsenic into DNA would be like taking a very large and complicated piece of software that was fundamentally dependent on the precise API of a component, and then changing that API to make it largely incompatible, thus requiring a large piece of your software to be re-written. It is not just like swapping in a suboptimal component that has no downstream effects.