Of course, the context of the selfish gene is evolution, and since evolution implies mixing and competition, the analogy is flawed (or just plain wrong). Specifically, evolution for populations separated by interstellar distances seems unlikely given reasonable transit times and current estimates of the time since the big bang ( 13.7B years last I checked).
Regardless, my bias (and at present it is just that, a bias) is that if we do find artifacts, they will be biological in nature and perhaps even part of the earth biome. Could such "alien incursions" be detected? I have no idea and current biological knowledge about the web of life suggests everything stems from a common ancestral web (as one prominent molecular biologist memorably states, "It's all one show"). However, it is a line of inquiry that might be fun to pursue since it makes for interesting musings about any and all the things we can't explain about terrestrial biological history. I like to think of it as Dawkins' Selfish Gene writ very large.
But the fundamental question of how information about biological incursions might be detected -- a communications problem at heart -- has not yet been formulated. It could be that such information is rapidly corrupted on earth through various natural processes (of which evolution in its various forms is one) and this might make detection from biological records virtually impossible. Or maybe it's easier than we might expect as suggested in an interesting essay ( New Scientist, August 7, 2004) by Paul Davies on searching our own genetic backyard for evidence of ET.
Of course, there is also the possibility of literal physical artifacts as well since explorative civilizations will almost certainly launch various sorts of probes to better understand the cosmos. And taking that thought just a bit farther, one could also imagine "stowaway" biological material aboard such probes (though survival in the harsh interstellar environment might seem to preclude such inadvertent "panspermia").
It is interesting to note that various forms of all these ideas have been around for a while both in the technical literature and in the science fiction literature, but I have yet to find quantitative treatments with a minimum of assumptions that point to one form of contact or another. So all the above (once more for emphasis ) comprise musings rather than any attempt at real science.