is DNA stupid?

says...


I have a theory (indulge me, I've had a half bottle of Pinot Grigio, and the theory is only half-formed) that e7 is an experimental AI reponder like Eliza. Whereas Eliza when stuck for anything to say would ask "Tell me about your mother", e7 just slips in staock phrases like "But Chomsky states..."
And, just like Eliza, e7 only appears intelligent, but never really learns anything.
Just my tuppence.
PeterS Remove my pants to reply.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

ELIZA just called. It demands an apology...
Perhaps this is really a copy of PARRY. See http://www.stanford.edu/group/SHR/4-2/text/dialogues.html
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Wrong again. The debate is still at the first point. You claimed that the human genome has 20,000 bits of information. I provided proof that it has billions of bits of information. You aren't allowed to post obvious howlers and then to move on without dealing with your error.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
e7 wrote:

------------------------- I notice that whenever you are opposed you assume a paranoid resistance to such opposition and respond by solidifying your belief in trivial pieces of misquoted trash as a kind of gospel to protect you from "losing your faith". You, sir, are terrified that some devious cabal will talk you into something you don't really fathom, and that you yourself are so fragile that you might actually become deceived by "them" and believe in what is actually false, you're frightened of everyone who doesn't believe whatever you say absolutely uncritically!!

---------------------------------- Go review ALife programming at the Santa Fe Institute.

--------------------------------------------- What I was trying to tell you is that IS NOT the way it was done!
It doesn't exist, it doesn't have to, evolution didn't do it that way, it learned by random variation and attrition of organisms which had less survivable properties of behavior. So you see, it didn't store any info on personality, only on the machine in which personality spontaneously grows, and it leaves it to that process to perfect the response of the organism. Now one high-level way of responding to an environment is to grow a software device that helps the organism, and this software device is called self-modeling/memory/awareness. That is what we call "US" our mind and existence in this story we tell ourselves called Life In The World.

-------------- Yes, I know that. I wish YOU REALLY did!
-Steve
--
-Steve Walz snipped-for-privacy@armory.com ftp://ftp.armory.com/pub/user/rstevew
Electronics Site!! 1000's of Files and Dirs!! With Schematics Galore!!
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Your first error is equating a gene with a bit. A HUGE error. You see, a gene as mentioned in the article is a string of DNA that encodes a single protein or enzyme. Each "letter" in the gene is three codons in length, and each codon is 2 bits of storage, representing one of four bases. So each genetic "letter" is 6 bits- 64 possible combinations. Most proteins have dozens or hundreds of amino acids in them, and each represents 6 bits of data, so each gene represents anywhere from a few dozens of bits to hundreds of bits. In other words, you are off by perhaps two orders of magnitude. Now, this is very important- some of the genes are control genes that treat other genes like subroutines in a program, so the basic complexity (how many bits) is way off base from the emergent complexity. You would never specify building a house by pointing out every nail hole dimension and location- and you certainly would not hand over a full sized, fully "nail-populated" and "bathroom tile-populated" model. Instead, you use a shorthand notation, such as "this area will be tiled with tiles this size". This eradicates the thousands of bits spent specifying where each and every dot of mastic or paint goes. The same is true of organisms- they are not "fully spelled out"- that would be patently and utterly ridiculous. Instead, they are defined by locations, volumes, and generalities, and each of the materials is written in the same manner as a subroutine- "use this tissue type here", and "coat these with proteins so-and-so"- and this leads to the emergent structure. Our genes show how to make the protein equivalents of hammers, saws, and boards, and then how they are used in a general way. Ever play Conway's "Game of Life"? It is completely impossible to predict the outcome without actually running the game. And so it is with living things. And, in Conway's game, structures of greater complexity than the starting conditions can (and do) emerge all the time. So it is with real living things. So, to state that 20,000 genes cannot encode a brain is not only wrong, but provably so. And, we understand just why it is wrong. Saying that 20,000 bits cannot encode your brain is true- but has nothing to do with the issue. Now, here is a perfect example you can try for yourself. Download POV-ray. Write a short program to generate a realistic graphic image. Now, would you rather send a 3D solid model by modem, rendered, made solid, and then broken down and analyzed at its real scale? Or, would you rather send a few dozen lines of code that produce the realistic solid model through description? You can choose to send many hundreds of gigabytes of "finished model representation" or you can send a kilobyte or so of code. That is the difference in actual genetic code and your understanding of it. The genetic code is descriptive, not absolute.
Cheers!
Sir Charles W. Shults III, K. B. B. Xenotech Research 321-206-1840
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Sir Charles W. Shults III wrote:

-------------- No, Charles, a codon is three base-pairs, and a base pair has four possible values, GC CG AT and TA. Thus it equates to two bits. Thus a codon is 6 bits with 2^6d possible values.
-Steve
--
-Steve Walz snipped-for-privacy@armory.com ftp://ftp.armory.com/pub/user/rstevew
Electronics Site!! 1000's of Files and Dirs!! With Schematics Galore!!
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

<snip>
of
Uhh.. that's exactly what I said. Read it again.
Cheers!
Sir Charles W. Shults III, K. B. B. Xenotech Research 321-206-1840
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
MIME-Version: 1.0 Content-Type: Text/Plain; charset=US-ASCII
http://www.math.gatech.edu/~lacey/courses/3012/genome.html
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

<snip>
Ok, genes are not made of single DNA sites or locations, and it can take thousands or more DNA sequences to have a gene. Saying DNA has 20k genes involved in brain functions is not close to saying there are only 20k DNA sites that comprise those genes. It's millions, if not billions.
Also, just for clarification, a given bit has 2 possible values 0 or 1. DNA has 4 per site, depending on which of the four amino acids is present.
Total information storable by 20k bits: 1.99 e6020 Total information storable by 20k DNA "sites": 3.96 e12040, or 1.99e6020 times greater than your original argument. And that's not taking into account that 20k genes is millions, if not billions of times greater than 20k DNA "sites".
Dave
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
From: http://pga.lbl.gov/Workshop/May2003/lectures/Babbitt.pdf
A lecture by Patricia Babbitt University of California, San Francisco April 24, 2003
"Using a standard measure for overall amino acid frequencies gives the information content of a random protein sequence as 4.19 bits/residue."
So there goes the 20,000 bits of information idea.
I still don't think 95% of the information we have in encoded in DNA though. (It's sort of my job)
--
- Alan Kilian <alank(at)timelogic.com>
Director of Bioinformatics, TimeLogic Corporation 763-449-7622
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
snipped-for-privacy@raceme.UUCP (Alan Kilian) wrote in message

4.19 bits/residue? Where did that number come from? DNA sequences would be similar to a base-4 numeric system as opposed to binary (base-2), but how can you have a base-4.19?
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

The paper explains it far better than I ever could, but here goes....
To tell how manmy bits are required to store a symbol take log-base-2(number of possible symbols)
So, for DNA, there are 4 possible symbols. log_2(4) = 2 (bits)
There are 20 amino-acid symbols. log_2(20) = (about) 4.3 (bits)
Since the amino-acid symbols don't occur equally in real proteins, Patricia and others have estimated that there are about 4.2 bits of information in each protein symbol.
Is that an OK explanation?
So, I'm going to go with 30,000 genes and a total guess of 400 symbols-per-gene conservatively. (Actually EXTREMELY conservatively) so I get a pure guess of 30,000 * 400 * 4.2 = Fifty megabits of "information" encoded in JUST THE PROTEINS of the Human genome.
There are regulatory-regions, structural DNA, promoter regions, multiple splice-sites, multiple-exon proteins, forward and reverse-strand information and many many other places for "information" in the human genome.
Also, I would like to re iterate that I personally do not believe that 95% of a human's knowledge is pre-programmed into our DNA.
I have too many friends doing infant-memory studies to believe that.
There are only 20 amino-acid symbols.
--
- Alan Kilian <alank(at)timelogic.com>
Director of Bioinformatics, TimeLogic Corporation 763-449-7622
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
snipped-for-privacy@raceme.UUCP (Alan Kilian) wrote in message

Ah, we're quantifying two different things then. I was soley saying that the actual ammount of info you could encode on a given DNA strand (regardless of species) was a base-4 numeric system. You were quantifying "informational storage" via amino acids at a higher level. We just had different answers for different viewpoints.
Dave
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
snipped-for-privacy@raceme.UUCP (Alan Kilian) wrote in message

4.2 (bits/aminoacid)*400(aminoacids/protein)*1(protein/gene)*20,000(genes/brain) = 33.6 MegaBits/brain = 4.2 MegaBytes/brain = 155 brains/CD
And I think Hans Moravec or someone claimed that the typical human adult's life memories at any one time could be fit on a CD.
Einstein (or any typical smart Jew) might take 2 or 3 CD's.
It is possible to get fractional bits per amino acid when you take into account the fact that each amino acid has a different expectation. (The more rare amino acids are conveying more information I think.)
One of my professors at UCLA claimed that the most efficient number of bits/aminoacid would be 2.71828 (e).
Or, if I've got it all confused again, then: Someone? claims some? argument? sounding similar to what I've written above is true.
- karen715j
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Karen J,
I think most of the problem is the statement that knowledge is encoded in our genome.
It is not.
If there are 20,000 genes active in a human brain, that does not mean that there are only 20,000 "Things" in the brain, and that is what's getting you confused I think.
Try this. Think of just a small number of genes: GeneTransistor; GeneWire; GeneSwitch; GeneLightbulb.
Those are 4 genes.
I think you can imagine several ways to build electronic circuits using just these 4 genes right?
You could build a 2-digit adding machine.
You could build a 4-digit multiplying machine.
You could build a 1-cellthreshold-triggered neuron simulator.
Using a small number of extra genes, you could create a simple wiring diagram showing how to connect a hundred billion of your 1-cell neurons
You wouldn't need a hundred billion bits of information to construct a mesh-layout array of neurons.
Then, it would be possible to imagine a learning process whereby inputs are presented to the array, outputs are generated, and some feedback is created that can either strengthen or weaken the connections between these neurons.
That is a very simplified description of how a brain is constructed, and how it learns using only several genes which can be easily encoded in a human genome.
The problem arises from the mistaken assumption that almost all a human's knowledge is pre-wired into the brain at build-time. That would indeed require more information than is contained in the Human genome. Incidentally, that's a pretty big demonstration that human knowledge is NOT pre-wired in in my opinion.
--
- Alan Kilian <alank(at)timelogic.com>
Director of Bioinformatics, TimeLogic Corporation 763-449-7622
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
snipped-for-privacy@raceme.UUCP (Alan Kilian) wrote in message

I think most of the problem is that some posters are writing about personality, some about genetic inclination to personality type, some about general intelligence, some about knowledge, and some are writing one thing and meaning another and some are reading one thing and thinking another.

I disagree. I was confused about other things; not that.

A very simplified, very excellent description.

I don't think that e7 or Noam Chomsky are saying that. Mainly they are saying that the ability to learn languages is prewired SPECIFICALLY and not simply part of general intelligence.
And they are saying a lot of other things are prewired also. I agree with them on genetic inclination towards anger, laughter, depression, elation and on and on.
But when Chomsky claims that the ability to learn numbers is prewired I part company. I suspect that that ability is better described as "genetic inclination to enjoy dealing with the abstract", or something. (Jews, for instance.) Most normal people prefer to deal with race cars, football and bingo.

I believe that some knowledge IS prewired. For instance the knowledge that blue, green and red are all different from each other. Or at least it is MUCH easier for them to pick up the concept than a totally color- blind person. But this is probably one of those stupid semantic quibbles that should be left to the philosophers who enjoy the abstract even more than I do.
At any rate, after reading Hans Moravec about three years ago, I concluded that my cybersoul could be stored on approximately two CD's: one for my genetics, and one for the current state to my memories.
That's all I really care about. Because it makes it very likely that I ACTUALLY AM a cybersoul.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

You seem to inject the word "Jews" into posts where the context seems to have nothing to do with Jews. Most people (Aardvark owners, for instance) don't do that. Why do you do that?
(Note: I am not enquiring about your opinion of Jews or about anything other than injecting unreleted issues into posts. Or Aardvarks. Talking about Aardvarks is OK too.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Guy Macon <http://www.guymacon.com wrote:

Guy!
I was going to mention this, but you beat me to it, and did a wonderful job of it. Thanks.
--
- Alan Kilian <alank(at)timelogic.com>
Director of Bioinformatics, TimeLogic Corporation 763-449-7622
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Guy Macon <http://www.guymacon.com wrote in message says...

The context "seems" related to Jews to me. For instance, I believe that genetics probably affects the tendency we humans have "to enjoy dealing with the abstract". The smart people who I have known who avoided abstract questions, tended to be non-Jews. For thousands of years, Jews have sorted themselves out according to whether or not they could deal with complex religious issues. So now Jews tend to love that sort of thing. IMO.
Also, I suppose that since Jews and Zionism are two of the most interesting and important topics in my politics, then examples of (whatever) involving Jews quickly come to my mind.

The main issue of the thread was (is there any pre-wired information in the human brain that most of us would probably normally expect to be learned info rather than pre-wired?).
So, apparently you think that "enjoy[ing] dealing with abstract issues" is either: 1 An environmentally acquired taste, or 2 A simple result of being smart
I claim that it's genetic because some smart people I have known seem to have not much interest in "abstract issues". They would rather fix the lawnmower, or repave the driveway, or something.... go camping maybe.
You would probably say: "That's an environmentally acquired taste." I would say: "What the hell makes you think that, you pompous ass?" You would say: "Anyone with any common sense, would take that as their opening position unless some good evidence could be given to refute the common sense position." I would say: "It's been my experience. I don't make a scientific study of every single interesting question that comes up. I don't have that much time to waste. The singularity is coming." You would say "What's the singularity?" I would say " Search Google, vernor vinge' and hans moravec'. I'm busy. I have to go now."
- karen715j

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

This is amazing. You put words in my mouth setting up a strawman argument (just like Aardvark owners do) when you have no idea (never having asked) of my actual position, then you call me a name, then you asume that I am unaware of the sigularity theory (Arrdvarks are as a rule unaware of the sigularity theory, BTW) all in one post. Now THAT'S entertainment!
Let me know if you ever wish to discuss my actual opinions rather than making up things and putting them in my mouth.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.