The article illustrates an important general concept in biology: all systems are dynamic. That is, no structure in a biological system can be modeled accurately as a static thing. In order to really understand something in biology, it is necessary to appreciate the fact that everything changes all the time. In introductory biology, we are taught that DNA is made up of a string of nucleotides, and that the sequence of the nucleotides is what carries the information needed to make proteins. That is only vaguely correct. It encourages one to think of DNA as being analogous to the printed letters on a page of text: static, unchanging. We are taught that mutations occur, but they are thought of as exceptional, accidental events.I'd venture to extend this principle to all systems, be they sociological, mechanical, biological, economic or political. But that may be seeing what I want to see, not what's actually there.
Now, we learn that mutations are not always accidental. In fact, some cells have a mechanism that exists for the specific purpose of causing mutations. No only that, but when the system is activated, it produces mutations by the millions. The idea that mutations in such large numbers might actually be beneficial seems, at first glance, to be completely counterintuitive. That illustrates a second general principle: intuition is a tricky thing.
In any event, anyone interested in Neural Networks and Genetic Algorithms had better take notice of this new datum. Biological systems still have an awful lot to teach us. Although we may not have the foggiest idea why something has been successful in the bloody world of fang, claw, tentacle and flagella, the fact that it has means that it bears watching.
No comments:
Post a Comment