Can natural selection create: Part 4: Waiting for other mutations: Lots of colons!
BQ: We are considering the question, "Can natural selection CREATE even a single useful gene?" Remember, it's quite easy to modify existing genetic code, and your body does all the time. That's why we get cancer. It's why we are all, in fact, unique mutants. But can natural selection CREATE novel information?
It took us 12,000,000 years to stabilize our first desired mutation, and we saw that such presented a large problem with the evolutionary timeline. But keep in mind, it took us that long to encode a SINGLE useful nucleotide. Genes are at a minimum 1000 nucleotides in length, and that's about 50 times too generous, as we're ignoring regulating elements and introns.
However, to give ourselves the best chance of creating new, useful information, we're setting up this advantage. We'll call it a nice, linear, piece-by-piece process. Our first change took 12,000,000 years, and now we need to do it at least 1,000 more times, and it has to be useful. How long will that take? Well, multiply 12,000,000 times 1000 more times and you get, 12,000,000,000 years. That's a little bit under the time that the Big Bang took place, and we just made one single, useful gene, in a world with no harmful mutations and perfect selection.
Tomorrow, we need to look at even MORE factors that we're leaving out. Also, remember, this is making a new, useful gene only 1000 nucleotides long. Our genome is composed of 6,000,000,000. How long would it take to make such a genome if it takes so long just to create one new gene? And, from a while ago, we found out that creating the first "life" exceeds the probabilistic resources of the entire universe for its stated "life"span. From that we saw that M-theory was a fallback, because if we have multiple (in this case, infinite) universes, then such an infinitely small (by definition, impossible) chance of it happening would have to take place. Oh, and in one universe you, dear reader, are the Pope, and have a miniature llama as a pet.
(F. Hoyle, Mathematics of Evolution. M. Lynch, J. Conery, R. Burger, Mutation accumulation and the extinction of small populations. J. Neel, The rate with which spontaneous mutation alters the electrophoretic mobility of polypeptides. J. Adell, J. Dopazo, Monte Carlo simulation in phylogenies: An application to test the constancy of evolutionary rates.)
(PN240)
Can natural selection create: Part 5: Surprisingly slow.
BQ: We are continuing to answer the question, "Can natural selection create even a single useful gene?" Yesterday we "made" our first cool, new nucleotide. It took us just this side of forever to do so. What we didn't mention is that the human genome occurs in large "chunks" of between 20K-40K nucleotides within which we have no recombination. This means that the nucleotides are not moving around. Essentially, this reduces the probability that we can easily create the gene, and adds even more billions of years into the calculation. (The nucleotide mutations have to occur in such a way that they adjoin each other.)
But now we have our mutant nucleotide and are trying to get it to "survive" reproduction and to be passed on; since we have 20K possible nucleotide alternatives within our population of 10K, we have a tough goal. We want to go from two copies of the mutation in the population to four, and then more from there. If the mutation managed to be reproduced at 10% per generation (this is an awesome mutation that turns you into a supermodel and gives you just the best qualities ever), it would take us ~105 generations (2100 years) to increase from 1 to 20K copies. (1.1^105=20K).
The problem is these advantages are unrealistic. According to Patterson in "Evolution," a recessive beneficial mutation that would increase fitness by 1% would take 100,000 generations to fix. Haldane in, "The cost of natural selection," estimated that the most nominal time for fixation of a beneficial new mutation to take place is at least 300 generations, or >6000 years. This is called, "Haldane's dilemma," as it is so slow that it means there is effectively no selection taking place.
But if we can create one new nucleotide in 6000 years, can't we create a new gene in 6 million, the time since we supposedly evolved from apes? No. Haldane's dilemma concerns unlinked, non-adjacent mutations. For 1000 specific, adjacent mutations, it would take well over 6 billion years; furthermore, selecting for one nucleotide reduces our ability to select for others, which is called "selection interference."
We can conclude from all this that we cannot create even one new gene within our evolutionary timeframe, even without harmful mutations to boot.
Coming up, we'll look at the implications of endless fitness valleys, deleterious mutations, and poly-constrained mutations, all in our attempt to make evolution act the way "it's supposed to." We have about two or three questions left on this, and they should be easier to understand.
(PN241)
Can natural selection create: Part 6: Inferiority and fitness.
BQ: Continuing to see if natural selection can create a single useful gene, we now need to look at the conclusion of some known facts that we've left out. We know that man and ape differ by approximately 150,000,000 nucleotides. As we've seen, as a percentage, that's not too much, but as a genetic change in raw nucleotide numbers, that's massive. These changes account for at least 40,000,000 mutations.
If we assume that man came from chimp-like creatures, there must have been 20,000,000 nucleotide fixations within the human lineage (40,000,000/2), but we found out yesterday that natural selection could maximally have accounted for ~1000 of these in 6 million years, and they would have been non-sequential. What does this mean? It means that the rest would have been fixed by random drift, not a natural "enhancement" by selection, which would make them...you guessed it, nearly-neutral but HARMFUL mutations. We'd have on the best of days 1000 "good" mutations and millions of harmful ones, which would mean that man DEVOLVED from chimps! In fact, we'd be statistically dead!
Before we move on to poly-constrained DNA tomorrow, we need to briefly consider fitness valleys. Evolutionists can show that the creation of a new gene takes a great deal of experimentation. When the gene is being developed, there is a period of time when the species' fitness declines. This is called a "fitness valley." A partially-completed gene is not neutral, and it's not helpful...it's harmful. The species, then, must be harmed in order to be enhanced. We can imagine a situation where a species could survive fitness valleys if they were rare and short-lived, but continuous evolutionarily-positive innovation of novel genes would mean continuous, harmful fitness valleys, or one longer valley with a negative trajectory. Indefinitely number, indefinitely long fitness valleys ultimately would destroy all species.
(PN242)
Can natural selection create: Part 7: Polyfunctional DNA (See Figure 5—Polyfunctional Complexity)
Continuing to see if natural selection can create a single useful gene, we now need to look at the conclusion of some known facts that we've left out. We know that man and ape differ by approximately 150,000,000 nucleotides. As we've seen, as a percentage, that's not too much, but as a genetic change in raw nucleotide numbers, that's massive. These changes account for at least 40,000,000 mutations.
If we assume that man came from chimp-like creatures, there must have been 20,000,000 nucleotide fixations within the human lineage (40,000,000/2), but we found out yesterday that natural selection could maximally have accounted for ~1000 of these in 6 million years, and they would have been non-sequential. What does this mean? It means that the rest would have been fixed by random drift, not a natural "enhancement" by selection, which would make them...you guessed it, nearly-neutral but HARMFUL mutations. We'd have on the best of days 1000 "good" mutations and millions of harmful ones, which would mean that man DEVOLVED from chimps! In fact, we'd be statistically dead!
Before we move on to poly-constrained DNA tomorrow, we need to briefly consider fitness valleys. Evolutionists can show that the creation of a new gene takes a great deal of experimentation. When the gene is being developed, there is a period of time when the species' fitness declines. This is called a "fitness valley." A partially-completed gene is not neutral, and it's not helpful...it's harmful. The species, then, must be harmed in order to be enhanced. We can imagine a situation where a species could survive fitness valleys if they were rare and short-lived, but continuous evolutionarily-positive innovation of novel genes would mean continuous, harmful fitness valleys, or one longer valley with a negative trajectory. Indefinitely number, indefinitely long fitness valleys ultimately would destroy all species.
(PN243)
Figure 5—Polyfunctional Complexity