RSS 2.0 Feed

» Welcome Guest Log In :: Register

Pages: (12) < 1 2 3 [4] 5 6 7 8 9 ... >   
  Topic: Thread for Cryptoguru, Evolution, Evolutionary Computing, etc< Next Oldest | Next Newest >  
QED



Posts: 41
Joined: July 2008

(Permalink) Posted: Jan. 18 2015,21:25   

Told ya'...fly pigeon, fly!

  
cryptoguru



Posts: 53
Joined: Jan. 2015

(Permalink) Posted: Jan. 19 2015,07:45   

wow ... OK, to avoid being labelled as running from a fight ... I'm going to try once more to see where we can get with a more structured approach
(I'm an amateur MMA fighter so not used to that sort of accusation)

I'm not expecting agreement, but at least it should help clarify where the sticking points are. If discussing a point, instead of posting an essay and saying it covers EVERYTHING, please refer to section and subsection so we all know which point you're contesting. Then we can get to a list of agreed principles, which should allow us to concentrate on the important controversial bits. (What was happening before was a full-scale assault on anything I said or the way I said it irrespective of whether there was agreement on the underlying principles or not.) Please also mark which points you agree on, otherwise I don't know where the common ground is.

Section A: How does the cell work (very very basic version)
1) living cells process a program called DNA
2) DNA is a code (a sequence of 3 symbols codes for amino acid)
3) DNA is processed in a non-linear fashion through regulatory mechanisms with multiple reading frames.
4) The cell eventually produces folded proteins, which are determined by which genes are being read, the reading frame, the regulatory process ... tonnes of stuff ... but it's deterministic (ignoring mutations) and the whole process is defined by the code. (this includes regulation in response to external stimuli ... i.e. input)
5) folded proteins are assembled into complex structures, again this process is controlled by mechanisms which are defined in the code.
6) the cells replicate and differentiate their function in a continuous morphological (yet continually compositionally functional) fashion based on their position in the organism, this process is also controlled by the code.

Section B:
What is Biological Evolution
1) given a population of self-replicating entities (a living cell)
2) random mutation of symbols in the raw code (usually point mutations ... about 50-100 de novo germline mutations per generation in humans)
3) advantages are introduced via the mutations by chance, the advantages are preserved and propagate through the population by "natural selection"
4) Natural selection is the name given to the process whereby environmental conditions apply a pressure for survival on the entities.
5) Entities therefore compete for resources and the ultimately surviving hereditary line is preserved, presumably carrying the genetic advantage that allowed it to overcome its competition.
6) Differential reproduction is a fancy way of saying that the hereditary line of some organisms, due to their inferiority in how they are suited to the environment, eventually (at some level of descent) produce no offspring before dying and therefore leave the competition.
7) Death is the only agent in natural selection, regardless of whether some organisms are only slightly better adapted than others, there will eventually be a point where an entire inferior hereditary line is removed (ultimately by death) from the population (i.e. a discrete process, not a continuous one). It is now as though that line had never existed. This is why death is the only important selector, comparisons between organisms do not come into the selection process ... it is simply a case of whether an organism is sufficiently fit to survive long enough in the environment to reproduce.

(I imagine point 7 will draw the most controversy ... TBH it's not important that you agree on that point for the sake of my argument, so happy to leave Natural Selection as an ambiguous and mysterious force if people would prefer so we can move on to where the actual disagreement should occur)

Section C:
What does AVIDA model
(I'm going to change my analogy here to focus the disagreement away from mutation)
1) 26 logical commands are provided and randomly selected and combined by "mutating" an original arrangement of them.
2) the commands are a list of functional operators (building blocks) that a computer code can be built from. They are Turing complete, so we should expect to be able to build any computer code from them in the correct combination
3) self-replicating digital entities, which have their own virtual processor run and mutate these programs
4) any entities are rewarded when a sequence of commands forms a logical function by increasing their resources

Section D:
comparison of AVIDA to biological evolution
1) an analogy must be drawn between AVIDA commands and the genome
2) it possibly makes most sense (I concede) to assume that AVIDA commands are analogous to codons (and not proteins), so that any mutation will always create a set of valid codons.
3) the level AVIDA is selecting at is therefore analogous to a folded functional protein.

Section E:
problems with model
1) the multi-dimensionality of the genome due to multiple reading frames means that a point mutation in the genome will likely affect the expression of multiple proteins (19K coding genes code for at least 100K different proteins in the human genome). Mutations in AVIDA are mutually exclusive and therefore don't have a regressive effect on the expression of other COMMANDS. This is not a trivial difference, it is analogous to the difference between a bisection method (AVIDA mutation) and a bisection method where the root can change at each iteration (Genome mutation).
2) It is not just multiple reading frames that introduce polymorphism into the genome, but regulatory genes can effect the expression of an entire coding gene. This non-linearity is not modelled in AVIDA, which is a linear sequential code (like assembler). That is, the Genome executes a higher-level language than a sequential instruction set.
3) AVIDA enforces selection by rewarding at the functional level ... it identifies a function as a logical operation and rewards the organism that is presenting it. This is equivalent to natural selection providing feedback scores to the organism on a per-protein level. e.g. protein 1 7/10, protein 2 3/10, protein 3 9/10. This kind of micro-management can't happen in real-life as natural selection is blind and is applied not even to an organism but an entire population. This feature artificially boosts the productivity of hopeful combinations of commands, which otherwise wouldn't be encouraged.
4) The level of variability in AVIDA compared to the genome is like comparing solving a Rubik's cube with cracking 2048-bit RSA encryption. The logical functions which AVIDA selects and guide the optimisation process are trivial ... and also non-distinct. There are infinitely many ways to implement the EQU function using the AVIDA instruction set. Proteins are specific in their form, not just in an abstract functional way. The likelihood of randomly selecting a combination of AVIDA commands that performs a logical function is extremely high, I don't know if it's even possible to work this out considering there are infinitely many combinations that could represent each logical function. (e.g. inc dec inc dec inc dec is equivalent to leaving a register unchanged). In the genome the possibility of just mutating 3 neighbouring nucleotides anywhere in the genome to produce a different codon is less than the chances of winning the national lottery twice in a row, and two proteins with different amino acid chains can likely never be equivalent in function. (unless randomness of function is the required function for the protein)

DISCUSS!

  
JonF



Posts: 634
Joined: Feb. 2005

(Permalink) Posted: Jan. 19 2015,08:16   

You need to supply some serious support for E.4. Particularly, two proteins with different amino acid chains are often equivalent in function. Nobody has any idea how many proteins there are that perform some specified function, but the evidence suggest there's loads of 'em. Plus there are proteins that perform multiple functions.

  
Zachriel



Posts: 2723
Joined: Sep. 2006

(Permalink) Posted: Jan. 19 2015,08:21   

cryptoguru: evolution mutates DNA on a nucleotide level affecting function (gene level) and selects on the organism level.

Evolution also involves recombination and the splicing of new structures from subcomponents.  

cryptoguru: The whole premise of natural selection can be simplified to be death.

That is also false. The whole premise of natural selection is reproductive success.



--------------

You never step on the same tard twice—for it's not the same tard and you're not the same person.

   
rossum



Posts: 289
Joined: Dec. 2008

(Permalink) Posted: Jan. 19 2015,08:33   

Quote (JonF @ Jan. 19 2015,08:16)
You need to supply some serious support for E.4. Particularly, two proteins with different amino acid chains are often equivalent in function. Nobody has any idea how many proteins there are that perform some specified function, but the evidence suggest there's loads of 'em. Plus there are proteins that perform multiple functions.

The calculation has been done for Cytochrome C: Yockey (1992) calculated that there are  2.3 x 10^93 different amino acid sequences that could make a working Cytochrome C.

--------------
The ultimate truth is that there is no ultimate truth.

  
cryptoguru



Posts: 53
Joined: Jan. 2015

(Permalink) Posted: Jan. 19 2015,08:39   

Cubism: (this is not directly about my last post but answering an earlier clarification from Cubism)

Quote
Does the mutated nucleotide sequence qualify as "new and novel genetic material"? Does the mutated nucleotide sequence contain any "new and novel genetic material"?


The issue here is one of scale ... changing a single nucleotide and getting new functionality is more adequately describing a side-effect of switching a control gene rather than amazingly arriving at a new functional sequence through mutation and selection.

So evolution claims that all the differences between a chimp's DNA and ours are caused by random mutation on 2 hereditary lines from a common ancestor. Find a sequence of unique DNA in the human genome (yup you can use the "junk DNA" now too, now we know it's not junk). Now that string you're holding has arrived by random mutation ... selection has preserved it, but mutation created it. That's what I mean by new/novel/unique information. You may be able to handwave that the switching of a few control genes could happen randomly .... we can all accept that.
But ALL that new (unique) code you're holding had to be created by random mutation and it all had to be preserved through natural selection. The mathematician in me is screaming IMPOSSIBLE! The sheer amount of new and structured information that works not in a sequential code way but in an interdependent, non-linear, overlapping code manner that has been randomly mutated piecewise and each intermediate stage is functionally advantageous is off-the-scale probabilistically impossible (in my opinion ... because those stats couldn't even be quantified).

So my point is ... it's easy to show in a small amount of variability that something magic can happen once in a blue moon (that's just normal probability) ... but for magic to be preserved and extended to the level that evolutionists assert is an obvious extrapolation (molecules-to-man), we need to be seeing a LOT of magic and all the time, everywhere. Nobody gets that much luck ... never-mind claiming that every currently existing organism has had this kind of luck. The computer models (like AVIDA) oversimplify the proposed process by hugely simplifying the organism and only modelling tiny problems, this does not demonstrate the kind of variability and specified complexity we observe ... you should be able to model it to these levels if evolution works.

So to summarise my answer ... when I'm talking about new information ... I'm talking about a unique DNA sequence in an organism. I can't see how it would ever be possible for random mutation to create that much dumb luck even given billions of years. That's not an argument from incredulity ... that's me saying from my knowledge of statistics I can't see how it could ever be possible, so the extraordinary claim of evolutionists needs to be demonstrated in a model that addresses the difficult-to-believe parts of the claim, not just that we can randomly arrive at something useful sometimes

  
N.Wells



Posts: 1836
Joined: Oct. 2005

(Permalink) Posted: Jan. 19 2015,09:14   

Quote
1) living cells process a program called DNA
2) DNA is a code (a sequence of 3 symbols codes for amino acid)

Analogies between DNA and either codes or blueprints are convenient, but they can derail very easily. DNA is a very complex organic molecule that interacts with a very large number of other complex molecules in exceedingly complicated ways.  It is more like a catalyst than a set of instructions.
http://www.huffingtonpost.com/agustin....36.html
http://www.science20.com/chatter....ot_code
http://livinglifewithoutanet.com/2009.......-a-code



   
Quote
5) folded proteins are assembled into complex structures, again this process is controlled by mechanisms which are defined in the code.
The process is affected chemically by both the environment and by other proteins formed from other bits of DNA: No "mechanisms" are "defined" by a "code".


   
Quote

1) given a population of self-replicating entities (a living cell)
2) random mutation of symbols in the raw code (usually point mutations ... about 50-100 de novo germline mutations per generation in humans)
3) advantages are introduced via the mutations by chance, the advantages are preserved and propagate through the population by "natural selection"

1) or multicellular organisms
2) They aren't symbols: if so, what does adenine symbolize?
You are talking about point mutations.  These can be dramatic (cf. myostatin blocker mutations).  However, even more interesting are larger mutations lilke duplications (read up on polyploidy in ornamental and agricultural plants) and translocations.  (You are clear that the advantages are by chance, and not necessarily the mutations, correct?)
3) Also by genetic drift.  Differential advantages are also introduced by recombination, far more than by new mutations (cf. dogs, pet hamsters, pet guppies, the genus Brassica, ornamental pigeons, varieties of chickens, etc., etc.).

   
Quote
4) Natural selection is the name given to the process whereby environmental conditions apply a pressure for survival on the entities.
Genetic drift can also be environmental.  If an ice storm wipes out 100% of the birds of one particular species overwintering in a large area, that will have huge effects via genetic drift and none via selection, at least in the area of 100% mortality.

   
Quote
5) Entities therefore compete for resources and the ultimately surviving hereditary line is preserved, presumably carrying the genetic advantage that allowed it to overcome its competition.
6) Differential reproduction is a fancy way of saying that the hereditary line of some organisms, due to their inferiority in how they are suited to the environment, eventually (at some level of descent) produce no offspring before dying and therefore leave the competition.
7) Death is the only agent in natural selection, regardless of whether some organisms are only slightly better adapted than others, there will eventually be a point where an entire inferior hereditary line is removed (ultimately by death) from the population (i.e. a discrete process, not a continuous one). It is now as though that line had never existed. This is why death is the only important selector, comparisons between organisms do not come into the selection process ... it is simply a case of whether an organism is sufficiently fit to survive long enough in the environment to reproduce.

Because of recombination, two genes in you are not forever linked in your progeny.  Even if they occur close together on the same chromosome, translocations etc. can separate them.  

It is far easier to understand natural selection in terms of differential reproductive success than differential mortality: for example, think of male praying mantises and similar species where reproductive success means getting eaten by their mates, or cecidomyiian gall midges, where reproductive success means the children eating their way out of the mother and killing her.  Yes, sometimes a mutation is very bad, and as a result its owner gets eaten or otherwise dies very quickly, or doesn't even get born in the first place.  However, most variations are not that dramatic, and result in slightly more or slightly fewer offspring, and the genes' fortunes rise and fall on that slight difference.  The less timid mouse may well get eaten sooner by a cat or an owl, but may well have found more food, mated with more females, and fathered more offspring than its neighbor who rarely ventured out of its burrow and died in its bed at a ripe old age (http://www.jstor.org/discover/10.2307/2462815?sid=21105655083353&uid=4&uid=2).

Mortality rates are important, but if you don't do evolutionary accounting in terms of reproductive success, you will severely misunderstand whole areas such as average age of first reproduction, age-specific expectation of future offspring, r & k strategies, kin selection, and so on and so forth
http://www.zo.utexas.edu/courses....ry.html
http://www.zo.utexas.edu/courses....th.html

  
cryptoguru



Posts: 53
Joined: Jan. 2015

(Permalink) Posted: Jan. 19 2015,09:14   

JonF:
you're right ... I can't quantify it. But are you asserting that the variability in the human genome to build a fully functional human, with a functioning brain (and preprogrammed reflexes) and nervous system is equivalent to a sequentially  processed assembler code that just has to make some logic functions?
These are galaxies apart in complexity. And you certainly can't extrapolate the simple case.

Zachriel:
Quote
Evolution also involves recombination and the splicing of new structures from subcomponents.

Sure, but that wouldn't account for significantly long unique sequences that haven't appeared before in the history of that genome. They could only be explained by random mutation coming in somewhere along the way.

Quote
The whole premise of natural selection can be simplified to be death.

People kept saying this ... but it didn't make sense to me to see anything else as the reason for a winner to emerge. Actually you guys may be right about this, now I've thought about it a bit more. If an organism can be removed from the competition through another means other than death (e.g. location) that would work too. So yes, let me retract my statements about Natural Selection only being able to be possibly enforced by death ... ooh I've learned something.

Quote
there are 2.3 x 10^93 different amino acid sequences that could make a working Cytochrome C.

This may be true for a few proteins but certainly isn't the case for all proteins, which require a specific form to be functional. We need to be able to account for the origin of all of them.

  
Zachriel



Posts: 2723
Joined: Sep. 2006

(Permalink) Posted: Jan. 19 2015,09:32   

Quote (cryptoguru @ Jan. 19 2015,09:14)
Zachriel:
Quote
Evolution also involves recombination and the splicing of new structures from subcomponents.

Sure, but that wouldn't account for significantly long unique sequences that haven't appeared before in the history of that genome. They could only be explained by random mutation coming in somewhere along the way.

The process of splicing subcomponents can create new functional structures.

Quote (cryptoguru @ Jan. 19 2015,09:14)
Zachriel:  
Quote
The whole premise of natural selection can be simplified to be death.

People kept saying this ... but it didn't make sense to me to see anything else as the reason for a winner to emerge.

An example was provided, of pacific salmon. They all die. Some die before reproducing. Some die immediately after reproducing. Some produce more or healthier eggs. But they all die.

 
Quote (cryptoguru @ Jan. 19 2015,09:14)
Actually you guys may be right about this, now I've thought about it a bit more. If an organism can be removed from the competition through another means other than death (e.g. location) that would work too. So yes, let me retract my statements about Natural Selection only being able to be possibly enforced by death ... ooh I've learned something.

Good. It's an important point.

--------------

You never step on the same tard twice—for it's not the same tard and you're not the same person.

   
QED



Posts: 41
Joined: July 2008

(Permalink) Posted: Jan. 19 2015,09:35   

"plus a sprinkling of magic selection (which is the secret ingredient that we should never really define properly lest it became understandably deficient) will eventually mutate a Banana into Barack Obama."

Yep- it was inevitable - your "fight" isn't about science at all - it's a socio-political one, crypto. But thanks for letting everyone know.

Next up - why ID should be taught in schools, in
3...2...1...

  
cryptoguru



Posts: 53
Joined: Jan. 2015

(Permalink) Posted: Jan. 19 2015,09:37   

Quote
Analogies between DNA and either codes or blueprints are convenient, but they can derail very easily


I've said this before and I'll assert this again ... DNA is not like a code ... it IS a code and the cell is a Turing-complete computer (not an analogy ... a definition).
It is NOT like a blueprint, a blueprint is a diagram that shows how to build something, it doesn't contain the ability to build the object and necessary materials for the object in itself.


Quote
No "mechanisms" are "defined" by a "code".

That isn't true. DNA code is functionally non-linear ... it is processed by the cell, the expression of proteins is a deterministic process, which is explicitly defined by the DNA code. It uses chemical processes to respond to the environment and toggle control genes and expression of genes, but this functionality is prescribed in the DNA ... if it wasn't deterministic, the DNA could accidentally make a coconut instead of a baby ... my wife would be livid, if she went through all that pain just for a coconut. DNA only does something different if it's responding (according to its own programming) to external events or if it is mutated. Chemistry only describes the physical processes, they do not describe the functionality, which is abstract. That is akin to trying to analyse the functionality of Mac OSX by looking at the flow of electricity through your motherboard, that does not explain the functionality in the program that it is running; just the medium it is using to execute it.

Quote
They aren't symbols: if so, what does adenine symbolize?

They ARE symbols in the representation of the functional code. 3 Symbols constitute a codeword (codon) ... this is what AVIDA works with as its commands. Adenine doesn't "symbolize" anything atomically anymore than the letter 'F' symbolizes anything on your keyboard ... but it is a symbol that is used to build words which are understood in the context of the language they are intended for.

Quote
Genetic drift can also be environmental

True ... I don't see how that affects the scope of what we are considering though.

  
cryptoguru



Posts: 53
Joined: Jan. 2015

(Permalink) Posted: Jan. 19 2015,09:54   

QED'oh:
Quote
Yep- it was inevitable - your "fight" isn't about science at all - it's a socio-political one, crypto. But thanks for letting everyone know.

Your fight with me isn't about science either, it's because I dare to disagree with your paradigm ... who's being socio-political? Science is meant to challenge paradigms, that's how it was started (by creationists).
Quote
wah wah wah
(from earlier)
Come along to an MMA sparring session with me and we'll see who cries first! This pigeon eats off-topic evolutionist evangelists for breakfast. (in a differential fitness sort of way)

So you can shut up now until the conversation ends and you can make your closing speech about how you scared me off good 'n' proper.

At least the other guys on here are having a civil conversation and explaining their view of how it all works.
I'm actually learning something new for once.

  
N.Wells



Posts: 1836
Joined: Oct. 2005

(Permalink) Posted: Jan. 19 2015,09:59   

You are wrong about DNA and codes, but you are far more wrong about
 
Quote
 
Quote

Genetic drift can also be environmental


True ... I don't see how that affects the scope of what we are considering though.

If you look at a bird book for North America, you will see a very large number of eastern and western pairs of forest-dwelling species or subspecies, consisting of distinctive differences in coloration and calls, which have varying degrees of interfertility.  The reason for all these pairs of species appear to be far less a matter of adaptive evolution and natural selection than isolation and genetic drift* following removal of forests across the great plains: e.g., red- and yellow-shafted flickers, lazuli and indigo buntings, black-headed and rose-breasted grosbeaks, Baltimore and Bullock's orioles, etc.  Or look at genetic-distance maps of various groups of chimpanzees or gorillas, mollusk sister-species across the Panamanian Isthmus, and so forth.  (*Yes, bird coloration and songs are reinforced by sexual selection, which is a form of natural selection, but the initial changes would logically have been drift, and mating preferences are not relevant in organisms like sessile clams.)

  
QED



Posts: 41
Joined: July 2008

(Permalink) Posted: Jan. 19 2015,10:00   

I'm not fighting with you, crypto. I'm just making observations that you're not producing any "new information". But have at it. I can wait.

  
cryptoguru



Posts: 53
Joined: Jan. 2015

(Permalink) Posted: Jan. 19 2015,10:03   

N. Wells:
Quote
You are wrong about DNA and codes, but you are far more wrong about

Oh OK ... sure I'll just take your word for it, you being the expert and all that.

Quote
If you look at a bird book for North America, you will see a very large number of eastern and western ... blah blah blah

This is irrelevant to the point we're discussing, which is about how well AVIDA models biological evolution and how a non-linear functional code that is randomly mutated can expect to gain new function over time.

  
N.Wells



Posts: 1836
Joined: Oct. 2005

(Permalink) Posted: Jan. 19 2015,10:07   

Quote (cryptoguru @ Jan. 19 2015,10:03)
N. Wells:
Quote
You are wrong about DNA and codes, but you are far more wrong about

Oh OK ... sure I'll just take your word for it, you being the expert and all that.

Quote
If you look at a bird book for North America, you will see a very large number of eastern and western ... blah blah blah

This is irrelevant to the point we're discussing, which is about how well AVIDA models biological evolution and how a non-linear functional code that is randomly mutated can expect to gain new function over time.

Don't take my word: I gave you three links, and many more can be found to discuss the same issue.

No, at that point you were generalizing about evolution, not discussing Avida.

  
cryptoguru



Posts: 53
Joined: Jan. 2015

(Permalink) Posted: Jan. 19 2015,10:16   

QED:
I explained what "new information" is:-

Quote
when I'm talking about new information ... I'm talking about a unique DNA sequence in an organism.


Not a point-mutated nucleotide or a spliced copy, look at unique continuous sequences of nucleotides in human DNA that are unrepresented in our purported genetic relatives and perform necessary function. You need to explain how those huge amounts of information originated. You guys claim we get those by successive random mutation, I want to see the possibility of that modelled; because to me when you're dealing with that amount of variability, it's completely impossible.
I worked in cryptography for years ... so I am used to evaluating the likelihood of difficult problems being solved by stochastic processes ... I've got a reasonable handle on the statistics and the process (as well as specialising in biological mathematical & computer modelling - PhD) ... to me this is impossible; so show me a working model that represents even some of the complexity of "real" evolution.

  
N.Wells



Posts: 1836
Joined: Oct. 2005

(Permalink) Posted: Jan. 19 2015,10:19   

DNA is not symbolic.  It is a chemical, not a code (or if it is a code, so is water.)  Strictly speaking it is not coded, nor is it decoded or interpreted.  It is a chemical that interacts chemically, which is not true of any other known codes.  (Unless you want to make an issue of smoke signallers coughing.......? :) )

From http://www.antievolution.org/cgi-bin....;t=9614
 
Quote
The way a computer code works is that the exact sequence of the code - the precise order of the binary 1s and 0s - spells out exactly what operations the computer must perform. But in genetics, the sequence is only part of the picture. Just as important are genetic regulatory networks - which genes are turned on at what times and in combination with which other genes. Phenotypes are not simply the result of particular gene sequences but the result of specific gene-gene (or gene network-gene network) interactions.

But DNA bears little relation to a "code" in a more fundamental way. Consider exactly what a "code" is. A code is a system of arbitrary symbols used to represent  ideas and objects. In a sense, language itself is a "code"; the symbol "dog" represents that furry tetrapod with a waggly tail, for example. In a code, the symbols themselves have no inherent meaning. The letter "d" is meaningless by itself, as are the letters "o" and "g". It is only in combination that they derive meaning, and their meaning is derived from the idea that they represent. Furthermore, they only have meaning because we give them meaning. "Dog" is merely the label we apply to Fido; in a universe without sentient beings, "dog" would be meaningless. DNA does not fit this description at all. DNA is not arbitrary in any way; each letter of the genetic "code" is an actual biological compound. ACCGTCGA might be the gene for determining how long your toe hair is, but unlike a code, A, C, T and G each have their own non-arbitrary meaning. And this meaning exists independently of human sentience - the sequence of nucleotides does not have meaning only because we give it meaning. It would have meaning even if humans didn't exist at all.

What DNA is, is a polymeric chemical that follows a dynamic chemical process, governed by universal physical rules. It is only a "code" in the same sense that nuclear fusion is a "code" for how stars produce light


From http://www.talkorigins.org/indexcc....80.html
 
Quote
    The genetic code is not a true code; it is more of a cypher. DNA is a sequence of four different bases (denoted A, C, G, and T) along a backbone. When DNA gets translated to protein, triplets of bases (codons) get converted sequentially to the amino acids that make up the protein, with some codons acting as a "stop" marker. The mapping from codon to amino acid is arbitrary (not completely arbitrary, but close enough for purposes of argument). However, that one mapping step -- from 64 possible codons to 20 amino acids and a stop signal -- is the only arbitrariness in the genetic code. The protein itself is a physical object whose function is determined by its physical properties.

   Furthermore, DNA gets used for more than making proteins. Much DNA is transcribed directly to functional RNA. Other DNA acts to regulate genetic processes. The physical properties of the DNA and RNA, not any arbitrary meanings, determine how they act.

   An essential property of language is that any word can refer to any object. That is not true in genetics. The genetic code which maps codons to proteins could be changed, but doing so would change the meaning of all sequences that code for proteins, and it could not create arbitrary new meanings for all DNA sequences. Genetics is not true language.

   The word frequencies of all natural languages follow a power law (Zipf's Law). DNA does not follow this pattern (Tsonis et al. 1997).

   Language, although symbolic, is still material. For a word to have meaning, the link between the word and its meaning has to be recorded somewhere, usually in people's brains, books, and/or computer memories. Without this material manifestation, language cannot work.


DNA is understandable as a specialized alternative to RNA, and RNA is simultaneously the information, the tools to make stuff, and the material out of which both the tools and the structures are made.  DNA simply holds the information and hands the rest of the capabilities off to proteins.

I mentioned "DNA as blueprints" not because you said it but because it is another common and partially useful analogy for DNA that can similarly derail horrendously when one extends the analogy to areas where it does not apply.

Quote
You need to explain how those huge amounts of information originated. You guys claim we get those by successive random mutation, I want to see the possibility of that modelled; because to me when you're dealing with that amount of variability, it's completely impossible.

Duplication and mutation create new "information", and so does recombination (because so many developmental outcomes result from chemical interactions between gene products rather than just the presence or absence of individual genes - yet another reason why DNA is not just a code).  Also, there is no single "target", but rather a selection of what works well enough or better than the alternatives in terms of resultant numbers of offspring, so simplistic probability calculations don't apply, and neither do simplistic life-or-death dichotomies.

  
JonF



Posts: 634
Joined: Feb. 2005

(Permalink) Posted: Jan. 19 2015,10:31   

Quote (cryptoguru @ Jan. 19 2015,09:39)
Cubism: (this is not directly about my last post but answering an earlier clarification from Cubism)

   
Quote
Does the mutated nucleotide sequence qualify as "new and novel genetic material"? Does the mutated nucleotide sequence contain any "new and novel genetic material"?


The issue here is one of scale ... changing a single nucleotide and getting new functionality is more adequately describing a side-effect of switching a control gene rather than amazingly arriving at a new functional sequence through mutation and selection.

Why?

Quote
The mathematician in me is screaming IMPOSSIBLE!

The mathematician in me is screaming SHOW ME THE NUMBERS!  So far all you've shown is personal incredulity.

  
JonF



Posts: 634
Joined: Feb. 2005

(Permalink) Posted: Jan. 19 2015,10:34   

Quote (cryptoguru @ Jan. 19 2015,10:14)
JonF:
you're right ... I can't quantify it. But are you asserting that the variability in the human genome to build a fully functional human, with a functioning brain (and preprogrammed reflexes) and nervous system is equivalent to a sequentially  processed assembler code that just has to make some logic functions?
These are galaxies apart in complexity. And you certainly can't extrapolate the simple case.

You are asserting a qualitative difference between Avida and biological evolution.  Quantifiable or not, yes, the human body is a lot more complex that what Avida can produce.  Partly due to the lack of time and lots of trials.  But I see no reason why the process that produced EQU in Avida can't produce a human being in real life.

  
JonF



Posts: 634
Joined: Feb. 2005

(Permalink) Posted: Jan. 19 2015,10:35   

Quote (cryptoguru @ Jan. 19 2015,10:37)
Quote
Analogies between DNA and either codes or blueprints are convenient, but they can derail very easily


I've said this before and I'll assert this again ... DNA is not like a code ... it IS a code and the cell is a Turing-complete computer (not an analogy ... a definition).

You don't get to just define a cell as a Turing-complete computer.  You need to demonstrate that the cell meets the requirements of a Turing-complete computer. Which you supposedly know.

I could define a pair of scissors as a Turing-complete computer. But I would be objectively wrong.  Calling a tail a leg doesn't make it one.

  
cryptoguru



Posts: 53
Joined: Jan. 2015

(Permalink) Posted: Jan. 19 2015,10:40   

N.Wells:
who wrote that nonsense??
This point isn't even worth arguing ... it's called "Genetic code" and they're called "codons" for a reason.
It fulfils all the criterion for being a code ... in fact, a computer code; as it is processed by the cell in a Turing-complete manner.
I don't think many people on here would agree with your point ... and interestingly AVIDA itself models the cell as a computer with a processed instruction set.

Information is the arrangement of matter, the DNA code is in the arrangement of the chemicals (not in the chemicals themselves). Information is abstract. DNA is not just a chemical zipper! It is addressable and uses boolean logic to regulate parts of the code, so is functional.
The information in a book is not described by the chemical constituents of the ink and the pages, it is the arrangement of these molecules that contains the information.

  
N.Wells



Posts: 1836
Joined: Oct. 2005

(Permalink) Posted: Jan. 19 2015,10:48   

Quote (cryptoguru @ Jan. 19 2015,10:40)
N.Wells:
who wrote that nonsense??
This point isn't even worth arguing ... it's called "Genetic code" and they're called "codons" for a reason.
It fulfils all the criterion for being a code ... in fact, a computer code; as it is processed by the cell in a Turing-complete manner.
I don't think many people on here would agree with your point ... and interestingly AVIDA itself models the cell as a computer with a processed instruction set.

DNA is extremely complicated, so we do not have language to do it justice, so we use analogies and metaphors.  That's fine, until we mistake the analogy or the metaphor for reality and drive the analogy into areas where it does not pertain.  That's what you are doing.  

Quote
Information is the arrangement of matter, the DNA code is in the arrangement of the chemicals (not in the chemicals themselves). Information is abstract. DNA is not just a chemical zipper! It is addressable and uses boolean logic to regulate parts of the code, so is functional.  The information in a book is not described by the chemical constituents of the ink and the pages, it is the arrangement of these molecules that contains the information.

I'm not sure what you are getting at.   The sequence of nucleotides in a gene determine what protein is produced and how it folds, but a) a lot of variation is tolerated and some makes no difference at all, and b) information in DNA is most specifically not in its arrangement in terms of the physical arrangement of the genes, but merely in the timing of production of proteins and the chemistry of how those proteins interacts.  (Proximity on a chromosome affects the likely of two genes getting separated, but does not mean that they will be expressed at the same time: as you said, not a zipper.)

  
JonF



Posts: 634
Joined: Feb. 2005

(Permalink) Posted: Jan. 19 2015,10:49   

Quote (cryptoguru @ Jan. 19 2015,11:16)
QED:
I explained what "new information" is:-

 
Quote
when I'm talking about new information ... I'm talking about a unique DNA sequence in an organism.


Not a point-mutated nucleotide or a spliced copy, look at unique continuous sequences of nucleotides in human DNA that are unrepresented in our purported genetic relatives and perform necessary function.

That explanation is incompatible with the previous definition. A point mutation produces a unique DNA sequence.  You are trying to be as vague as possible. Some mathematician.

Do you know what an operational definition is?  Your first try was one, your second try isn't.

Quote
You need to explain how those huge amounts of information originated. You guys claim we get those by successive random mutation, I want to see the possibility of that modelled; because to me when you're dealing with that amount of variability, it's completely impossible.

Yeah, good ol' personal incredulity.  Much better than evidence.

Quote
I worked in cryptography for years ... so I am used to evaluating the likelihood of difficult problems being solved by stochastic processes ... I've got a reasonable handle on the statistics and the process (as well as specialising in biological mathematical & computer modelling - PhD) ... to me this is impossible; so show me a working model that represents even some of the complexity of "real" evolution.

You need to define exactly want you mean by representing some of the complexity of "real" evolution.  You need to have an operational definition of the product of such a model, and you need to demonstrate that such a model is practical.

Meanwhile we've got lots of examples of real and modeled complex evolution, not as complex as human evolution, but nobody's found any reason why the results we have cannot be extrapolated to "real " evolution. And there have been lots of studies of what rates of various processes are required to produce what we see.  Are you familiar with those studies?  Would you care to guess what their results are?

  
cryptoguru



Posts: 53
Joined: Jan. 2015

(Permalink) Posted: Jan. 19 2015,11:06   

JonF:
Yes I am personally incredulous ... I am not asserting that proves anything ... I'm saying that the burden of proof is on you to show that this (relatively new) proposed mechanism which seems to defy laws of logic could even theoretically do what is suggested

Quote
But I see no reason why the process that produced EQU in Avida can't produce a human being in real life.

That's because you want to believe that ... it hasn't been demonstrated that it's even possible. People are claiming that AVIDA proves biological evolution ... it doesn't! We just showed that ... It proves you can find a logical function using stochastic process on a Turing-complete set of instructions.


Quote
You don't get to just define a cell as a Turing-complete computer.

It's actually probably even Turing-universal
Efficient Turing-universal computation with DNA polymers

  
cryptoguru



Posts: 53
Joined: Jan. 2015

(Permalink) Posted: Jan. 19 2015,11:19   

Quote
That explanation is incompatible with the previous definition. A point mutation produces a unique DNA sequence.  You are trying to be as vague as possible. Some mathematician.

You misunderstood what I said there. I said "point mutation" singular ... not sequence of point mutations.

EXAMPLE

GCATTUCAGUUGATUCCATATGGCTU
GCATTUCAGTUGATUCCATATGGCTU

The U has been point-mutated to a T ... I don't consider the whole sequence new information. I'm not even going to consider it "new information" until there is a reasonably sized sequence of unique information. Let's set it to be bigger than you could create by chance.

For example, a sequence of 200 nucleotides that is unique to that organism requires 1 in 4^200 chance of arriving at randomly in the traditional way of monkeys typing randomly on a base-4 keyboard. So I want to see how the process of evolution could theoretically produce something of that complexity ... i.e. too difficult to get using monkeys, yet functional.

  
midwifetoad



Posts: 4003
Joined: Mar. 2008

(Permalink) Posted: Jan. 19 2015,11:48   

Quote
The U has been point-mutated to a T ... I don't consider the whole sequence new information. I'm not even going to consider it "new information" until there is a reasonably sized sequence of unique information. Let's set it to be bigger than you could create by chance.


What's the minimal size sequence change necessary to make a difference in viability or relative reproductive success?

--------------
Any version of ID consistent with all the evidence is indistinguishable from evolution.

  
JonF



Posts: 634
Joined: Feb. 2005

(Permalink) Posted: Jan. 19 2015,11:51   

Quote (cryptoguru @ Jan. 19 2015,12:06)
JonF:
Yes I am personally incredulous ... I am not asserting that proves anything ... I'm saying that the burden of proof is on you to show that this (relatively new) proposed mechanism which seems to defy laws of logic could even theoretically do what is suggested

I don't see it as defying the laws of logic. Eaxactly what laws nd how do they defy them?

It is nice that you admitted all you have is armchair philosophizing.  

Quote
 
Quote
But I see no reason why the process that produced EQU in Avida can't produce a human being in real life.

That's because you want to believe that ... it hasn't been demonstrated that it's even possible.

Indeed we have not directly observed biological evolution on the scale of hundreds of thousands of years and widely varying environments. There's good reasons for that; it's totally impractical.  I haven't seen any objective reasons why the extrapolation isn't reasonable. From you or anyone.

Quote
People are claiming that AVIDA proves biological evolution ... it doesn't! We just showed that ... It proves you can find a logical function using stochastic process on a Turing-complete set of instructions.

The stochastic process of descent with modification and selection.  I.e. evolution.

Quote
Quote
You don't get to just define a cell as a Turing-complete computer.

It's actually probably even Turing-universal
Efficient Turing-universal computation with DNA polymers

DNA can be used as a computer in a test tube. That's not what it does in the cell. Try again.
xxx

  
Doc Bill



Posts: 1039
Joined: April 2007

(Permalink) Posted: Jan. 19 2015,12:01   

It appears that CreeptoGuru suffered a concussion to have written this creationist boilerplate tripe:

Quote
You misunderstood what I said there. I said "point mutation" singular ... not sequence of point mutations.

EXAMPLE

GCATTUCAGUUGATUCCATATGGCTU
GCATTUCAGTUGATUCCATATGGCTU

The U has been point-mutated to a T ... I don't consider the whole sequence new information. I'm not even going to consider it "new information" until there is a reasonably sized sequence of unique information. Let's set it to be bigger than you could create by chance.


Now Creepto is defining his own terms again in fine creationist style.  You'll have to better than this to impress us, boy-o!

Creepto ate the dog.
Creepto ate the hog.

In Creepto-speak there is no new information.  I guess you'd need something like this:

Creepto ate the dog.
Creepto prepared the mongoose in a rich butter sauce.

Oh, now I get it.  The mongoose recipe is definitely bigger than I could get by chance, although instead of butter I would have fried that puppy in tard.

  
JonF



Posts: 634
Joined: Feb. 2005

(Permalink) Posted: Jan. 19 2015,12:02   

Quote (cryptoguru @ Jan. 19 2015,12:19)
 
Quote
That explanation is incompatible with the previous definition. A point mutation produces a unique DNA sequence.  You are trying to be as vague as possible. Some mathematician.

You misunderstood what I said there. I said "point mutation" singular ... not sequence of point mutations.

EXAMPLE

GCATTUCAGUUGATUCCATATGGCTU
GCATTUCAGTUGATUCCATATGGCTU

The U has been point-mutated to a T ... I don't consider the whole sequence new information. I'm not even going to consider it "new information" until there is a reasonably sized sequence of unique information. Let's set it to be bigger than you could create by chance.

For example, a sequence of 200 nucleotides that is unique to that organism requires 1 in 4^200 chance of arriving at randomly in the traditional way of monkeys typing randomly on a base-4 keyboard. So I want to see how the process of evolution could theoretically produce something of that complexity ... i.e. too difficult to get using monkeys, yet functional.

Then you need a new operational definition of information, since by your original definition a point mutation adds information.

Gene duplication followed by point mutation.  Much longer than 200 nucleotides. and observed over and over again.

{ABE}Gene Genesis: Scientists Observe New Genes Evolving from Mutated Copies
Evolution by gene duplication: an update

  
  336 replies since Jan. 16 2015,08:04 < Next Oldest | Next Newest >  

Pages: (12) < 1 2 3 [4] 5 6 7 8 9 ... >   


Track this topic Email this topic Print this topic

[ Read the Board Rules ] | [Useful Links] | [Evolving Designs]