Code as an Analogy of DNA?


(John Mercer) #182

There are a couple of problems with that definition.

First, in the immune response, “memories” are WRITTEN in the genome by variation+selection, so “ROM” is completely inappropriate.

Secondly, the term “quaternary” is well-defined in biology in the context of protein structure, and you are just creating confusion by using it here.

(Dale Cutler) #183

You asked for a definition. I gave you one. You didn’t like it, even though it made sense. Sorry.

(John Mercer) #184

If we agree on a code to keep our communications secret, there is no need to store any data in any memory, so you are in fact using a highly-specialized definition of “code.” All computer instructions are code, but not all codes are computer instructions.

(Timothy Horton) #185

You gave me one you made up yourself. That tells me you know the accepted definition of code won’t apply to the twisted way you are trying to use if for DNA.

(Dale Cutler) #186

This has nothing to do with secrecy. It’s about data storage.

Yeah, you’re probably right about ROM. Call it EPROM instead.

(Dale Cutler) #187

You asked me for a definition but I didn’t parrot one back to you that you liked. Maybe your request was disingenuous.

(Timothy Horton) #188

It seems to be about you inventing your own definitions since that’s the only way you can make an argument. Less than impressive for sure.

(John Mercer) #189

Secrecy is not necessary, nor is data storage. I was merely giving you an example that did not fit your definition, therefore invalidating it.

Also, NO term with “Read Only” is appropriate for DNA sequence, as your immune system writes new data every time you get a cold.

(Timothy Horton) #190

Or maybe you don’t like being called on the dishonesty of making up your own definitions as your supporting “evidence”.

(Dale Cutler) #191

I’m amazed at all this flack. Trifonov uses the term. (The video is fun to watch, btw.)

(Dale Cutler) #192

Okay, call it flash memory, I don’t care. It is still a data storage system, whatever the definitions or analogies make you happy or upset.

(Chris Falter) #193

In Shannon information theory, a code is a mapping between information (bits) and the structures of some medium.

A biological cell stores the instructions for making proteins in DNA. There is a mapping between the information and the structures. Ergo, per Shannon, the mapping between instructions and medium (DNA) is properly referred to as a code. Biologists commonly refer to this as “the genetic code.”

The amount of freedom available for the mapping process is irrelevant to Shannon.

At the same time, Shannon deliberately excludes any inference to meaning or purpose. Thus anyone who would try to infer an intelligent codesmith from the genetic code profoundly misunderstands Shannon information theory.

I believe in a divine codesmith, but that is an act of faith, not an inference from information theory.

(John Mercer) #194

It is, but no data storage system is required to have a code.

(Retired Professor & Minister.) #195

That’s a fascinating article you posted about Harvard scientists using DNA for incredibly compact data storage. Great reading. (And I thank you for that link.) And it also illustrates how various levels of abstraction are added in such man-made coding systems.

This thread is continuing to amplify equivocation confusions over the word code. [I’m not implying anything deliberately misleading or ominous about that. I’m approaching this as a linguist who is fascinated by ambiguities in language.] The Harvard article reminds us that DNA in its natural contact is only a code analogy, while once DNA gets manipulated by the bioengineers for more “conventional” data storage, those DNA sequences become actual code in the computer science sense and not just as an analogy. (Meanwhile, I also concede that in an average American layperson’s understanding of the English word code, both senses can fit such a general term.)

I do think that Dr. Swamidass et al suggesting the word template as a better analogy has much merit.

(Dale Cutler) #196

The fully automated factory analogy does not require any abstractions in its data handling, as I mentioned earlier. It is only the outside observer that that needs labels for communication to others outside of the systems. And that is what language is about, abstract symbolic terms and communication, and ‘code’ is still the most obvious term. The fact that DNA storage can be used in more than one way rather exemplifies that. I’m fine with ‘template’, too, in living systems (is that all it is?).

(Retired Professor & Minister.) #197

I’m not sure what you are asking with “Is that all it is ?” Obviously, a label is never just “all it is” because a label is not the same thing as that to which it is associated.

To reiterate what I assume will be obvious to most readers (but I just want to make sure, considering that Peaceful Science has a diverse readership): Calling a biological phenomenon a template in no way reduces the “sense of wonder” people like me have in regarding DNA, codons, and amino acids as amazing examples of God’s creative works.

(John Mercer) #198

Correct. Therefore it is not a code. Template is much more accurate.

(Dale Cutler) #199

@Mercer: ‘Template’ is an abstraction as well, and in an automated factory, information is passed digitally (‘code’), as well as materials being manipulated and processed physically (which could include the use of templates), so both terms (both are abstract) can and do apply. The Trifonov video might influence you to accept the term ‘code’ a little more easily.

(Dale Cutler) #200

I am not a ‘good biologist’, so I don’t pretend to have a handle on the extent of all the known functionalities of different cellular components, including DNA. So that was an honest question and maybe somewhat rhetorical, ‘is that all it is’ was my wondering about other functions of base pair sequences. I guess I know that some are ‘section markers’, if you will, so might not they be considered ‘stop’ codes instead of templates? That is a question for you, too, John @Mercer.

(Chris Falter) #201

Per Shannon information theory, it is in fact a code, as I explained 10 posts ago.

The confusion arises from the fact that a lot of folks do not understand Shanmon theory, do not accept the definitions of Shannon theory, and/or draw analogies and philosophical conclusions from genetic code that lie far beyond the domain of Shannon theory.