---Pop, here are my thoughts and reactions to that 'Big Numbers' computation
Andy
---Problems with this computation: 1) It is massivly over simplified, rendering it inefficient. IF DNA did specify exactly the length and spacial direction of all the dendrites in the human brain, then the Base Pair chain could be as long as calculated, also since DNA is inherited, this would mean that children would show identical dendrite length and spacial orientation (or simple combination) of their parents dendrite lenthn and spacial orientation. This is clearly NOT seem. 2) DNA sequences provide information for nucleotide sequences, which in turn produce protien and other molecules which then determine dendrite growth. The way these molecules react with each other to produce dentrite characteristics is determined by the molecules physical and chemical properties. These molecules have, because of their nucleotide chain sequences, unique shapes sizes and electron cloud density & distribution. The environment affects the way these molecules are oriented in relation to each other (spacial distance and orientation) this inturn affects how they react with each other. So two sets of identical protiens, produced from a the exact same piece of DNA can react differenly because the environment may orient them in different ways to each other. So DNA does not need to carry all the information for all the dendrites. Just carrying th information for molecules that make a dendrite is enough, the microenvironment around the chemicals will cause different dendrites to form from the same information.
To have all the information to fully predict how all the dendrites will form, we need to know every quantum mechanical detail of every molecule involved and the other environmental details: light, gravity, ect. Considering that we can't know the exact position, velocity, and spin of the electrons around a single hydrogen atom, there is no way we can gain all the information needed to predict exact biological growth. I suppose that if we could find the mechanics of the most sub-atomic particles, such that we could not break it down any further, and included every detectable environmental variable (on this sub-atomic level) we could build a computational model that would exactly predict everything from electron spin to the exact formation of biological forms in a closed system. Since the only truly closed system consists of everything, we would have to keep track of ever sub-atomic particle. To consider this even theorectically possible my a species that has trouble keeping track if it socks is thinking way to highly of the human race.
---> > "If you want to specify the connections of the human brain you need, a > 10 > billion numbers one for each cell, and then 4 numbers for each dendrite. > One > number specifies the length of the dendrite and three numbers specify > the > direction it is supposed to grow. We will assume that each neuron has > 2000 > synapses and each number requires only 11 bits of information. (this > actually is favorable to the design crowd. This eleven bits is only > enough > to represent numbers between 0 an 2048) The quantity of numbers > required to > specify the wiring diagram for the brain is > > 10 x 10^9 x 2000 x 3 = 6 x 10^13 numbers. > > I am going to rework the DNA code to encode the requisite information. > We > need to have numbers between 1 and 2000 so the triplet code won't work > because it will only represent numbers between 1 and 64. A six > nucleotide > code is required. As in the triplet code there will be some positions > not > needed. Now to represent the wiring diagram for the brain requires a > string > of DNA which is > > 6 x 6 x 10^13= 3.6 x 10^14 nucleotides long Just for the dendrites. > > The total human genome is only 3.5 x 10^9 nucleotides long. Obviously, > there is not enough information in the entire genome to encode for our > human > brain. Thus, since I believe that God did design us, the design must > lie > somewhere below this level. > > > ** > The fourth number comes from this: there are 10^10 numbers needed to > specify the individual neuron from which the dendrite grows. To > represent a > number as large as 10 billion in base four (which is what DNA > is)requires a > 17 digit number. thus to ennumerate each neuron with a unique number > requires an additional 170 billion dna locations. I hope this is correct > As > with last night, I did it in a hurry. Corrections will be appreciated. >
-- Brian W. Neuschwander