Posted: Jun 14, 2012 8:42 am
The problem with his scenario above is that amylase production seems to match dietary habits and there is no reason to infer that duplication of the amylase gene is a harmful mutation. So the accident occurs in just those groups that can make use of it.
Here's the source of your confusion. The mutation doesn't occur in groups that make use of it, it is fixed in populations that do. I'll let you figure out what the difference is between a mutation occurring for a certain reason and a random mutation getting fixed in a population for a particular reason.
You essentially look at what mutations were fixed and go "Ooh, the mutation occurred to give us an advantage" but as your very own source points out, you would be well advised to read and assimilate this.
. If we consider all the mutations that led to these pivotal point in our evolution, human origins begin to look like a trail of unfeasible coincidences. But that is only because we do not see the harmful mutatios that were weeded out, points out John Hawks at the University of Wisconsin-Madison. "What we're left with is the ones that were advantageous." It is only from today's viewpoint that the mutations that give us our current physical form appear to be the "right" ones to have. "
I suggest you find some of Susu.exp's posts on the distribution of mutations and you will know exactly where your idea of organisms developing mutations to meet future needs goes awry. I will search for some of them later and will post them here.
You have given the example of the stromal environment in cancer as an example of top-down regulation. I disagree - a process that involves cross-talk between cells and feedback from said cross-talk is not top-down. It is bottom up with feedback.
Also, the role of stromal cells has been known for a long time, as have the disadvantages of monolayer studies in elucidating the process of carcinogenesis, since gene expression patterns are known to vary with spatial architecture, being regulated by the cell modifying transcriptional programs by sensing for external cues.
And that tumour stroma influences cancer cells has been known for a long time as well.