Science!: Genetics and Robotics
Two stories that particularly caught my eye as I catch up on reading science magazines:
-From Technology Review: Geneticists are finding some surprising behavior in the genome. Apparently it's possible for living things to pass on genetic... effects, without actually passing down the seemingly relevant genes. For instance, female mice inherited some features thought to be controlled by the Y chromosome of their fathers, despite not even having a Y themselves. The explanation seems to involve even more complex control systems hidden in the supposed "junk" portions of the DNA than anyone previously suspected, with a large role played by changes in gene expression that can be inherited separately from actual genes. I know enough about this topic to speculate about mRNA, methlyation of genes, and stuff like that, but it's still pretty confusing. Another article in the same issue says, "Geez, we didn't say the Human Genome Project would instantly cure everything, so get off our case about it!"
-From Scientific American: "Rise of the Robo Scientists". Biologists are experimenting with robotic systems that can themselves do experiments, putting yeast samples into petri dishes and making simple observations. What's interesting is that the the robots are doing some of the work of proposing the tests. They can search their data to propose that gene X is involved in producing protein Y, and go from that idea to "it can be tested if we put yeast strain Z here and test its growth" to the specific commands for making a robot arm do that. The article asks, can we fairly say a robot is a scientist yet? True, the bots are helpless without humans to do some basic support tasks for them, but ask a grad student about his boss... A funny out-of-context quote mentions that "based on our experiences with Adam, we were able to make Eve a much more elegant system". This article ties in with things I've read about computers and creativity.
(Only excerpts are available to non-subscribers, sorry.)
-From Technology Review: Geneticists are finding some surprising behavior in the genome. Apparently it's possible for living things to pass on genetic... effects, without actually passing down the seemingly relevant genes. For instance, female mice inherited some features thought to be controlled by the Y chromosome of their fathers, despite not even having a Y themselves. The explanation seems to involve even more complex control systems hidden in the supposed "junk" portions of the DNA than anyone previously suspected, with a large role played by changes in gene expression that can be inherited separately from actual genes. I know enough about this topic to speculate about mRNA, methlyation of genes, and stuff like that, but it's still pretty confusing. Another article in the same issue says, "Geez, we didn't say the Human Genome Project would instantly cure everything, so get off our case about it!"
-From Scientific American: "Rise of the Robo Scientists". Biologists are experimenting with robotic systems that can themselves do experiments, putting yeast samples into petri dishes and making simple observations. What's interesting is that the the robots are doing some of the work of proposing the tests. They can search their data to propose that gene X is involved in producing protein Y, and go from that idea to "it can be tested if we put yeast strain Z here and test its growth" to the specific commands for making a robot arm do that. The article asks, can we fairly say a robot is a scientist yet? True, the bots are helpless without humans to do some basic support tasks for them, but ask a grad student about his boss... A funny out-of-context quote mentions that "based on our experiences with Adam, we were able to make Eve a much more elegant system". This article ties in with things I've read about computers and creativity.
(Only excerpts are available to non-subscribers, sorry.)