tag:blogger.com,1999:blog-30971820.post646429320567479525..comments2023-10-31T21:30:25.737+08:00Comments on Fresh Brainz: Identification Of Design MotifsThe Key Questionhttp://www.blogger.com/profile/05426898630563791849noreply@blogger.comBlogger2125tag:blogger.com,1999:blog-30971820.post-41483653415237268522007-12-12T14:24:00.000+08:002007-12-12T14:24:00.000+08:00The assignment of AAs to a one-letter alphabet is ...<B>The assignment of AAs to a one-letter alphabet is essentially arbitrary. There is no *physical* reason why Glycine ought to be H or X or G. Thus to truly validate your claim, you would need to test all possible mappings to all possible alphabets. (In the case of 20 AAs mapping onto 26 letters you would need to evaluate 26!/(26-20)! = 21543347282404147200000 possible permutations)</B><BR/><BR/>21.5 sextillion permutations? Why, that would just take up a year of supercomputer time!:P<BR/><BR/>The main assumption of IDM is that the putative genome designer exhibits systematic habits. If there are hidden English sentences within the protein sequence, once any comprehensible word of sufficient length (say 10 letters) appears, we immediately have a cipher that can be used to examine the rest of the sequence. No need to reset the permutations for each letter that we move forward.<BR/><BR/>This is broadly similar to what codebreakers did at Bletchley Park. Though unlike this case, they know that the 1. target language is German, and that 2. the message is usually not gibberish.<BR/><BR/><B>And heaven forbid that English isn't the Chosen Language and you had to map AAs to (say) Russian, Chinese or Japanese.</B><BR/><BR/>That's true. However proponents of intelligent design may feel that the genome designer is more likely to use certain languages (for example the 22-character Aramaic), which would support their version of a Designer God.<BR/><BR/>As for the 40,000+ character Chinese...yup, let's examine that language later.<BR/><BR/><B>Even so, the presence of sequences of sensible letter streams can still occur purely by chance, given the large combinatorial space you are working with. Books like the Bible Code have been thoroughly debunked, as they ought to have been.</B><BR/><BR/>Indeed, which is why it will be more persuasive if topographical continuity is demonstrated. It's one thing to fish out a few words here and there; totally something else if a comprehensible sentence appears in sequence. <BR/><BR/><B>Your idea of series of numbers will be very hard to prove one way or another. All sorts of mathematically rigorous sequences have been shown to exist in biology, e.g. the famous Fibonacci sequence. Understanding why such sequences occur is interesting in itself, but does not obviate the need for a creationist explanation.</B><BR/><BR/>I originally had a simple arithmetic sequence in mind (10,20,30...), like those used in BASIC programming, but while riding on the MRT last night I also remembered the Fibonacci sequence. I don't know if this sequence has been found in the genome, but it would be quite bizarre if running numbers were found in non-coding, non-regulatory regions.<BR/><BR/>The first candidate I would suggest to examine using the IDM approach would be the mysteriously massive dystrophin gene. It'll be cool to find out what secrets lie inside its numerous introns.<BR/><BR/>Personally I find that the chances of finding anything interesting this way is very, very slim, since there is significant genome variation even between healthy individuals (SNPs, CNPs and other mutations), demonstrating that genomes don't have to be exactly in a certain configuration in order to work properly, unlike most computer programmes which can be garbled with a bad line of code.<BR/><BR/>But I was trying to demonstrate that a design hypothesis is not ontologically impossible, even if my methodology is shaky. I'm happy to hear of any other possible strategies.The Key Questionhttps://www.blogger.com/profile/05426898630563791849noreply@blogger.comtag:blogger.com,1999:blog-30971820.post-63430838969501721802007-12-12T06:03:00.000+08:002007-12-12T06:03:00.000+08:00I call bullshit on both.1. The assignment of AAs t...I call bullshit on both.<BR/><BR/>1. The assignment of AAs to a one-letter alphabet is essentially arbitrary. There is no *physical* reason why Glycine ought to be H or X or G. Thus to truly validate your claim, you would need to test all possible mappings to all possible alphabets. (In the case of 20 AAs mapping onto 26 letters you would need to evaluate 26!/(26-20)! = 21543347282404147200000 possible permutations) And heaven forbid that English isn't the Chosen Language and you had to map AAs to (say) Russian, Chinese or Japanese. Even so, the presence of sequences of sensible letter streams can still occur purely by chance, given the large combinatorial space you are working with. Books like the Bible Code have been thoroughly debunked, as they ought to have been.<BR/><BR/>2. Your idea of series of numbers will be very hard to prove one way or another. All sorts of mathematically rigorous sequences have been shown to exist in biology, e.g. the famous Fibonacci sequence. Understanding why such sequences occur is interesting in itself, but does not obviate the need for a creationist explanation.Elia Diodatihttps://www.blogger.com/profile/03834465156174945738noreply@blogger.com