At 10:12 PM 11/24/02 +0100, Eugen Leitl wrote: >On Sun, 24 Nov 2002, Tyler Durden wrote: > >> I believe Daniel Hillis (or was it Jaron Lanier?) inserted time-capsule >> information into a cockroach's DNA and released it into the Boston subways. >> He calculated that this would be the way to preserve information for the >> longest period of time.
Sounds like a gedankenprank that neither are capable of doing without extra training. Especially since they probably haven't tested it by catching, grinding, and sequencing more roaches. >This assumes the insert doesn't result in negative fitness (could very >well be, if the insert kills a gene). > >Also, a fitness-neutral insert is likely to be lost, or severely garbled. >I hope very much he used a really good redundant encoding. Either the message is neutral, and encoded with lots of redundancy (because its going to be changed at the standard 1-in-a-thousand-base mutation rate, and not selected for) or the message is beneficial and is maintained by natural selection. The latter being tough to do, your best hope is an error correcting code. If the message is maladaptive (other than taking up space on the chromo, which for many critters isn't a big hassle) you're fucked.
