Here is a likely poorly-specified question for biologists, prompted by wanting to buy Alan Weisman’s The World Without Us and then reading a story about genetically modified mice. Weisman’s book asks how the world would change and what of us would survive if humans were all wiped out overnight or just disappeared by something (a virus, the Rapture). The premise is unlikely (something that kills people—all people—but leaves the rest of the world standing) but intriguing.
So I wondered, what if, long, long after our disappearance, some other species arose on earth at least as intelligent as us and eventually started doing evolutionary and molecular biology. Let’s say they have a working theory of evolution much like our own. Now say for the sake of argument that a bunch of transgenic organisms produced by humans have survived and prospered in the interim. So our future biologists find things like a bacteria that produces insulin, or a plant that secretes insecticide, or rice that is high in beta carotene, or more exotic stuff as needed.
I’m wondering, would such organisms even present themselves as empirical anomalies? (That is, how much would you have to know about genomes and evolution for them to seem odd?) And if they did seem odd, how would they be explained? That is, would the evidence of their intelligent design by a previous, now-extinct species be clear? You can see that I’m just irony-mongering here. Would some Arthropod-staffed functional-equivalent of the Discovery Institute point its claw at some of these organisms, saying they were anomalies that could only be explained by the intervention of a divine intelligence? Would Charles Crustacean find a story that could account for their evolution by natural selection? I’m particularly interested in whether the artificial provenance of transgenic organisms would be clear on internal evidence alone. I don’t know anything about this stuff, so probably the answer is “Yes” for reasons obvious to experts. But if it weren’t …
From the sound of Weisman’s book, though, internal evidence wouldn’t be all that was available. Our putative Arthropod successors would likely be able to conjecuture as follows: “The lost civilization who did this is probably the same one responsible for leaving those giant goddamn piles of steel-belted rubber rings and miscellaneous plastic items piled around the place.” To which someone would no doubt reply, “Come off it, no organism that spent its time making rubber tubes and piling them up in giant mountains would have ever been smart enough to figure out genetic engineering.”
 It occurs to me that rice requires a lot of cultivation to prosper, but there aren’t any humans to take care of it. Hence, “insert example as needed.”
I think it very likely that genetic modifications would survive and be detectable. I might mention that I have a BA and an MA in biology although I am an economist.
One reason to think this is that there are known exceptions to a very simple random mutation plus selection version of the neodarwinian synthesis. Many organisms contain what appear to be retrovirus genomes (or something similar). They do not help the host but do to little harm to be eliminated by selection. Members of the same species often have very different numbers of these things. It has been concluded that they have made copies of themselves, that is, the DNA is, in a sense, a product of a parasite organism not the normal mutation of the hosts DNA. These little things are not intelligent, but they were once anomalous and are a new addition to the standard model -- and they were detected.
One important issue is that much of our genome (and much more of Xenopus' and much much much more of lillies) is apparently selectively neutral. There is a lot of DNA (as a fraction of the genome) that doesn't appear to do anything useful. It imposes almost no cost as the amount of DNA in grams in an organism is tiny.
A bacterium which contains the coding region for insulin, but does not produce insulin (do to a mutation in the promoter) would not be at an evolutionarily relevant disadvantage compared to wild type. It would take a very long time for random drift to cover up the strange presence of a Eukaryotic gene in a bacterium. It might not be clear that it got there via intelligent manipulators, but it would an anomaly.
I'd guess that b Thuringiensis toxin (insecticide) producing plants would have an advantage (and didn't exist before humans as the relevant mutation didn't occur). The appearance of the same gene in totally unrelated organisms and in nothing in between would be very very strange. the arthropods might even notice that the plants are good mammal food.
Golden rice is clearly manipolated (and this would be clear even if the rice benefited). Many genes not found in other grains are all in the same spot in golden rice (I think). this is a result of popping a whole metabolic pathway in with one fell swoop. This is not a normal pattern for eukaryotes (roughly things other than bacteria) and is anomalous.
I'd say we are already well past the point that our interventions in genomes will be undetectable for tens of millions of years.