Recombination and function:
I have recently finished one of my headlong campaigns in cyberland, and as has often been the case I have learned something. But I am reminded (Kevin Brick et al Genetic Recombination is directed away from functional genomic elements in mice NATURE vol. 485 no. 7400 May 31, 2012 page 642) of something I learned years ago but could find no use for. And of course I didn’t keep careful, accessible notes, so I must rely on memory or spend a long and tedious time repeating the work. Welcome to the shaky world of my memory.
Starting out with my genetic simulation I had an idea of where I needed to go. I had to simulate real life. And I knew that I would have to follow the genetic laws of Mendel because they are beyond question even if they are not the whole story. But how many chromosomes should I have?
Nature provides humans with dozens of chromosomes but not thousands. That makes intuitive sense. Were there but one pair of chromosomes they would be fragile. I once dropped a two by four board, tossed it actually, and to my surprise it broke. The thing was that this two by four was seventeen feet long. Had it been three feet long I doubt my best efforts would have broken it. Big is vulnerable. On the other hand if there were thousands of chromosomes then the machinery for sorting them out during meiosis and mitosis would be bulky in aggregate and quite unlikely to function flawlessly. Hence there is a compromise. So how many to have?
In those halcyon days (actually they were tormented, hag ridden, demon infested days, but no matter) I decided in effect to make the number variable. There is a process we used to call “crossing over” and now more commonly call “recombination.” Two chromosomes during meiosis get together and twist about each other. Occasionally there is a break and the strands recombine at that site so AAAAAA and BBBBBB chromosomes produce AAABBB and BBBAAA. In effect this means the two ends are acting as if they were different chromosomes. The chance of a recombination event occurring between a pair of adjacent bases is about one in a million, but since the genome is very large that means a lot of such events each generation, three thousand the last time I checked.
In the process of ever seeking a better genetic structure, having multiple chromosomes is an advantage. Genes can get together in many combinations. If a good combination turns up, that can be selected for during the lifespan of the organism.
So even though having thousands of chromosomes seems awkward, nature has given us a similar effect by allowing crossing over. So in my model I chose to have a single pair of chromosomes but chose to have a selectable probability of recombination events.
I found that I could eek the effect I sought from the program, but when I twiddled with the crossing over rate I found that no crossing over at all was the best approach. Accordingly I suspect that the genes that, together with their methylation or other epigenetic process, that produce the effect are all crowded together in one site or maybe a few sites. At all events they are not likely to be diffusely distributed throughout the genome.
It turns out that nature was there way ahead of me. The advantage of recombining genes is not enhanced if the event happens within the gene. Oh there may be a rare event where one end of one gene combines with the other end of another gene. But that has to be uncommon.
Do you remember the old Studillac? It was a Studebaker known for its sturdy frame – the company originally made heavy wagons – and a Cadillac engine known for its generous size. The machine filled one notable automotive niche. It went like blue blazes. A similar thing happened when somebody tucked a Ford Cobra engine into a British AC sports car. That combination became known as the Shelby. I saw one a few hours ago. I have never seen a Studillac to know it. (What I offer about cars is received rumor. I am no expert in the field, only being interested in the principle.)
But by and large, most of the time, you are best off choosing a model of car, not putting together bits of a number of models.
The article says that in normal wild type mice there is a mechanism that directs recombination events away from the genes, their promoters and other functional elements of the genome. I cannot say I am surprised. Nor can I say, “I told you,” but at least I feel like I am in good company with nature.
There have been 66,780 visitors so far.