I've posted this elsewhere, but it received no treatment whatsoever. So, I'm going to make this my fast breaking serious post in this wonderful, beautiful place. I'm hoping that our genetics boys arrive soon and tear me apart for my simplistic notions. Hercules Rockefeller, this means you... An idea that has occurred to me recently. The Human Genome Project. It stands. Feat accompli. Well... Mostly. 99% of the human genome, some 2.85 billion base pairs, is now known and is being busily scrutinized by hordes of geneticists. And yet. More remains to be done: 1.) 341 gaps remain and will remain until a new method is developed in order to process the long repeating stretches that encompass the gene-poor heterochromatin and the remaining euchromatin. 2.) Errors exist. 1 error in every 100,000 base-pairs. 3.) A new understanding of epigenetic influences on cellular development is poised to revolutionize our understanding. So, while the mapping of the genome is a huge step, it is only a step, and perhaps not the most vital step. 4.) An exact count of genes is still being worked out, as is the proteome. The latest estimates are 20 to 25 thousand genes coding some 90 thousand different proteins. (The mechanism how this works and it's implications for understanding of cellular development is a thread in itself.) I've discussed this issue with several scientists who actually work in this field (I am not one) and the number of genes is disputed. One practically took offense at the idea of a mere 25 thousand protein-coding genes while others accepted as a likely number. But, I digress. Here's what I've been thinking. In in vivo genetic transcription, the error ratio in an E. Coli bacteria is 1 in a million base pairs. With the built in error correction this is reduced to 1 in a billion base pairs. With this understanding, the 1 in 100,000 of the Human Genome Project becomes laughable at best. But it's the best that science has to offer so far. Now. Think about a lifeform that possesses a polymerase that is so error prone as 1 in 100,000 base pairs. What would happen with this creature? Could it survive? Could it be considered stable? Would we call it a mutant? How many species would diverge from a single genetic source? And in what amount of time? I think it's safe to say that this creature would undoubtably produce a large number of inviable young, but possibly a number of wildly disparate viable young. We could say that this creature is evolution on a fast track. Given a world full of such creatures and we might well see something on par with the Cambrian Explosion. Evolution on a vast scale. So. If a living being with such a state of affairs can be said to be an evolutionary bullet. What of the Human Genome Project? Isn't it something like evolution? And what is it evolving towards? We can think of the ongoing efforts of the project to be something on par with the limit of calculus. Coming closer and closer to some ideal but never reaching. But, what is that ideal? There is no such thing as the 'Human Genome". This is an artificial concept with no true meaning in the natural world. We have divided organisms up by means of phenotypes and other artificial categories in order to explain certain observed patterns in genetic inheritence. But, this is not to say that this is a truly accurate way of describing the reality. Evolution occurs within species. And as such, the 'human genome' contains a vast assortment of variation. There is no 'ideal' to which we can limit towards. To attempt to do so is foolhardy at best. We can attempt to limit towards a perfect map of any one human, but that is all we can do and still maintain any sense of 'preciseness'. To work towards an ideal is to work towards an abstract generalization. Also, consider the way in which the Human Genome Project is undertaken. It is not a precise rendition of one human's genetic chain. It is an amalgam of many. Cloned fragments of a variety of genetic donors being pieced together like a jigsaw puzzle. What a novel method of integrating disparate genetic material into a single organism. It makes me think of viruses. This creature that is evolving from the Human Genome Project is evolving like a virus not a human. Now. I am not saying that it will turn into some strange creature that doesn't resemble it's host progenitor within a few generations. Why not? Because of the watchful eye of the researchers always fudging the results to push them back onto what they think is the right track. They are the ones purposefully seeking patterns and the condition they are looking for is man. So the errors that push away from man will be removed when found. But. It could, if left to its own devices, evolve into something quite different than man. It could become something entirely alien to our experience. It could turn into several somethings. If it were automated, if a blind eye were turned towards it, what wonders it might create. And what horrors. This was my thought. A simple one, I think. But interesting. The Human Genome Project is not really delineating the 'human genome' but is rather evolving towards it. Poorly (compared to natural processes.) And even if a method of perfect transcription should be found, then even then there is no ideal to move towards. No 'genome' in physical fact. So, perhaps, it is best to maintain some error. It works for nature. It should work for man.
I think that what you’re proposing could make a great experiment on a vast scale – after the human genome, we investigate ultra-fast evolution. With such a vast mutation rate (I would argue that because the base pairs are being screwed up that they’re mutants) we’d see extraordinary divergence. However, we’d also expect quite a bit of death, as most mutations tend to result in that type of thing. It depends on which ‘letter’ gets mutated. Codons are redundant (presumably) to protect against mutation, ergo some mutation won’t really matter. However, ‘failure mutation’ aside, we’d see a large level of variation. This in itself is only a step in evolution, but interesting none the less. So: - in an experiment with a control environment, ‘infect’ a simple organism with the error prone polymerase. - Weed out deaths - Collect ensuing data on variation - Repeat with same group
But why do you want to correlate the transcriptional errors in E. coli to the errors in a sequencing experiment? You mean an RNA polymerase? (you were after all talking about transcription a while back). An extremely error-prone RNA polymerase likely leads to what's called an "error catastrophe" i.e. where the replicational, transcriptional and translational machinery is itself all messed up. I must have misunderstood you. Transcriptional errors are not inherited. So do you mean DNA polymerase errors? Yes they are inherited. Hmm..that's more interesting. Let's see. Well, remember that for evolution to occur, two conditions have to be satisfied. There must be genetic variation, and there must also be selective pressure (I hate that expression, but we all use it). And the the two sort of walk along-side each other - there's no point in increasing one if the other doesn't increase concommitantly I really don't see the connection between a laboratory experiment and evolution. Have I lost the plot? This is of course perfectly truse, until we sequence the genome of every single creature, living, dead and unborn!. Ah. You got there before me. Although I would say "precision"!!
