tag:blogger.com,1999:blog-37148773.post6763501645391022519..comments2024-03-27T14:50:47.345-04:00Comments on <center>Sandwalk</center>: New Ways of Looking at EvolutionLarry Moranhttp://www.blogger.com/profile/05756598746605455848noreply@blogger.comBlogger4125tag:blogger.com,1999:blog-37148773.post-69710572506571370352007-04-20T00:43:00.000-04:002007-04-20T00:43:00.000-04:00Many thanks. Let me respond to just one snippet (...Many thanks. Let me respond to just one snippet (though much else is worth comment, of course). You wrote<BR/><BR/><I>Fair enough, but what other possible pathways could have been followed? If the actual pathway is only one of several million possibilities then why was that one particular design selected? Is there no possibility that it could have been accidental and selected? Isn't it possible that the actual end-product is as much due to the random mutations that occurred as it is due to natural selection?<BR/><BR/>Dawkins seems to ignore this possibility when he draws attention to the non-randomness of evolution. I think that's a mistake. We can easily agree that natural selection is non-random but it doesn't necessarily follow that evolution now becomes predictable.</I><BR/><BR/>In a <A HREF="http://www.iidb.org/vbb/showthread.php?t=197955" REL="nofollow">train wreck of a thread</A> on Infidels I was inspired recently to re-examine some of the data from Lenski, et al.'s 2003 <A HREF="http://myxo.css.msu.edu/papers/nature2003/" REL="nofollow">Evolutionary Origin of Complex Features</A>. While it's not part of that re-analysis, there's one striking result of that study that both illustrates your point above and modifies it. In a static selective environment identical across runs, with an identical Ancestor over runs, and varying <I>only</I> the initial random seed that determined the timing, kind and location of mutations through the runs, in 50 runs Lenski's Avida simulation evolved artificial agents that could perform a fairly complicated logic function (EQUALS) in 23 of 50 runs. What is of interest in this context is that the 23 lineages that evolved to perform EQUALS did so in 23 <I>different</I> ways -- the 23 evolutionary pathways were different from one another and the underlying opcode programs that were the artificial agents' 'genes' were different from one another in all 23 cases. They performed the same function, but they evolved that capability by very different routes, and in the end did it in very different ways.<BR/><BR/>So in one sense the system is predictable: under conditions where there are complicated functions that are reproductively advantageous, 'Darwinian' mechanisms can often assemble complicated critters that perform those functions. On the other hand, it's an unpredictable system in that the particular evolutionary route (history) and the particular structure (genome) that enables performance of the functions are wholly dependent on the initial random seed: one can't predict the pathway, one has to run the model to see what it does given the particular randomization.<BR/><BR/>That provides a nice illustration of your argument that <BR/><BR/><I>The main conclusion of this essay is that a large part of ongoing evolution is determined by stochastic events that might as well be called "chance" or "random." Furthermore, a good deal of the past history of life on Earth was the product of chance events, or accidents, that could not have been predicted. </I><BR/><BR/>I will not invoke Conway Morris' catalog of convergent structures here, but I'm put in mind of it. :)<BR/><BR/>RBHRBHhttps://www.blogger.com/profile/13562135000111792590noreply@blogger.comtag:blogger.com,1999:blog-37148773.post-82399665552351436962007-04-19T20:41:00.000-04:002007-04-19T20:41:00.000-04:00You can start here : Evolution by AccidentYou can start here : <A HREF="http://bioinfo.med.utoronto.ca/Evolution_by_Accident/Evolution_by_Accident.html" REL="nofollow">Evolution by Accident</A>Larry Moranhttps://www.blogger.com/profile/05756598746605455848noreply@blogger.comtag:blogger.com,1999:blog-37148773.post-73866113239319153262007-04-19T18:29:00.000-04:002007-04-19T18:29:00.000-04:00One of the "new ways" of looking at evolution is t...<I>One of the "new ways" of looking at evolution is to consider mutation pressure, loosely defined as differences in the frequency of mutation. I'm not a big fan of this but it does emphasize that modern evolutionary theorists are thinking outside the Darwinian box—not surprising since Darwin died 125 years ago (today). I prefer mutationism, which is a way of emphasizing the imprtant role of mutation in directing evolution. Mutationism and mutation pressure are not the same thing.</I><BR/><BR/>I'd sure like to see more on this from you, as and when you have time. Or maybe some references?<BR/><BR/>RBHAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-37148773.post-58647847269918561942007-04-19T17:45:00.000-04:002007-04-19T17:45:00.000-04:00Lynch's model isn't selection free. Rather, he cho...Lynch's model isn't selection free. Rather, he choses to emphasize relaxed selective constraint due to small population size in taxa with complex gene structure. It's Ohta's nearly neutral theory extended to the evolution of genomes.RPMhttps://www.blogger.com/profile/00344508931818143159noreply@blogger.com