DNA : When Is A Code Not A Code ?

About Patrick Lockerby

Retired engineer, 60+ years young.
Computer builder and programmer.
Linguist specialising in language acquisition and computational linguistics.
Interested in every human endeavour except the... Full Bio

Patrick
Lockerby

DNA : When Is A Code Not A Code ?

Stephen C. Meyer is an intelligent design advocate and a co-founder of the Discovery Institute.

The core argument of Stephen Meyer’s book, Signature in a Cell, written in advocacy of intelligent design, is this: DNA is a code and a computer instruction is a code. Since computer code requires an intelligent designer, and DNA is a code, it follows that DNA is a product of, or is controlled by, an intelligent designer.

This argument has no foundation if one does not accept its basic premises: that DNA is a code that a computer instruction is a code, and that the term 'code' is applicable in exactly the same way to both uses.

Men take the words they find in use amongst their neighbours; and that they may not seem ignorant what they stand for, use them confidently, without much troubling their heads about a certain fixed meaning; whereby, besides the ease of it, they obtain this advantage, That, as in such discourses they seldom are in the right, so they are as seldom to be convinced that they are in the wrong;

John Locke. An Essay Concerning Human Understanding, Book III, Chapter X.
http://www.gutenberg.org/browse/authors/l#a2447

Before leaping to any conclusions based on our use of the word 'code', we must, if we are to be scientific, first define 'code'.

A code is a member of the class 'symbols'. A first level symbol is a label which is used in place of the thing which it identifies. For example, suppose a building with a sign over the window which bears the word 'pharmacy'. We can use the symbol 'pharmacy' in language as a symbolic substitute for any real pharmacy. Suppose now that we invent a slang term 'pill-farm' to mean 'pharmacy'. We now have a secondary label 'pill-farm' which is a second-level symbol for 'pharmacy'. 'Pharmacy' in its turn is a first level symbol for a real building of a specific type.

By convention, a primary symbol is a name, but any secondary symbol is a code: a symbol which stands in place of another symbol. For purposes of clarification, I will give another example. 'And so forth' is a primary label or symbol for an idea. By converting it into Latin, a language spoken by few speakers of English, we encode it as 'et cetera'. We now abbreviate it to 'etc.', a second level coding.

A code is not a symbol. A symbol is not a code. A symbol stands in place of an object or idea. A code stands in place of a symbol: it is a symbol for a symbol.

In computer instructions, we start with the simplest possible representations of what is going on inside a computer chip. We observe that a location in a computer chip can be at one of two voltages. Taking these voltages as our idea we invent symbols for the two voltages: '1' and '0'. These are our primary symbols and they can only be written as binary expressions.

As a convenience, we can use a form of abbreviation which is easier for humans to handle than binary. The most common such abbreviation is hexadecimal code, or hex. As an example, the binary 1010 0101 can be written as A5 in hex. Note that hex, being a secondary symbol level is a code.

When dealing with binary as computer instructions rather than as numbers it is convenient to use mnemonic codes. It may be that the binary string 1111 0000 1100 0100, or F0C4 in hex, is an instruction to the computer core, expressed as F0, to jump to memory location C4, but only IF a previously computed result was non-zero. We can write that as a mnemonic code: JNZ C4.

Such mnemonics are called assembly language. The 'assembly' part of the name comes from the fact that this mnemonic code needs to be assembled into a package of binary numbers in order for the computer to be able to use it as a program.

DNA is a string of molecules. There are four main components: guanine, adenine, thymine and cytosine. Those names, the words 'guanine', 'adenine', 'thymine' and 'cytosine' are primary symbols invented by humans to identify the physical molecules which are found in DNA.

For convenience, we often abbreviate these symbols to CAGT, so that we can more readily handle the huge volume of data which we have accumulated about DNA. Please observe: there exists a long molecule of a type which we label DNA. It has four major components to which we assign symbols as names. We next assign symbols to the name symbols as an abbreviating code. We humans have agreed to assign the four letters CAGT as a code for the symbols which in turn stand for the molecular components of DNA.

A code is a symbol which stands in place of a symbol. The four letters CAGT most definitely form a code, being symbols for the names of the four major components of DNA. The names guanine, adenine, thymine and cytosine are not codes: they are primary symbols. Primary symbols stand for real things and not for symbols. The real physical entities guanine, adenine, thymine and cytosine are not codes. If anyone wants to call them codes, let them point to the symbols which might be replaced by these 'codes'.

A computer code is a set of numerical values sufficient and necessary to the production of an end state from an initial state.

DNA is necessary but not sufficient to the production of an end state from an initial state.

To claim that computer code and DNA are both codes is an abuse of the power of words. It is decidedly not scientific.

Concluding remarks:

Anyone who already believes in intelligent design will derive no new knowledge from Stephen Meyer's book.
Anyone who believes in a rigorous approach to science will derive no new knowledge from Stephen Meyer's book.

I conclude that no value is to be obtained from Stephen Meyer's book by any thinking person.

Edit: further reading
More articles in the debate about whether or not DNA is "really" a code.
http://www.cosmicfingerprints.com/dnanotcode.htm
http://hambydammit.wordpress.com/2009/07/05/dna-is-not-a-code/
http://www.naontiotami.com/?p=612
http://ds9a.nl/amazing-dna/ - a highly recommended read!

Comments

I like that :)

Gerhard Adam | 07/06/09 | 21:39 PM

The writer's objections are trivial.
"A computer code is a set of numerical values sufficient
and necessary to the production of an end state from an
initial state.
DNA is necessary but not sufficient to the production of
an end state from an initial state."
The computer code is NOT sufficient, either. Both the computer code and DNA require a system of some sort to read the information and take corresponding actions -- a Turing machine or computer in the case of computer code, and a living cell of the appropriate organism in the case of DNA.

Both DNA and computer code contain information, and in all our experience and theory information comes only from an intelligent agent. Semantic quibbles about what is or isn't technically a "code" are irrelevant.

Dennis Murphy (not verified) | 07/07/09 | 10:55 AM

... in all our experience and theory information comes only from an intelligent agent. Semantic quibbles about what is or isn't technically a "code" are irrelevant.

And what experience would that be? Our experience with human inventions? Every interaction between an organism's sense organs and the environment is an exchange of information. It is presumptious in the extreme to suggest that this information exchange occurs only between intelligent agents.

In addition, the DNA argument is a cop-out since the ID position presumes that ALL of the laws of nature contain information and are therefore the product of an intelligent designer.

Gerhard Adam | 07/07/09 | 11:39 AM

"... the ID position presumes that ALL of the laws of nature contain information and are therefore the product of an intelligent designer. "

That's funny. I've been reading literature from Intelligent Design advocates for several years, and never have come across a statement like that. Thanks for enlightening me.

Dennis Murphy (not verified) | 07/08/09 | 08:28 AM

.. in all our experience and theory information comes only from an intelligent agent.

I'm late to the game as usual, but Gerhard, you've picked up on the key weak point of the design argument. Most of our experience with complex, information-rich systems shows that such systems are generated without intelligent agents. Human inventions make up a small minority of the complex systems around us - there is no justification for claiming that our experience with human systems counts as 'all our experience.'

As for the claim that theory tells us information only comes from an intelligent agent, the only theory that does that is the theory put forward by intelligent design pseudoscientists. There is plenty of theoretical work by real biologists laying out how new biological information can arise without intelligent engineering.

Michael White | 07/08/09 | 11:17 AM

Thanks for your reply.

I would like to know where I can find an example of information generated without intelligent agents. I don't mean that as sarcasm. I am genuinely interested. But it would have to be the creation of real information without an intelligent source. In other words, it would have to occur as a result of natural, undirected processes. If it's just copying or rearranging portions of existing information, especially if the process itself is directed or controlled by existing information (eg under control of a living cell), that would not count.

I admit I should have written my original comments about information more carefully. By "all our experience," I meant our experience with information for which we know the origin. In every such case, the origin or cause of that information is an intelligent agent. I don't think an example exists in which non-trivial information is created by natural processes.

And I should have omitted mention of "theory." Obviously there are many theories, because this remains a mystery.

I'd also like to respond to this comment you made:
"This is a site for genuine science, and not a place for posts advocating creationism."

I don't see where creationism has been advocated in this discussion. I've supported Mr Meyers' statements that natural processes alone are inadequate to create the massive amounts of information stored in living cells. Mr Lockerby mentioned the possibility of "an Other - a 'highest force' for good," but I don't think he intended that as an explanation or topic for further discussion. Who is advocating creationism, unless you have defined that term to include anyone who disagrees with a strictly materialist philosophy?

Have a good day.

Dennis Murphy (not verified) | 07/09/09 | 11:21 AM

Dennis: I am confident that Michael's comment was not in reference to this particular article or comments.

I don't think an example exists in which non-trivial information is created by natural processes.

Here is a simple example of information in physics. A photon has a specific frequency which relates to the change of state of the electron which emitted it. We can use that information to determine the chemical nature of the atom that emitted the photon. Before helium was discovered on Earth its presence in the sun was determined by spectral analysis. The spectrum of a star contains non-trivial information.

In a specific semantic sense of the word 'information' we can say that stars 'inform' us, through the light they give out, about their chemical composition, mass and velocity - the latter due to frequency shift effects. However, we must never forget that it is not the 'purpose' of a star to inform us humans about their physical chemistry.

Patrick Lockerby | 07/09/09 | 11:40 AM

Thanks for replying again, Patrick.

I think we're talking past one another in this discussion of information and what is required to create it. Your example of "information in physics" is more interesting but not qualitatively different than your earlier use of "red sky at morning," to which I already responded. I agree that the light from a star can, in a manner of speaking, 'inform' us of much about that star, but the situation is not qualitatively different than the way that a red sky in the morning 'informs' sailors that bad weather can be expected that day. The spectrum of a star is a direct result of the fusion process occurring within it and the presence of specific elements. If another star has the same composition, it will also produce the same spectrum. You will not have that kind of star without that given spectrum, nor will you find that same spectrum without that kind of star. One (the spectrum) is a direct consequence of the other (the star). I'll let you label what kind of 'information' its spectrum contains, to distinguish it from the next kind.

When I observe that star, I use strings of numbers to designate the wavelengths of lines I detect in its spectrum. Unlike the specific star and its resulting spectrum, there is no relationship or law to cause a '5' to appear after a '1' in a certain position, or for any particular digit to appear after any other. The only reason they appear in the order they do is that I put them that way to represent 'information' I want to preserve.

So there are the two kinds of 'information' we seem to be independently describing. As you wrote, stars can 'inform' us about themselves by the light they produce, but it is not the purpose of the stars to do that. Likewise, the numbers I recorded about the star can 'inform' you about it, but it *is* the purpose of my numbers to do that. In contrast to the stars, which each produce their unique light as a necessary consequence of their specific composition, there is no necessary reason why one number symbol follows another in my recorded data, except that I chose their sequence to convey specific meaning.

I will let you assign the appropriate terminology to distinguish between those very different kinds of 'information.' The question we're discussing, that Mr Meyers' book is about, is the cause of the latter type of information. The sequence of bases in the DNA molecule within every living organism on earth is an example of that type of information. His position is that it cannot be created except by the deliberate action of an intelligent agent.

I apologize if I'm being repetitious, or 'beating a dead horse.' If we still can't connect on the subject, I'll be glad for the conversation we've shared, and watch for some other opportunity to discuss the subject in a different way.

Have a good day.

Anonymous (not verified) | 07/09/09 | 22:59 PM

I apologize for the above "anonymous" posting. I was writing from a different computer, and didn't realize that my name wouldn't appear.

Dennis Murphy (not verified) | 07/10/09 | 07:41 AM

Consider memorizing the string of numbers you generated. In this case, you will have an electro-chemical process defined by the physiology of your brain that encodes "information" without any external intelligent agent . You can't count yourself since you are the product of those electro-chemical processes. Similarly your memory could be destroyed by other processes or interventions, so the storage of that information takes place without anything except electro-chemical processes intrinsic in the functioning of the brain.

Similarly there are oscillating chemical reactions that behave in a particular fashion that clearly convey information to an external observer, however the reactions themselves determine what the subsequent steps will be and how they cycle back around (this is also a mechanism present in many cellular operations).

The point is that there are many instances of where "information" is encoded which can be used downstream by other processes without any intelligent intervention required. The problem is that DNA is often compared to computer programs (which it is not) and therefore we presume that we understand the operation as a precise set of instructions instead of the chemical reactions that it actually contains. In other words, if a particular molecule expresses itself by producing a structure in a cell, this is no different than another chemical reaction forming a crystal or producing another chemical byproduct used in another reaction (without intervention).

Gerhard Adam | 07/10/09 | 13:14 PM

The problem with the statement "information generated" is that you have to define what precisely you're looking for as "information". The reason being that almost anything can be information based on how it is interpreted by an observer.

Every interaction with your senses results in information being transferred for analysis by your brain. Concepts like the color "red" are impossible to convey without that filter since any blind person my understand the technical details about the wavelength, but they wouldn't know what "redness" was because they lacked the ability to interpret that information.

If we consider something like forensics, then we are taking a collection of pieces of information and attempting to construct a story or fill gaps in a story. This allows us to use discrete pieces of information to formulate more complexity.

What makes something like DNA appear so complicated,is that people see it as a single step process and then jump to the conclusion that something like this couldn't just materialize. They are correct, since to make such a leap in a single step would truly be a miracle. However, over time, as processes build on each other and more pieces accumulate, a great deal of complexity can be introduced without any previous intent being required.

This is demonstrated by the existence of "conserved processes" within the genome where many genes were not invented over and over but rather they were reused in a variety of ways, so that they could be adapted to numerous physical variations. That's why the hand bones don't require a redesign, whether they are used in a dolphins flipper, or a chimpanzee's hand.

Gerhard Adam | 07/09/09 | 13:20 PM

in all our experience and theory information comes only from an intelligent agent

Actually, Dennis, 'information' is an anthropocentric concept abstracted from observations by an intelligence - in this case, us.

"Red sky in the morning, sailors take warning." Is the information about the coming foul weather to be found in the red sky, or in the sailors' interpretation of what a red sky means? Nothing which might be called 'information' can exist without an intelligence, this much is true. But the intelligence is only required to identify the entity as symbolic information, not to create the entity.

When we see a face in a cloud, is the face an aspect of our own intelligence at work, or is it evidence for clouds being the product of intelligent design?

It is in our nature to see 'meaning' in everything. That is to say, we humans are endowed with an ability to see an entity and assign it a symbolic value as a portent or indicator of another entity. The good news is that this ability to decide that things have 'meaning', that some things are symbols for other things led to the development of all human languages. The bad news is: since meaning is subjective, it follows that every word in our language imputes a subjective 'meaning' to its instantaneous user. Every word in every human language can convey a meaning from one human to another only through anthropocentric, ethnocentric or egocentric commonality of purpose in the communicators.

It may well be that there is more to the universe than our purely mechanistic notions permit us to know. It may well be that a study of Writings - inspired works which I honour with a capital letter - can give insights into the presence of an Other - a 'highest force' for good, if you will. But to even attempt to discuss that Other is an exercise in futility. As soon as we use our inefficient human language to discuss a topic, the anthropocentric, ethnocentric and egocentric components of the meanings of words intrude stealthily into the matter under discussion and lead us to believe that we really and truly know something which is, and forever shall be unknowable.

Patrick Lockerby | 07/07/09 | 11:52 AM

Amen, Brother.

Brian Taylor | 07/07/09 | 20:27 PM

I think the author of this article should verify the arguments that he wrote down here. They don't seem to hold up.

I see no difference between "JNZ C4" and "ACATG...", even after reading this article three times.

* DNA (physical DNA molecule) is not code.
* DNA code as written down by humans (e.g. ACATG...), is code.
* Humans have the capability to produce DNA based on code.

* Voltages in hardware are not code.
* Information stored in voltages can be written down by humans as assembly code (e.g. JNZ C4), which is code.
* Humans have the capability to produce certain voltages in hardware, based on code.

Even moreso (although perhaps not relevant) - it is possible (although likely not done by any human yet) to store DNA information (code) on computer hardware, or store Assembly code on DNA.

PS: Computer code is no more sufficient than DNA in producing an end state from an initial state.
PPS: I do think, however, that there can be no such direct implication, that code implies it was written by an intelligent designer.

Anonymous (not verified) | 07/07/09 | 11:15 AM

Computer code is no more sufficient than DNA in producing an end state from an initial state.

Perhaps I can clarify this. In a computer, it is possible to construct a whole from numerical parts with high accuracy. Think of a CAD design. From a set of numbers which describe vectors, scales and colours, we can run a program which will output a technical drawing of a complex object. Every time we run the program, the output is identical, unless the hardware develops a fault.

In the case of DNA, components are collected together to create long chain molecules. Molecules are not rigid structures. They are highly eleastic and can fold into various shapes. The way that a long molecule reacts with other molecules is governed to a very high degree by its shape. Even in the absence of a fault in a cell, no two molecules produced by any one cell can be absolutely identical.

A comparison between the product of a cell based on DNA and the product of a computer program based on binary code is not a fair comparison. A computer process produces identical results for every iteration using only binary code. A cell process requires much more than DNA for its operation and can never produce identical results for any two iterations.

References:
http://nobelprize.org/educational_games/medicine/dna/a/translation/index...
http://nobelprize.org/nobel_prizes/chemistry/laureates/1984/speedread.html

Patrick Lockerby | 07/07/09 | 12:11 PM

"Stephen C. Meyer is an intelligent design advocate"

Stephen C. Meyer is a creationist. He believes in magic.

bobxxxx (not verified) | 07/07/09 | 12:25 PM

The moron Dennis Murphy wrote "Both DNA and computer code contain information, and in all our experience and theory information comes only from an intelligent agent."

Translation: THE MAGIC MAN DID IT.

Bloody god-soaked idiot.

bobxxxx (not verified) | 07/07/09 | 12:29 PM

It is never my wish to do anybody's thinking for them. I prefer to present argument and let people discover for themselves whatever 'truth' they feel comfortable with. Accordingly, I have just edited this article to add some further reading links at the foot.

Patrick Lockerby | 07/07/09 | 14:09 PM

Thanks for your thoughtful reply, Mr Lockerby.

I still think you're placing too much emphasis on abstractions, and as a result you overlook the real issue and the substance of Mr Meyers' book.

Suppose the 'red sky' proverb is actually true. I mean that, whatever atmospheric conditions necessarily develop into bad weather for sailors always first creates a red morning sky. Natural laws determine that One condition (red sky) therefore precedes the other (bad weather). One should not call that 'information.' One element is dependent upon the other. If you have the first, you will have the second. You cannot generate 'information' if the elements you will use are not independent of one another. Or, if you insist on defining that as 'information,' it is of the most trivial nature.

(Continuing your example, the sailors' intelligent choice of a 'symbol' for red sky could be followed by an almost infinite number of second symbols, but they choose one to indicate that after red sky comes bad weather. There is your information.)

Subjective, perceived faces in clouds are one thing. Assemblies of independent characters into a specific, highly complex, and recognizable pattern are something else. Are you seriously trying to associate cloud pictures with the binary code used in computers or the 4-bit code used in DNA? There is surely a significant and meaningful difference between those examples.

I think your arguments are reflected in Mr Meyers' recognition that even random, natural processes can create patterns that might appear to be the work of an intelligent agent, but the probability that they did so decreases as the complexity and specificity of the pattern increases. Mr Meyers would recognize that some things may be difficult to determine as the result of an intelligent agent, and other things will be clearly indisputable as such. We might quibble over what quantity of perceived information is the threshold, but it does exist, and things like software or DNA are clearly beyond it.

Regarding your second reply:
"Molecules are not rigid structures. They are highly elastic and can fold into various shapes. The way that a long molecule reacts with other molecules is governed to a very high degree by its shape. Even in the absence of a fault in a cell, no two molecules produced by any one cell can be absolutely identical."

I think you're wrong there. Proteins are extremely long molecules that are folded into very specific shapes, determined by the DNA code and the infrastructure of the living cell. Many billions of identical copies are made, otherwise our bodies and intelligent minds could not function.

Have a good day.

Dennis Murphy (not verified) | 07/08/09 | 08:11 AM

Patrick, interesting article as usual! I never really thought about the semantic difference between code and symbol, and I think you explain it well.

Dennis, I think you make a valid point - protein folding is very specific, true. (I remember spending an entire semester in grad school just on folding, the Levinthal paradox, kinetics, folding funnel, etc.) I think, and here I'm assuming, that Patrick's point was that individual molecules are elastic - but when strung together, they take on more requirements.

Becky Jungbauer | 07/08/09 | 08:53 AM

Thank you. Actually, though, I was not disputing Mr Lockerby's statement about elasticity. I meant only to question the claim that no two molecules (suitably complex, of course) produced by a cell can be "absolutely identical." The composition and the functional shape of the molecule seem to be the aspects that would have to be identical, at least most of the time, for the cell to function. Two elastic objects (molecules) that differed only in their current flexed position would still be considered identical, wouldn't they?

Perhaps he meant something more, and in any case I didn't mean to attach too much importance to the subject. It was more of a side issue, and I don't think it impacts the main question of this thread one way or another.

Dennis Murphy (not verified) | 07/09/09 | 10:23 AM

For "Bobxxx," his insulting posts add nothing to the discussion here, and don't even deserve this reply.

Dennis Murphy (not verified) | 07/08/09 | 08:11 AM

Patrick -- Thanks for a wonderfully thoughtful review of Stephen Meyer's book. I hope that your readers will compare it to the review posted by Jeffrey Dach elsewhere in Scientific Blogging. Regarding Dennis Murphy's concern about proteins, it is true that a gene is translated into a precise sequence of amino acids in a given protein, and that the resulting proteins fold into specific three-dimensional structures. However, that structure is not like a hard little piece of plastic, but has a certain amount of flexibility that is essential to protein function. For instance, when enzymes bind substrate molecules, they typically undergo a conformational change in their structure and fold more tightly around the substrate. The amount of motion is surprising -- when a hexokinase enzyme binds a glucose molecule, it looks like a nanoscopic PacMan gobbling a morsel. Bottom line: protein sequences are identical, but the molecular shape of a protein molecule varies over time within certain limits.

Dave (not verified) | 07/08/09 | 09:23 AM

Patrick -- Thanks for a wonderfully thoughtful review of Stephen Meyer's book. I hope that your readers will compare it to the review posted by Jeffrey Dach elsewhere in Scientific Blogging.

I'd actually been wondering how that stayed up. Turns out one of our biology moderators is on a trip and one was doing some personal stuff but I believe the life expectancy of that piece is getting short.

Hank Campbell | 07/08/09 | 11:35 AM

Yeah, sorry for being slow on the trigger. This is a site for genuine science, and not a place for posts advocating creationism.

Michael White | 07/08/09 | 12:22 PM

Ahh...suddenly everything becomes clear. I was really curious as to how that was still around.

Becky Jungbauer | 07/08/09 | 12:54 PM

I'm gone for a few days and creationists crawl out of the woodwork!

Michael White | 07/08/09 | 16:15 PM

Michael said

I'm gone for a few days and creationists crawl out of the woodwork!

I suppose you would call any physicist the perjorative name, "creationist", if they advocate the Big Bang theory. Your atheist bias is showing. This is not about science. Its about advancing your atheist agenda.

Jeffrey Dach | 07/09/09 | 09:56 AM

And suddenly everything became clear. I was puzzled by the vaguely familiar form of the statue shown in your previous photo. It was fun to recognize your new picture!

Dennis Murphy (not verified) | 07/09/09 | 10:27 AM

Thanks! It's one of my favorite monuments in D.C. but most people don't even know it's there.

Becky Jungbauer | 07/09/09 | 10:31 AM

Thank you all for your comments.

Regarding the comparison between numbers and molecules as products of processes, I thought I would add a few remarks.

In a computer program, an intermediate result may be used as a factor to produce further results. The accuracy of any intermediate or final result depends on the choice of data type, e.g. signed or unsigned 8, 16, 32, 64 bits, IEEE single/double precision floating point, and such. However, once the data type is rigidly defined, repeat runs of a (fully debugged) program with the same starting data will always produce the same final results.

Compare this with what happens in a cell. If a protein is taken as an intermediate result, then there is a source of compounded errors. If two proteins having the same chemical constituents are folded differently then it may happen that the next stage of processing, being highly sensitive to the precision of the folding, will cause the two proteins to be processed differently.

An example of a process where such divergence is possible is posttranslational modification. A protein, having been constructed by use of the DNA/RNA mechanism then goes on to be modified by the attachment of other biochemical components not directly mediated by DNA/RNA. A mis-folding at this stage may cause an attachment point for a chemical component to be buried inside the molecule, where attachment is physically blocked. Even worse, a mis-folded protein may actually be harmful to the parent cell.

This, then is the essential difference between computer code and the DNA-mediated functions of a cell. Computer code produces 100% predictable and replicable results. The final state of a cell resulting from some initial condition and the operation of the DNA transcription mechanism is not 100% predictable.

Further reading:
http://en.wikipedia.org/wiki/Posttranslational_modification
http://folding.stanford.edu/English/Science

If any biochemist wants to correct any errors I may have made, I would be delighted to learn more.

Patrick Lockerby | 07/09/09 | 11:07 AM

Part of the issue is that the cell is a chemical system, while a computer isn't. Coding, computation, and information in chemical systems differ in many ways from coding, computation, and information in computer systems.
This is where analogies between man-made and natural systems frequently break down. Human machines really aren't that similar to protein 'machines' - a major difference is chemistry.

Michael White | 07/09/09 | 13:26 PM

Mr. Lockerby said:
"A computer code is a set of numerical values sufficient and necessary to the production of an end state from an initial state.
DNA is necessary but not sufficient to the production of an end state from an initial state.A computer code is a set of numerical values sufficient and necessary to the production of an end state from an initial state."

This is not correct. A computer code is no more sufficient and necessary to the production of an end state from an initial state than DNA. they both require a system and interpretation.

"To claim that computer code and DNA are both codes is an abuse of the power of words. It is decidedly not scientific."

This is not a justified conclusion. This is an irrational, biased attack following a sleight of hand using semantics, which clearly is in opposition to the standard definition of the word "code".

Code (disambiguation)
From Wikipedia, the free encyclopedia
"A code is a rule for converting a piece of information into another object or action, not necessarily of the same sort."

Genetic code
From Wikipedia, the free encyclopedia
Main article: Genetic code
"Biological organisms contain genetic material that is used to control their function and development. This is the DNA, which contains units named genes that can produce proteins through a code (genetic code) in which a series of triplets of four possible nucleotides are translated into one of twenty possible amino acids."

Stephen Swires (not verified) | 07/15/09 | 23:20 PM

Stephen: my point was that to compare any two disciplines fairly and scientifically using the same term, that term must be defined to mean exactly the same thing in both disciplines.

As you have pointed out with your links to the " the standard definition of the word code ", aka Wikipedia, there is one definition for computer code and a different one for biological code.

My point, exactly.

Patrick Lockerby | 07/16/09 | 00:20 AM

Patrick,
As I said, you are just playing a semantics game and using sleight of hand. You are obviously biased against anything Stephen Mayer has to say because you label him an intelligent design advocate, and therefore you must grasp at even the tiniest straw you think you can to discredit what he has to say. Why not be truly objective and consider that some of what he has to say may have some merit? I have not read his book, but now I am interested in reading it.

The definition given for "code": "A code is a rule for converting a piece of information into another object or action, not necessarily of the same sort." is equally suitable for defining the usage of the word "code" in both computer programming and in genetics.

A key element of language is the fact that we can use words to describe things that fit a pattern. Otherwise, we would have to invent a new word for everything in every discipline and everything we observe. That would make language impossibly verbose.

No doubt there are differences between the disciplines of computer programming and genetics, but that does not mean that are not many parallels to the point of being quite amazing, that we should look at the parallels, and consider the implications of them. That is, if we truly have an open, objective mind.

Best regards,
Stephen

Stephen Swires (not verified) | 07/16/09 | 12:25 PM

"my point was that to compare any two disciplines fairly and scientifically using the same term, that term must be defined to mean exactly the same thing in both disciplines."

Hi Patrick,
I was trying to accommodate your concern about terminology when I asked you last week to provide the terms you wished to use to properly distinguish between "information" as we might perceive it in the necessary causes and effects of natural processes, versus "information" in the form of arbitrary strings of characters arranged for the purpose of storing and conveying meaning. I haven't seen a reply about that. It may help us to get past semantic technicalities if you'll answer it.

Thanks, and have a good day.

Dennis Murphy (not verified) | 07/16/09 | 13:00 PM

A computer code is no more sufficient and necessary to the production of an end state from an initial state than DNA. they both require a system and interpretation.

Not true.

A computer code (machine language) does not require interpretation (1). By definition, it is a specific action that is to be performed by the machine. The interpretation is made by a human that is looking at a sequence of codes to perform a function, but by itself, any given instruction in a computer is specifically changing the state of the system to produce a finite and fixed result. In addition, this specific action will always result in exactly the same state change. Moving the contents of storage into a register has nothing to do with the values involved, but only that some pattern of bits will be transferred into the register resulting in the register reflecting that same pattern of bits.

DNA does not require a system, although it can generate one. This is evidenced by activities like Polymerase Chain Reaction (PCR) which allows duplicates of DNA to be generated without any system or interpretation at all. Similarly when selected strands of DNA are introduced into cells they control general activities, but their state changes are not fixed. They will vary depending on the cell type and organism to which they were introduced (for example, HOX genes which control the type of development that will occur based on the anterior/posterior axis for placement of structures like leg, eye, or antenna).

(1) High level languages are not computer code, but simply languages that must be translated into computer code before they are executable.

Gerhard Adam | 07/15/09 | 23:54 PM

"A computer code (machine language) does not require interpretation (1). By definition, it is a specific action that is to be performed by the machine."

Referring to Lockerby's original post, "Note that hex, being a secondary symbol level is a code."
Computer programs are written by humans in a computer language, or code, such as hexadecimal, C++, etc. which must be "interpreted" by the hardware "system". The computer language or code is what is referred to when we refer to "computer code". The code is useless unless it is interpreted by a system which acts as instructed by the code.

"DNA does not require a system, although it can generate one. This is evidenced by activities like Polymerase Chain Reaction (PCR) which allows duplicates of DNA to be generated without any system or interpretation at all. "

If a strand of DNA is isolated from any and all systems, the information contained in the DNA will be useless. It will not by itself produce the living organism or anything else programmed into it. It requires a system.

While PCR is not the process of producing the organism the DNA is coded for, it is does require a system suitable for PCR containing primers, polymerase, and a means of heating and cooling. PCR is not the implementation of the information coded into the DNA, but is only a replication of information. It is like a computer programmer copying a segment of computer code from a program and replicating it.

Stephen Swires (not verified) | 07/16/09 | 10:47 AM

Computer programs are written by humans in a computer language, or code, such as hexadecimal, C++, etc. which must be "interpreted" by the hardware "system".

You need to understand something more about computer hardware before you start throwing in compiled languages and suggesting they are the same thing. Hardware systems no more interpret code than a light switch interprets your intentions when you flip the light on.

While PCR is not the process of producing the organism the DNA is coded for, it is does require a system suitable for PCR containing primers, polymerase, and a means of heating and cooling. PCR is not the implementation of the information coded into the DNA, but is only a replication of information.

True, but it was never suggested otherwise. However if you take your computer program and place it in memory alongside as many instructions as you like, the one thing you won't get is a replica.

If a strand of DNA is isolated from any and all systems, the information contained in the DNA will be useless. It will not by itself produce the living organism or anything else programmed into it. It requires a system.

It, at least, still has an existence, although your computer program wouldn't without a system. The point is that regardless of how different ways you say it, you cannot postulate that an intelligence was required to create DNA by using a simplistic example like a computer program to which biological organisms don't bear the least resemblance. It isn't even a close approximation.

Gerhard Adam | 07/16/09 | 16:07 PM

"You need to understand something more about computer hardware before you start throwing in compiled languages and suggesting they are the same thing. Hardware systems no more interpret code than a light switch interprets your intentions when you flip the light on."

Your insistence on disagreeing with anything and everything someone says certainly does not impress me with your intelligence or knowledge, if that is your intention.

The CPU is hardware. The accompanying chip set is hardware. The bios chip is hardware. The video processor is hardware. The sound processor is hardware. These pieces of hardware have been constructed with built in programs (or burned in programs as in the case of the bios and an EPROM) that interpret and process the information fed to them from the programs written in higher languages.

"True, but it was never suggested otherwise. However if you take your computer program and place it in memory alongside as many instructions as you like, the one thing you won't get is a replica."

You most certainly did suggest otherwise. Go back and read it. You said that PCR is an example of DNA not requiring a system, and I showed you that it does require a system. If a computer operator using the system of a computer and a text editor selects and copies a section of computer code he can replicate it as many times as he wishes or a programmer can write a program to cause a section of code to be replicated, just as a person working in a laboratory with a system designed for PCR can produce many copies of a DNA segment. PCR does naturally occur: the system is designed and implemented by human intelligence.

"It, at least, still has an existence, although your computer program wouldn't without a system."

While that is not true, what the heck is your point in making such a statement???

Your argumentative rabbit chasing style is pointless and endless. I see no point in communicating with you any further.

Stephen Swires (not verified) | 07/16/09 | 23:16 PM

I can print computer code (or DNA sequences, for that matter) on paper and make an infinite number of copies of it just by using a photocopier. No computing activity is being done. Neither are DNA operations being done just because you make copies of the molecule using your lab's PCR equipment.

Dennis Murphy (not verified) | 07/16/09 | 13:28 PM

That's simply ludicrous. You aren't making copies of anything except images of real world items. If I copy of a photo of a person, I'm not creating copies of the person.

Gerhard Adam | 07/16/09 | 16:09 PM

It is simple, and it's not ludicrous. The information exists in the sequence of numbers (or DNA base letters) on the page, not in the "real world" paper and ink. When I make a copy of it, I'm making a genuine, second set of the information that is exactly the same as my original. (Disclaimer for nitpickers -- it's possible that limitations of the copier may corrupt the copied information.) I can then take that copy and transfer its information back into computer memory or use it to make a DNA string. It's still the same information, it was genuinely copied, and it was then moved from one medium into another. *Information* is a different thing than the "real world item." Your casual dismissal of my example shows that you're not comprehending the issue or deliberately avoiding it.

Dennis Murphy (not verified) | 07/16/09 | 16:26 PM

When I make a copy of it, I'm making a genuine, second set of the information that is exactly the same as my original.

OK, good. Here's a copy of DNA snippet

CAGTTAGTAGCCTAG

Call me when you have something that replicates or grows.

Gerhard Adam | 07/16/09 | 16:52 PM

Lessee ... I claim that one can make a copy of a set of information (e.g., computer code or DNA sequence) and the informatin in the copy is exactly the same as the original. You reply with a challenge that I should take your DNA snippet and turn it into something that replicates or grows. Can you see the non-sequitor there?

You wrote: "CAGTTAGTAGCCTAG"

CAGTTAGTAGCCTAG
CAGTTAGTAGCCTAG
CAGTTAGTAGCCTAG

I cheated. I used a copying system to make three additional sets of your 30-bit string. That's all I said I could do. Is the information in my three sets the same as in your example? Could I use the information in any one of the sets to construct a real DNA snippet of the corresponding sequence?

You're the one who said he could replicate DNA "without any system ... at all."

You're missing the forest for the tree bark, Mr Adam. I really think I'm wasting my time trying to respond to your objections.

Dennis Murphy (not verified) | 07/17/09 | 07:22 AM

Look, it's pretty obvious that you don't understand the difference between copying symbols and actually making a copy. Similarly, there is a difference between requiring a system versus having a system (i.e. procedures).

All you've created as a mnemonic copy of what you think is a DNA string, you haven't copied anything except a symbol of a symbol.

Other than describing more procedures, what is the "system" required for DNA to "run"?

Gerhard Adam | 07/17/09 | 15:36 PM

"... what is the "system" required for DNA to "run"?"
>sigh<
You figger it out.

Dennis Murphy (not verified) | 07/17/09 | 15:50 PM

Yeah, I thought so.

Gerhard Adam | 07/17/09 | 15:56 PM

I'm surprised at the distractions and non-sequitors being offered as arguments on this topic. Our discussion is going nowhere if we have to discuss side-issues about predictability, or chemistry-vs-electronics, or the true definition of a 'code.' We know that intelligent agents are sufficient as a cause to produce information, and we know of no other causes for it. When we find complex information (1) somewhere, such as sequences of letters or DNA bases, the best explanation we have is that an intelligent source caused it. That is Mr Meyers' position in his book.

Based on my experience here, I conclude that one could learn quite a bit from Mr Meyers' book, regardless of how he uses the word "code."

Have a good day.

(1) In contrast, 'simple' information, such as short or repeating sequences, can easily resemble the results of some natural processes, e.g. crystals and polymers.

Dennis Murphy (not verified) | 07/16/09 | 13:34 PM

I think I need to write an article on how the electronics behind computer logic is used to prescribe the computer code. It is substantially different from the way DNA works.

In brief, computer code is unidimensional, shapeless: all possible outcomes of all stages of computation produce 100% predictable outputs. DNA code is 3-dimensional: the outcome of any stage or process depends on the physical 3-dimensional shape of the input. Micro-variations in the shapes lead to an amplification of error: at no stage is the outcome 100% predictable based on the chemical sequence, or "code".

At the mechanistic level, the voltages that represent logic 1 and 0 are separated by a wide gap wherein the value of the voltage is indeterminate. In DNA, if there is such a 'band gap' it exists at the quantum level and there is no mechanism in DNA transcription to detect such a gap and avoid it.

A comparison between the DNA "code" and computer code is a comparison between an abacus and a set of dice.

Patrick Lockerby | 07/17/09 | 16:59 PM

Definining code is a facinating problem. Dictionary definitions look so poor. Other definitions around the same subject are poor too. Consider "code", "data", "symbol", "representation" and of course "art / literature".
To define something effectively it, is usefull to describe what it is and what is does. So having failed in the first
part may I encourage you too explore the second part. What does code do and with what?
Language is severly restricted at this time in these matters. Science has failed so far to sort it out. I predict that
when these words are properly defined we will have a new insight into the human brain and the universe.

Tesom (not verified) | 07/31/09 | 10:31 AM

So basically, information is all there is in the universe.
DAVIES, Paul:
That's right. The universe is just a big information processor. Wheeler calls this "it from bit." Now if you take that view -- that the universe is a gigantic computer -- then it leads immediately to the conclusion that the resources of that computer are limited. The universe is finite. It's finite because the speed of light is finite. There's been a finite time since the big bang. So if we have a finite universe, we have a computer with finite resources, and hence, finite accuracy. So once you recognize that the universe is a gigantic computer, then you see that the laws of physics can't be infinitely precise and perfect. There must be a certain amount of wiggle room or sloppiness or ambiguity in those laws.

Anonymous (not verified) | 08/13/09 | 04:15 AM

More The Chatter Box articles

About Patrick Lockerby

Comments

Add a comment

Know Science And Want To Write?

What People Are Saying

Books By Writers Here

Not a Scientific Blogging member yet?

Register Now for a Free Scientificblogging.com Account

Already a Scientific Blogging member?

Sign In Now

More The Chatter Box articles

About Patrick Lockerby

Related Articles Here On Scientific Blogging

Comments

Add a comment

Know Science And Want To Write?

What People Are Saying

Books By Writers Here