I need Matt: (Introduction)

by xeno6696 , Sonoran Desert, Thursday, September 02, 2010, 12:16 (5196 days ago) @ David Turell

David, the question bothered me overnight, so I got up extra early to tackle it. -> I am thrilled that we have a qualified mathematician with us. Matt look at this argument on protein folding and the conclusion that the specificity at the tertiary level requires a process other than Darwinian chance. Is this mathematically reasonable?
> 
> http://bio-complexity.org/ojs/index.php/main/article/view/BIO-C.2010.1-At base, I checked all the calculations he provided, and yes--they are sound. However he has one critical flaw here (from a mathematical perspective.) He claims that the evolutionary "search algorithm" wouldn't be capable of the success needed. That's fine, except that he provides no actual mathematical function beyond that which I described last night: he declares an f(x), but provides no machinery for it to work. Nor does he provide an alternate g(x) search function for comparison. He throws out plenty of math, and some good discussion of biochem, but no actual comparison ever ensues; it's not a mathematical argument, only an argument that uses some math.-Why it is important that he defines functions explicitly is this:-compute a sequence f(x) = x^2, and g(x) = x*log(x). Plot the results. (Don't worry I'll show you. (These also provide examples of what an analysis of algorithms SHOULD look like!) -f(x) = http://webspace.ship.edu/cawell/Sorting/bubanal.htm
g(x) = http://www.iti.fh-flensburg.de/lang/algorithmen/sortieren/quick/quicken.htm-http://www.... For a view of a very formal (and complete) algorithmic analysis. -If he doesn't provide input sets with functions, with computed output sets... then there's no comparison, and more importantly, no time-complexity analysis. -His analysis implicitly asserts that evolution is an exhaustive search function, that his function is a linear search that must search through all elements. He asserts f(x) = x. The time complexity of this would have a theoretical upper bound of x, but in practice linear searches rarely have to search through the whole set. And--some searches will have issues depending on the algorithm--the quicksort I showed you above, has an average time case and a worst time case. And I haven't even started talking about how his data sets are modeled... data structures themselves can have an immense impact on search time. -So, I leave you with this: it's a convincing argument until you ask yourself what exact search functions is he comparing? His argument is akin to showing you bubble sort as the "Darwinian" search algorithm, and suggests that something better is out there... without telling you what it is.

--
\"Why is it, Master, that ascetics fight with ascetics?\"

\"It is, brahmin, because of attachment to views, adherence to views, fixation on views, addiction to views, obsession with views, holding firmly to views that ascetics fight with ascetics.\"