Important applications in microscopy research
on
Quantum Holography
·
· Score: 2, Insightful
It's baffling to me that no one seems to have pointed this out, but the important applications of this technology are likely to be in light microscopy, not scanning luggage. Particularly light microscopy of biological tissues. This could be a very important advance.
I understand why bush made this decision, but I am opposed to ANY federal funding of ANY program that uses tissue from human fetuses
Stem cell research does not use any tissue from human fetuses. It uses cells from a blastocyst, which is an extremely early stage of development, when the cells have only just become an embryo. Not only is it not a fetus, it does not have tissue, which has not developed yet. It is a small ball of cells. Most of the opposition to stem cell research and cloning is because the public is not well-informed about how the science really works and what the potential really is. To be fair, this is largely the fault of scientists, who as a whole press on without making the effort to communicate with the public. I think the real problem is that "life" is really an artificial construct that doesn't map one to one onto physical reality. People want it to be, however, and that's why they go after things like cloning. They argue that it's fundamentally wrong because some seriously fucked up things could be done with the technology, but the techniques involved in "cloning" that people want to ban are essentially ways of manipulating the differentiation of a cell. Every cell in your body (with a few notable exceptions) has the entire genome in it. The information is there to many any type of cell from any other cell. What "cloning" essentially amounts to is things like taking the nucleus from a cell and injecting it into an egg with the nucleus taken out. Different factors present in the cytoplasm of the 2nd cell that are not all mapped out (but involve the repertoire of cytoplasmic transcription factors, proteins that bind to DNA under various circumstances and control which genes are transcribed, or made into RNA) cause the nucleus to differentiate back to a completely undifferentiated cell. This is technically an embryonic stem cell. People object to this idea because it's theoretically possible to take that one cell, grow it for a while, implant it into a uterus, and have a clone of the person come out. However, I find it hard to argue that reversing the differentiation of one of my cells back to the point where a clone of me could theoretically be created is "creating a life." It's just a way of taking out some of my cells and growing me a new liver (or whatever) with them. However, in order to have an experimentally tractable system to work this out in, researchers have to start with embryonic stem cells derived from sperm + egg. Bush's decision is in no way acceptable (largely because stem cell treatments would often involve creation of a new ES cell line tailored to each person), and there's no question that it will be changed once the public's interest dies down. Bush knows as well as anyone that the US can not afford to fall behind in technology, and there's certain things that are done in federally funded academia that simply will not be done in industry, no matter how much money they have. The president wants to please the public, but deep down he knows what's good for the country as a whole, and handicapping American scientists relative to the rest of the world is not it.
Traditional antibiotics when ingested orally, will attempt to destroy ALL bacteria in the gut, possibly leading to a condition called Candidiasis
This is not true. Different antibiotics in use today do have specificities for different kinds of bacteria, and candidiasis is a textbook condition in certain immunosuppressed patients contrary to your assertion that the medical community believes it doesn't exist. However, it is not a common side effect of normal antibiotic use as far as I know.
Antibiotics and growth hormones are given to farm animals, and passed along to humans in the food they eat. I have never had a doctor recommend that I supplement antibiotic use with bifidus and/or acidophilus, and with the amount of training in clinical nutrition/alternative treatments that most doctors get in medschool
While it is very true that antibiotic use in farm animals breeds resistant bacteria and should be banned, it is primarily due to the resistant bacteria being spread to people, not the antibiotics themselves. Even if the antibiotics and recombinant growth hormones were present at high levels in the meat, the growth hormones would never survive cooking and most antibiotics wouldn't either. As far as medical school, I'm unfortunate enough to be in medical school, and they beat us over the head with clinical nutrition and complimentary medicine stuff for hours daily, not that either of those things are in any way related to commonsense ideas like eating yogurt when you're taking antibiotics that cause GI upset by disturbing the normal GI flora.
I haven't read the paper itself, but as far as the specificity issue goes, there would probably be a highly supralinear relationship between membrane concentration of the drug molecules and cell killing ability since they have to multimerize to form the pores, so that would make it easier. I would suspect that these could give you much better specificity than antibiotics commonly used in clinical practice. The mechanism is similar to proteins bacteria make to kill each other (such as gramicidins, colicins, etc.), and those have much higher specificity than any common antibiotics.
I'm a grad student in computational neuroscience, and I just want to point out that AI is not about simulating the brain. That's what we do. AI is about abstracting out the principles that underlie "intelligence" and programming them into a computer. Sort of a top-down approach. AI people will often use so-called "neural nets" and other optimization procedures, but do not confuse this with actually modeling the brain. Of course, all lines are blurry in this kind of work. That said, it's also entirely pointless to draw comparisons between computers and the brain at this level. This is partly because the physiological substrates of computation in the brain are not entirely understood even now (depending who you ask), but also because the brain does things in a way that's best for the brain, not most efficient for a digital computer. For example, if the visual system wants to do some sort of image processing in the spatial frequency domain, it effectively calculates a fourier transform using millions of cells, each exhibiting some spatial frequency tuning. To actually model this process with reasonable biophysical accuracy would bring any supercomputer to its knees, yet functionally equivalent calculations could be carried out on a fast PC. The point is, making computers fast enough to model a whole brain accurately in real time will probably not happen during our lifetimes. However, if one could figure out what the different parts of the brain were doing and perform functionally equivalent computations in silicon, we could do it. That's the goal of AI, not realistic modeling of the nervous system.
Summer movies (like the previous two Jurassic Parks, all of the Batman movies, and The Matrix), according to critics and the Academy of Motion Pictures, suck.
What you're saying is usually true, but it doesn't have to be true. "Decent film" and "fun movie" are not mutually exclusive, and the first Batman is a perfect example of this. A good movie, to me anyway, is one that can deliver both. The Matrix, for example, could have been a truly great movie if they had handed the screenplay and a red pen to any humanities grad student and given him/her half an hour with it. But as it stands, the sheer cheesiness and heavy-handedness of certain parts ("But Neo, I love you! You can't die!") really detracted from what easily could have really good, along the lines of Batman I, the City of Lost Children, the first half of Fight Club, etc. I guess what I'm trying to say is the Mummy Returns doesn't HAVE to be stupid in order to be entertaining, and I think it's fair for Katz or anyone else to criticize it for being a stupid movie despite the fact it makes no pretense of being anything else.
One thing about Linus's argument is that not all of science is done according to open-source ideals. Many of the greatest scientific developments of the past 50 years have come out of industry because there are certain things only industry can do. I work in biological research and my examples will reflect that, but take things like the development of viagra, for example. Most of the work figuring out the pathways of nitric oxide in penile erection, etc. was done in basic science labs in universities. But there reached a certain point where the large-scale high throughput screens for interacting molecules had to be done in industry because they've got the money to throw at a problem like that (this is kind of a poor example, I guess, since viagra had already been developed as a blood pressure medication prior to people figuring out the erection thing). In any case, even discoveries made with public funds lead to patenting of genes and that kind of thing. While I personally agree with Linus's argument, I don't find it entirely convincing. Open source may make better software, but I'm not sure it makes more money, and that's microsoft's goal.
Re:So where does the information come from?
on
A Map to Nowhere?
·
· Score: 1
People have portrayed the task of unravelling the human genome as a Herculean task. Well, the entire genome can be fitted on a CDROM. That isn't very much data at all. Are we really saying that the human body is no more complex that a copy of Windows 2000? Obviously, it is. So where is this extra information located?
Arguments along these lines show a profound lack of understanding of information theory. To be fair, 99.9% of people don't even know what info theory is, and I can't claim to understand all aspects of it myself. But look at it this way: the location of a SINGLE POINT on a number line has an infinite information carrying capacity if you can resolve its location to infinite precision. Say you want to encode War and Peace in the location of a point. You could just number every letter 1 to 26 and stick the point at.12152322... etc. where every two digits corresponds to the appropriate letter in War and Peace. Let's imagine that you wanted to express the location of this point P in a different way, such as giving directions how to get there given rules X and Y. The actual number of bits of information you need to successfully designate point P will depend greatly on what X and Y are. This example is almost so trivial as to be worthless, but I think it gets across the point I'm trying to make. You can't look at "information" in just the number of bits used to encode it, without looking at the system used to decode it. If the human body is more complex than Windows 2000, it's because cellular machinery does a better job of using the genome than a computer does of using code.
I think the common assumption that introns are junk may well turn out to be one of the most glaring fallacies of turn-of-the-century genetics.
Who ever said introns are junk? This is one of those things they teach you in 8th grade science class that no one in molecular biology has believed for as long as I can remember.
Unfortunately, this attitude is rare amongst genetic researchers.
Really? How many "genetic researchers" do you know? The last thing that ever surprises a biologist is when they discover that a system has an extra layer of complexity. But keep in mind how much we (meaning researchers in the biological sciences) know and can do in the lab; it's difficult to argue that we could do those things if our grasp of the basics was not correct.
Re:Actually, the genetic map could tell all for AI
on
A Map to Nowhere?
·
· Score: 1
Forgive me if I'm wrong (and I invite criticism)
I can't say this with absolute certainty, but I worked on HIV for two years, and I'm 99% sure you're wrong. I don't know that much about integration of viral DNA, but HIV and other lentiviruses (a class of retroviruses) have special methods of integration that enable them to do things that other retroviruses can't, namely to integrate their genome into the chromosomal DNA of a cell that is not actively dividing. There was some work being done using integrase inhibitors, which block an enzyme necessary for this process, but it's also not clear whether integration is necessary for transcription of the viral genes (for example, certain white blood cells called macrophages are known to have high levels of circular viral DNA sitting around, but no one knows if it's active or not). In any case, I am pretty sure that the integration of the HIV genome is not base pair specific. Your idea is very interesting, but it would be very difficult to implement in practice, and the virus would mutate to circumvent this type of intervention rather quickly.
I'm sick and tired of this BS about there being less genes than predicted, the failure of the human genome project, etc. The project needed a lot of hype to get the funding it needed, and they succeeded in that, but the final sequencing of the genome is not really a discovery in itself. Biology is a big puzzle, and the human genome project was the project of uncovering all the pieces. Now that the pieces are there, we're ready to make some progress. A lot of people are like "okay, so let's cure AIDS now," but it doesn't work that way. A lot of the discoveries that will result are not going to be directly attributable to the HGP. I do admit that geneticists tended to have an oversimplified view of biological development, but the whole "one gene one protein" thing was shot down DECADES ago with the discovery of things like mRNA splice variants, post-translational modifications, etc. The HGP has nothing to do with that.
While it might have surprised a few people at the very molecular end of the scientific spectrum, the gene number thing didn't surprise most biological scientists, including myself, all that much. The reason is this: if you compare a human vs. a mouse, for example, the extra complexity of a human is not at the level of an individual cell, but at the level of organization of cells. A certain percentage of the genome codes for proteins like collagen and keratin, which are important for formation of tissues. Then some more of it codes for proteins involved in stuff that goes on in all cells, such as polymerases, metabolic pathways, etc. Then there's some that codes for things like liver enzymes, hemoglobin, neurotransmitter synthesis pathways, etc--things specific to certain tissues. At these levels, the requirements for a mouse and a human are very similar. What makes a human more complex is that we have the same building blocks arranged in a much more complex fashion. This requires a greater number of genes involved in mediating interaction between cells, namely various receptors and signaling molecules. Taking this into account, it is an extremely primitive idea to think "Oh, we're ten times as complex as a mouse, so we must have ten times as many genes." It's more likely that we just have ten times as many of the subset of genes involved in development, complex formation of tissues, etc. And I shouldn't even say ten times because it's also important to realize that the complexity of this system will almost certainly scale supralinearly with the number of elements/genes involved.
In any case, it's extremely premature (and annoying) to make arguments against the success of the human genome project. Maybe the HGP people are guilty of making a lot of hype that they never had any intention of living up to in the short run, but it's simply incorrect to say that scientists the world over are shocked and dismayed by the unexpected findings of the genome project. Or that this shoots down the "one gene one protein" thing had been defunct for years. What the HGP opens the door on is scientific examination of regulation of expression of genes, etc. How genes interact with each other. This is where real breakthroughs in understanding are going to be made.
Oh, one last thing: very few biologists bought the "junk DNA" thing either. That's public perception, but we've known for YEARS that it plays important roles in gene regulation, recombination, etc. etc. Also, nothing discovered in the genome project provides any evidence supporting any religious ideas whatsoever. Anyone who thinks that does not understand what they're talking about. On the other hand, it doesn't really provide new evidence on evolution since those arguments were already convincing before the HGP was finished.
Something along these lines has already been tried. It was called the Opticon, I think, and it's most famous for being used by the blind computer guy in Sneakers. However, in real life the devices were a failure because the engineers who designed them didn't know anything about somatosensory neurophysiology. People have several different "touch" receptors in their fingertips that send information up to the brain through different axons (the individual fibers of a nerve). Each of these different types of receptors responds to mechanical stimuli at different frequencies, and the ones the brain uses to read Braille are primarily Merkel disks, which are sensitive to a different frequency range than the device operates at. Apparently because of the design of the device, which uses piezo crystals, you can't tune the frequency very well. I heard they were working on an improved version using shape memory alloys, but I don't know how far that got.
Does anyone know why all audio codecs seem to be using frequency domain representations? It seems to me that you would have problems representing certain waveforms (I understand there's something called wavelets to circumvent this, but I haven't looked through the math yet). Does anyone know if there are any methods based on something different, such as principal components/eigenspace projection?
As a neuroscience grad student, I just wanted to make a few comments in defense of biological scientists. First, we are often accused of pushing ethics aside in favor of things we think are neat on an abstract scientific level. While there is certainly some truth to that, the primary difference between a scientist's perspective and a layperson's perspective is that the scientist generally has a better understanding of what the technological advances might actually mean in the not-too-distant future. For example, when most people hear cloning, they think of armies of duplicate slaves or something, but they don't realize that the techniques involved are likely to play important roles in the cures for many diseases, such as Parkinson's, Huntington's, and Alzheimer's (which will DESTROY the American economy if we don't figure it out before it hits the baby boomers). Likewise, technologies like transgenic rabbits expressing green fluorescent protein or whatever make many laypeople cringe, thinking of big industry selling bizarre genetically engineered novelty pets or something (I mention this because a previous poster did). They don't realize that certain breeds of rabbits are one of the number one model organisms for heart disease, a huge public health problem in the developed world, and there are various valid approaches for treatment with gene therapy. These things have to be worked out in animals before we try them on humans. Biotech has been portrayed as evil because of stupid marketing mistakes made by big companies making transgenic crops and etc., and people act like "golden rice" is the only good thing that's ever come of biomedical research. My point is, if that's what you think, you need to become better informed. Try working in a hospital for a while (I'm a med student too) and look at what ridiculous advances in medical technology have been made in even the past five years. Or try having a relative dying of a disease because new treatments are being delayed by ethical protests against testing it in mice. In any case, the thing I find most disturbing is that you never seem to see many people on/. talking about the ethical implications of quantum computing. Quantum computers that could quickly factor products of huge primes, for example, could completely turn the world upside down. It could topple governments or start world war 3. Then again, it might not, but a lot of poly sci and international relations people I know think it would, and in any case it's a good example of the biases irrational trends in pop culture place on the lay person's judgment of the ethics of various scientific developments.
The G3 is chambered in 7.62x51mm. The first poster said you could get 10k fps with this round, which is ridiculous with a 7.62mm bullet, but there are special rounds that use a 5.56mm bullet surrounded by a plastic sabot, and I think you can get over 10,000 fps with those. Of course, they're not going to "penetrate a foot of steel." Also, I should point out that while the AK-47 and G3 both use 7.62mm bullets, the AK uses 7.62x39mm rounds, which are shorter and lower power, so the bullet itself is usually shorter and lighter despite having the same diameter.
And heart "disease" is just junk building up in the arteries and such, isn't it?
Type I diabetes mellitus and rheumatoid arthritis are regarded as autoimmune diseases. Atherosclerosis, the building up of junk in the walls of blood vessels, is largely dependent on immune system white blood cells called monocytes leaving the bloodstream and differentiating into cells called macrophages. Although it is probably not an autoimmune disease in the sense of some host protein being mistaken for foreign, it is an immune system process gone awry, and the best approaches for fighting it are ones that interfere with the improper functioning of the immune system.
First, when people talk about genetic similarity between species, they're generally doing fairly meaningless things like looking at sequences of coding regions of certain genes (the parts that get transcribed and made into protein). This is pretty meaningless for a couple of reasons: first, sequence similarity does not correlate 100% with similarity at the protein level. The dolphin hemoglobin gene might have greater sequence homology to the human gene than mouse hemoglobin (it probably doesn't, but it could), but one could most likely replace every hemoglobin molecule in a human body with mouse hemoglobin with no problem. Dolphin hemoglobin, OTOH, has different O2 and CO2 binding characteristics, and would probably kill the person if the switch were made.
The more important thing to note, however, is that real differences in species are primarily due to different patterns of gene expression rather than different sequences in the coding regions of the genes. When you think about it, the difference between a human and a puffer fish is way beyond the genetic level, anyway. Fibroblasts or white blood cells from either species would look the same growing in a dish--it's the organization of the cells into tissues, the tissues into organs, etc. that makes the difference. If someone's sequencing the fugu genome, it's probably not because the fish is so similar to humans, but because the fish is cheap and easy to work with in a certain experimental preparation.
Re:C) scientists like you are unwittingly...
on
Golden Rice
·
· Score: 1
The reason why I (as a scientist) disagree with this line of reasoning is that there really is no unified scientific community. If I could think of any way to prove that golden rice was dangerous, I would in a second because my career would benefit GREATLY as a result. There are many scientists who would love to prove that GMOs are dangerous so they could be on the cover of Time magazine too, but so far none of them have been able to do it.
a biologist's opinion of the GM food debate
on
Golden Rice
·
· Score: 1
As a biologist, I usually avoid reading/. discussions on anything biology-related. Usually it regresses to simple-minded discussions about nanobots from people who don't understand the difference between DNA and RNA. However, I was pleasantly surprised to see a rational response (for the most part) to the golden rice story. As far as social issues go, I think a good analogy is to compare vitamin A deficiency to heart disease in the US. Yes, it's not a genetic problem, it's a problem of overeating, lack of exercise, etc. However, heart medications save lives. Golden rice does not answer the questions of world poverty, but it will save many lives and prevent many many more cases of blindness, which is a huge strain on many third-world economies.
However, the real reason I'm posting is that many people are still misinformed about the technical aspects of genetic engineering. Everyone should should have their own opinions on social issues, but the underlying technical information should not be incorrect.
Misinformation stems largely from distribution of information that caters to naturalistic ideas and human arrogance, which love to have us believe that "life" is this thing of complexity too grand for mere mortals to ever understand. Well, to be honest, most biologists (myself included) get a little pissed off when people keep telling us over and over again what we don't understand, how we "tinker" mindlessly, "play god," etc. I'm not saying there are no dangers to GMOs whatsoever. But as biologists, trust me when I say that WE KNOW WHAT WE'RE DOING. Just as it is beyond my comprehension that people can split atoms, make rockets fly to the moon, or build skyscrapers, what we as biologists do is also completely beyond the comprehension of the general public. I truly wish this were not the case, but it is, and everyone must accept this. As a biologist, I have to say that the dangers of GMOs are greatly outweighed by the potential benefits. And the vast vast majority of biological scientists I know agree on this (and I am in nonprofit academia, not in industry).
I can't remember all the dangers of GMOs people usually talk about off the top of my head, but trust me that they are all things that can be controlled. As far as the general ones, here's a link on antibiotic resistance:
http://vm.cfsan.fda.gov/~dms/opa-armg.html
There's also the issue of allergenicity...I don't have a link for that one. It is a potential danger, but it will not be a problem if we do safety testing for it. Plus, we can predict which proteins would be likely to cause allergic responses. Opponents of GMOs love to say that this will be impossible to control for because of fusion proteins, upregulation of endogenous gene products adjacent to the promoter inserted with the transgene, etc., but it is easy to check where the insertions occur if we do them randomly, and more importantly, we do not have to do them randomly.
Those are the only two dangers I can think of that would apply to ALL GMO foods including golden rice. The rest are common sense things like don't plant herbicide resistant oats across the street from wild oats. I'll find a couple other scientifically sound URLs on this and post them later.
I second the idea to use ZIP drives. In the medical school I go to, every computer has a ZIP drive in it, and every student is given a zip disk with a copy of Eudora on it. So everyone carries one zip disk around, with all of their homework, all of their email, etc. all on one disk. This is really nice, as you can't telnet in to the email server, so it probably means less work for the admins not giving us all shell accounts. I mean, it's far from a perfect system, but it bypasses many of the problems I experienced in college with public hard disks, broken floppy drives, etc.
Re:Implant a receiver, have clip-on alternate sens
on
End To Blindness?
·
· Score: 1
Stimulation of visual cortex for a visual prosthesis for the blind is not a new idea. It was first done back in the 60's (I think) by Brindley and Lewin. There even used to be a neural prosthetics division working on stuff like that over at the NIH, but I guess the consensus was that there needed to be more animal testing done first. Stimulating primary visual cortex is going to be a lot trickier than the retina, though. First of all, it's huge--you would have to get a large portion of it in order to have any real sight, plus it's subdivided on a microscopic level into cells that get input from one eye, ones that get input from both, ones that see certain colors, ones that are tuned for orientation of a contrast boundary, ones tuned for temporal and spatial frequencies, ones for spatial phase, etc. etc. etc. It's not as easy as the retina, where the optics make things a lot more obvious. I'm guessing that within our lifetime there will be prostheses for primary visual cortex that will allow people to see the same way cochlear implants allows people to hear; that is, well enough to function, but not as well as a normal person.
Re:When prosthetic eyes are better...
on
End To Blindness?
·
· Score: 1
It is highly unlikely that prosthetic eyes will ever be better than real eyes. I'm in grad school for visual neurophysiology, and when you start to learn what the visual system can actually do, you realize there is no man-made device that even comes close. Photoreceptor cells can reliably detect the presence of one photon, and an individual rhodopsin molecule falsely registers a photon approximately once per 500 years at physiological temperatures (I forget the numbers for a whole cell, but it's still on the order of hours, if not days). The eye has a dynamic range orders of magnitude greater than any electronic camera I'm aware of in addition to motor tracking that is very difficult to implement in hardware. The retina also has mechanisms to compensate for motion of rapidly moving objects, which would also be difficult to implement in hardware. But the real challenge would be implementing the feedback mechanisms that your brain uses to tell your eyes exactly where to look, to focus and filter differently, etc. Not to mention depth and motion perception by circuits that integrate things like feedback from head movement with changes in binocular disparity (how different what your two eyes see is), differential movement of objects across the visual field, etc. Even if you could make something small enough that would be optically better than an eye, it would still be lagging far behind the eye in real vision because of the problems of interfacing it with everything else already there. Prosthetic eyes better than natural ones are unlikely to exist in the lifetimes of our grandchildren.
hi, I'm a neuroscience grad student, and I have to comment that basically all of the discussion going on everyone here seriously misunderstanding what this result means. I'm in a hurry, so I can't go into great depth. But to make things simple, the protein they overexpressed in those mice is involved in synaptic modification. So they made mice with more modifiable synapses. The title of the paper is "Enhanced LEARNING after genetic overexpression of a brain growth protein." Learning is not the same thing as intelligence. No one is claiming that anyone made smarter mice except the journalists. The researchers made the brain more modifiable, so the mice could learn a piece of information more quickly. That doesn't mean they could remember it as long, that they could think as well as normal mice, etc. There are a million other factors, but the bottom line is that all they demonstrated was that the transgenic mice were better at a certain task than wild type mice. This could be because they learned faster, or it could be for any number of unrelated side effects (such as ones involving changes in vision). This is like demonstrating that fighter pilots can do some maneuver 95% on cocaine and only 92% normally, and from this extrapolating that it would be beneficial for everyone to install cocaine dispensers that keep them high 24/7. This particular study has no serious implications for evolution or a society of superhumans. The relationship between genotype and brain phenotype is way too complex for anyone to come up with a one-gene solution to creating greater "intelligence." Anything like that will likely involve many genes and large-scale screening of individuals followed by correlations with some measure of the person's intellect, but I won't even start on the problems involved in a study like that...
There are a few misconceptions about evolution that I think need to be cleared up:
First, one can think of evolution as being dependent on a large percentage of the population dying. On earth, this is still true. 40,000 children die every day of preventable illnesses. 16,000 people are infected with HIV every day (which will undoubtedly cause selected populations of people with disproportionate levels of mutations in the CCR5 chemokine receptor that renders them resistant to infection through mucosal membranes). But maybe a more important thing is the idea of unborn children. In the end, the groups having the most children will win out. Evolution still occurs, just not by the same factors as 15,000 years ago.
Another thing to remember is that humans evolve/have evolved as a group, not as individuals. This is because homo sapiens is a pack animal, like ants or wolves. Long fangs and aggressiveness might help out a bear in the wild, but would probably just have gotten a homo erectus clubbed to death by the rest of the group. Traits like kindness and self-sacrifice (or propensity for religious belief) were selected for because they promoted the survival of the race as a whole. So in this sense, "social evolution" is nothing new.
A third thing, men's attraction to large breasts is based on breasts being one of the things that differentiates fertile women from prepubescent girls and other men, not women with large breasts from women with small breasts, as there is no serious correlation of breast size with ability to nourish a child (at least not a more significant correlation than there would be with other obvious markers of health). This is a subtle distinction, but an important one, I think.
As far as I can tell, he just hates Bob Gallo, the guy whose group first discovered that AIDS is caused by HIV (for those who don't know, Mullis and Peter Duesberg's whole thing is that HIV does not cause AIDS). Gallo is the guy played by Alan Alda in that ridiculous inaccurate "And the Band Played On" movie, if anyone's seen that. I've never met Kary Mullis or even looked at his data if he has any; this is just my impression from reading his web pages. I can't remember the URLs, but they're full of quotes like "People are going to know I'm right, as soon as they find out what an asshole Gallo is!" As far as Gallo, I could see why people would dislike him since he seems a little too politician. I've seen him around several times but never met him personally (yet I get gold-leaf xmas cards stamped ROBERT C. GALLO AND STAFF), but my boss and all the other faculty doing HIV stuff in my department know him really well and think highly of him. FWIW, the whole stealing data from the French thing was basically entirely fictionalized for the movie. As far as I can tell, Mullis's motivation seems to be proving that AIDS is caused by the use of amyl nitrite poppers, or ingesting semen, or a complex combination of retroviruses, or whatever they say it is now, as long as it isn't what Gallo and probably some other enemies said it was, HIV.
Sorry, I didn't see the link in the article. That would certainly decrease the cost of it, but I still have to say that the thought of doing PCR without a decent thermal cycler or even pipettes makes me cringe. My real problem with the article wasn't even the cost issue as much as the fact that the person who wrote it obviously didn't know what he was talking about, and I think it would be a lot tougher to people to get this working in their kitchen than he implied. Anyway, I think it would be much more interesting if someone took this one step further and packaged a kit that actually DID something, such as amplify a region of DNA and test it for a restriction fragment length polymorphism ("hey Dad! You're predisposed for heart disease!"). You could sell some polymerase, dNTPs, primers, and a bit of restriction enzyme to test the product afterwards. You could even put a shrink-wrapped agarose gel in there so people wouldn't have to boil ethidium bromide solutions in their microwave.
Slashdot Mirror
It's baffling to me that no one seems to have pointed this out, but the important applications of this technology are likely to be in light microscopy, not scanning luggage. Particularly light microscopy of biological tissues. This could be a very important advance.
I understand why bush made this decision, but I am opposed to ANY federal funding of ANY program that uses tissue from human fetuses
Stem cell research does not use any tissue from human fetuses. It uses cells from a blastocyst, which is an extremely early stage of development, when the cells have only just become an embryo. Not only is it not a fetus, it does not have tissue, which has not developed yet. It is a small ball of cells. Most of the opposition to stem cell research and cloning is because the public is not well-informed about how the science really works and what the potential really is. To be fair, this is largely the fault of scientists, who as a whole press on without making the effort to communicate with the public. I think the real problem is that "life" is really an artificial construct that doesn't map one to one onto physical reality. People want it to be, however, and that's why they go after things like cloning. They argue that it's fundamentally wrong because some seriously fucked up things could be done with the technology, but the techniques involved in "cloning" that people want to ban are essentially ways of manipulating the differentiation of a cell. Every cell in your body (with a few notable exceptions) has the entire genome in it. The information is there to many any type of cell from any other cell. What "cloning" essentially amounts to is things like taking the nucleus from a cell and injecting it into an egg with the nucleus taken out. Different factors present in the cytoplasm of the 2nd cell that are not all mapped out (but involve the repertoire of cytoplasmic transcription factors, proteins that bind to DNA under various circumstances and control which genes are transcribed, or made into RNA) cause the nucleus to differentiate back to a completely undifferentiated cell. This is technically an embryonic stem cell. People object to this idea because it's theoretically possible to take that one cell, grow it for a while, implant it into a uterus, and have a clone of the person come out. However, I find it hard to argue that reversing the differentiation of one of my cells back to the point where a clone of me could theoretically be created is "creating a life." It's just a way of taking out some of my cells and growing me a new liver (or whatever) with them. However, in order to have an experimentally tractable system to work this out in, researchers have to start with embryonic stem cells derived from sperm + egg. Bush's decision is in no way acceptable (largely because stem cell treatments would often involve creation of a new ES cell line tailored to each person), and there's no question that it will be changed once the public's interest dies down. Bush knows as well as anyone that the US can not afford to fall behind in technology, and there's certain things that are done in federally funded academia that simply will not be done in industry, no matter how much money they have. The president wants to please the public, but deep down he knows what's good for the country as a whole, and handicapping American scientists relative to the rest of the world is not it.
Traditional antibiotics when ingested orally, will attempt to destroy ALL bacteria in the gut, possibly leading to a condition called Candidiasis
This is not true. Different antibiotics in use today do have specificities for different kinds of bacteria, and candidiasis is a textbook condition in certain immunosuppressed patients contrary to your assertion that the medical community believes it doesn't exist. However, it is not a common side effect of normal antibiotic use as far as I know.
Antibiotics and growth hormones are given to farm animals, and passed along to humans in the food they eat. I have never had a doctor recommend that I supplement antibiotic use with bifidus and/or acidophilus, and with the amount of training in clinical nutrition/alternative treatments that most doctors get in medschool
While it is very true that antibiotic use in farm animals breeds resistant bacteria and should be banned, it is primarily due to the resistant bacteria being spread to people, not the antibiotics themselves. Even if the antibiotics and recombinant growth hormones were present at high levels in the meat, the growth hormones would never survive cooking and most antibiotics wouldn't either. As far as medical school, I'm unfortunate enough to be in medical school, and they beat us over the head with clinical nutrition and complimentary medicine stuff for hours daily, not that either of those things are in any way related to commonsense ideas like eating yogurt when you're taking antibiotics that cause GI upset by disturbing the normal GI flora.
I haven't read the paper itself, but as far as the specificity issue goes, there would probably be a highly supralinear relationship between membrane concentration of the drug molecules and cell killing ability since they have to multimerize to form the pores, so that would make it easier. I would suspect that these could give you much better specificity than antibiotics commonly used in clinical practice. The mechanism is similar to proteins bacteria make to kill each other (such as gramicidins, colicins, etc.), and those have much higher specificity than any common antibiotics.
I'm a grad student in computational neuroscience, and I just want to point out that AI is not about simulating the brain. That's what we do. AI is about abstracting out the principles that underlie "intelligence" and programming them into a computer. Sort of a top-down approach. AI people will often use so-called "neural nets" and other optimization procedures, but do not confuse this with actually modeling the brain. Of course, all lines are blurry in this kind of work. That said, it's also entirely pointless to draw comparisons between computers and the brain at this level. This is partly because the physiological substrates of computation in the brain are not entirely understood even now (depending who you ask), but also because the brain does things in a way that's best for the brain, not most efficient for a digital computer. For example, if the visual system wants to do some sort of image processing in the spatial frequency domain, it effectively calculates a fourier transform using millions of cells, each exhibiting some spatial frequency tuning. To actually model this process with reasonable biophysical accuracy would bring any supercomputer to its knees, yet functionally equivalent calculations could be carried out on a fast PC. The point is, making computers fast enough to model a whole brain accurately in real time will probably not happen during our lifetimes. However, if one could figure out what the different parts of the brain were doing and perform functionally equivalent computations in silicon, we could do it. That's the goal of AI, not realistic modeling of the nervous system.
Summer movies (like the previous two Jurassic Parks, all of the Batman movies, and The Matrix), according to critics and the Academy of Motion Pictures, suck.
What you're saying is usually true, but it doesn't have to be true. "Decent film" and "fun movie" are not mutually exclusive, and the first Batman is a perfect example of this. A good movie, to me anyway, is one that can deliver both. The Matrix, for example, could have been a truly great movie if they had handed the screenplay and a red pen to any humanities grad student and given him/her half an hour with it. But as it stands, the sheer cheesiness and heavy-handedness of certain parts ("But Neo, I love you! You can't die!") really detracted from what easily could have really good, along the lines of Batman I, the City of Lost Children, the first half of Fight Club, etc. I guess what I'm trying to say is the Mummy Returns doesn't HAVE to be stupid in order to be entertaining, and I think it's fair for Katz or anyone else to criticize it for being a stupid movie despite the fact it makes no pretense of being anything else.
One thing about Linus's argument is that not all of science is done according to open-source ideals. Many of the greatest scientific developments of the past 50 years have come out of industry because there are certain things only industry can do. I work in biological research and my examples will reflect that, but take things like the development of viagra, for example. Most of the work figuring out the pathways of nitric oxide in penile erection, etc. was done in basic science labs in universities. But there reached a certain point where the large-scale high throughput screens for interacting molecules had to be done in industry because they've got the money to throw at a problem like that (this is kind of a poor example, I guess, since viagra had already been developed as a blood pressure medication prior to people figuring out the erection thing). In any case, even discoveries made with public funds lead to patenting of genes and that kind of thing. While I personally agree with Linus's argument, I don't find it entirely convincing. Open source may make better software, but I'm not sure it makes more money, and that's microsoft's goal.
People have portrayed the task of unravelling the human genome as a Herculean task. Well, the entire genome can be fitted on a CDROM. That isn't very much data at all. Are we really saying that the human body is no more complex that a copy of Windows 2000? Obviously, it is. So where is this extra information located?
.12152322... etc. where every two digits corresponds to the appropriate letter in War and Peace. Let's imagine that you wanted to express the location of this point P in a different way, such as giving directions how to get there given rules X and Y. The actual number of bits of information you need to successfully designate point P will depend greatly on what X and Y are. This example is almost so trivial as to be worthless, but I think it gets across the point I'm trying to make. You can't look at "information" in just the number of bits used to encode it, without looking at the system used to decode it. If the human body is more complex than Windows 2000, it's because cellular machinery does a better job of using the genome than a computer does of using code.
Arguments along these lines show a profound lack of understanding of information theory. To be fair, 99.9% of people don't even know what info theory is, and I can't claim to understand all aspects of it myself. But look at it this way: the location of a SINGLE POINT on a number line has an infinite information carrying capacity if you can resolve its location to infinite precision. Say you want to encode War and Peace in the location of a point. You could just number every letter 1 to 26 and stick the point at
I think the common assumption that introns are junk may well turn out to be one of the most glaring fallacies of turn-of-the-century genetics.
Who ever said introns are junk? This is one of those things they teach you in 8th grade science class that no one in molecular biology has believed for as long as I can remember.
Unfortunately, this attitude is rare amongst genetic researchers.
Really? How many "genetic researchers" do you know? The last thing that ever surprises a biologist is when they discover that a system has an extra layer of complexity. But keep in mind how much we (meaning researchers in the biological sciences) know and can do in the lab; it's difficult to argue that we could do those things if our grasp of the basics was not correct.
Forgive me if I'm wrong (and I invite criticism)
I can't say this with absolute certainty, but I worked on HIV for two years, and I'm 99% sure you're wrong. I don't know that much about integration of viral DNA, but HIV and other lentiviruses (a class of retroviruses) have special methods of integration that enable them to do things that other retroviruses can't, namely to integrate their genome into the chromosomal DNA of a cell that is not actively dividing. There was some work being done using integrase inhibitors, which block an enzyme necessary for this process, but it's also not clear whether integration is necessary for transcription of the viral genes (for example, certain white blood cells called macrophages are known to have high levels of circular viral DNA sitting around, but no one knows if it's active or not). In any case, I am pretty sure that the integration of the HIV genome is not base pair specific. Your idea is very interesting, but it would be very difficult to implement in practice, and the virus would mutate to circumvent this type of intervention rather quickly.
I'm sick and tired of this BS about there being less genes than predicted, the failure of the human genome project, etc. The project needed a lot of hype to get the funding it needed, and they succeeded in that, but the final sequencing of the genome is not really a discovery in itself. Biology is a big puzzle, and the human genome project was the project of uncovering all the pieces. Now that the pieces are there, we're ready to make some progress. A lot of people are like "okay, so let's cure AIDS now," but it doesn't work that way. A lot of the discoveries that will result are not going to be directly attributable to the HGP. I do admit that geneticists tended to have an oversimplified view of biological development, but the whole "one gene one protein" thing was shot down DECADES ago with the discovery of things like mRNA splice variants, post-translational modifications, etc. The HGP has nothing to do with that.
While it might have surprised a few people at the very molecular end of the scientific spectrum, the gene number thing didn't surprise most biological scientists, including myself, all that much. The reason is this: if you compare a human vs. a mouse, for example, the extra complexity of a human is not at the level of an individual cell, but at the level of organization of cells. A certain percentage of the genome codes for proteins like collagen and keratin, which are important for formation of tissues. Then some more of it codes for proteins involved in stuff that goes on in all cells, such as polymerases, metabolic pathways, etc. Then there's some that codes for things like liver enzymes, hemoglobin, neurotransmitter synthesis pathways, etc--things specific to certain tissues. At these levels, the requirements for a mouse and a human are very similar. What makes a human more complex is that we have the same building blocks arranged in a much more complex fashion. This requires a greater number of genes involved in mediating interaction between cells, namely various receptors and signaling molecules. Taking this into account, it is an extremely primitive idea to think "Oh, we're ten times as complex as a mouse, so we must have ten times as many genes." It's more likely that we just have ten times as many of the subset of genes involved in development, complex formation of tissues, etc. And I shouldn't even say ten times because it's also important to realize that the complexity of this system will almost certainly scale supralinearly with the number of elements/genes involved.
In any case, it's extremely premature (and annoying) to make arguments against the success of the human genome project. Maybe the HGP people are guilty of making a lot of hype that they never had any intention of living up to in the short run, but it's simply incorrect to say that scientists the world over are shocked and dismayed by the unexpected findings of the genome project. Or that this shoots down the "one gene one protein" thing had been defunct for years. What the HGP opens the door on is scientific examination of regulation of expression of genes, etc. How genes interact with each other. This is where real breakthroughs in understanding are going to be made.
Oh, one last thing: very few biologists bought the "junk DNA" thing either. That's public perception, but we've known for YEARS that it plays important roles in gene regulation, recombination, etc. etc. Also, nothing discovered in the genome project provides any evidence supporting any religious ideas whatsoever. Anyone who thinks that does not understand what they're talking about. On the other hand, it doesn't really provide new evidence on evolution since those arguments were already convincing before the HGP was finished.
Something along these lines has already been tried. It was called the Opticon, I think, and it's most famous for being used by the blind computer guy in Sneakers. However, in real life the devices were a failure because the engineers who designed them didn't know anything about somatosensory neurophysiology. People have several different "touch" receptors in their fingertips that send information up to the brain through different axons (the individual fibers of a nerve). Each of these different types of receptors responds to mechanical stimuli at different frequencies, and the ones the brain uses to read Braille are primarily Merkel disks, which are sensitive to a different frequency range than the device operates at. Apparently because of the design of the device, which uses piezo crystals, you can't tune the frequency very well. I heard they were working on an improved version using shape memory alloys, but I don't know how far that got.
Does anyone know why all audio codecs seem to be using frequency domain representations? It seems to me that you would have problems representing certain waveforms (I understand there's something called wavelets to circumvent this, but I haven't looked through the math yet). Does anyone know if there are any methods based on something different, such as principal components/eigenspace projection?
As a neuroscience grad student, I just wanted to make a few comments in defense of biological scientists. First, we are often accused of pushing ethics aside in favor of things we think are neat on an abstract scientific level. While there is certainly some truth to that, the primary difference between a scientist's perspective and a layperson's perspective is that the scientist generally has a better understanding of what the technological advances might actually mean in the not-too-distant future. For example, when most people hear cloning, they think of armies of duplicate slaves or something, but they don't realize that the techniques involved are likely to play important roles in the cures for many diseases, such as Parkinson's, Huntington's, and Alzheimer's (which will DESTROY the American economy if we don't figure it out before it hits the baby boomers). Likewise, technologies like transgenic rabbits expressing green fluorescent protein or whatever make many laypeople cringe, thinking of big industry selling bizarre genetically engineered novelty pets or something (I mention this because a previous poster did). They don't realize that certain breeds of rabbits are one of the number one model organisms for heart disease, a huge public health problem in the developed world, and there are various valid approaches for treatment with gene therapy. These things have to be worked out in animals before we try them on humans. Biotech has been portrayed as evil because of stupid marketing mistakes made by big companies making transgenic crops and etc., and people act like "golden rice" is the only good thing that's ever come of biomedical research. My point is, if that's what you think, you need to become better informed. Try working in a hospital for a while (I'm a med student too) and look at what ridiculous advances in medical technology have been made in even the past five years. Or try having a relative dying of a disease because new treatments are being delayed by ethical protests against testing it in mice. In any case, the thing I find most disturbing is that you never seem to see many people on /. talking about the ethical implications of quantum computing. Quantum computers that could quickly factor products of huge primes, for example, could completely turn the world upside down. It could topple governments or start world war 3. Then again, it might not, but a lot of poly sci and international relations people I know think it would, and in any case it's a good example of the biases irrational trends in pop culture place on the lay person's judgment of the ethics of various scientific developments.
The G3 is chambered in 7.62x51mm. The first poster said you could get 10k fps with this round, which is ridiculous with a 7.62mm bullet, but there are special rounds that use a 5.56mm bullet surrounded by a plastic sabot, and I think you can get over 10,000 fps with those. Of course, they're not going to "penetrate a foot of steel." Also, I should point out that while the AK-47 and G3 both use 7.62mm bullets, the AK uses 7.62x39mm rounds, which are shorter and lower power, so the bullet itself is usually shorter and lighter despite having the same diameter.
And heart "disease" is just junk building up in the arteries and such, isn't it?
Type I diabetes mellitus and rheumatoid arthritis are regarded as autoimmune diseases. Atherosclerosis, the building up of junk in the walls of blood vessels, is largely dependent on immune system white blood cells called monocytes leaving the bloodstream and differentiating into cells called macrophages. Although it is probably not an autoimmune disease in the sense of some host protein being mistaken for foreign, it is an immune system process gone awry, and the best approaches for fighting it are ones that interfere with the improper functioning of the immune system.
A couple of comments on this:
First, when people talk about genetic similarity between species, they're generally doing fairly meaningless things like looking at sequences of coding regions of certain genes (the parts that get transcribed and made into protein). This is pretty meaningless for a couple of reasons: first, sequence similarity does not correlate 100% with similarity at the protein level. The dolphin hemoglobin gene might have greater sequence homology to the human gene than mouse hemoglobin (it probably doesn't, but it could), but one could most likely replace every hemoglobin molecule in a human body with mouse hemoglobin with no problem. Dolphin hemoglobin, OTOH, has different O2 and CO2 binding characteristics, and would probably kill the person if the switch were made.
The more important thing to note, however, is that real differences in species are primarily due to different patterns of gene expression rather than different sequences in the coding regions of the genes. When you think about it, the difference between a human and a puffer fish is way beyond the genetic level, anyway. Fibroblasts or white blood cells from either species would look the same growing in a dish--it's the organization of the cells into tissues, the tissues into organs, etc. that makes the difference. If someone's sequencing the fugu genome, it's probably not because the fish is so similar to humans, but because the fish is cheap and easy to work with in a certain experimental preparation.
The reason why I (as a scientist) disagree with this line of reasoning is that there really is no unified scientific community. If I could think of any way to prove that golden rice was dangerous, I would in a second because my career would benefit GREATLY as a result. There are many scientists who would love to prove that GMOs are dangerous so they could be on the cover of Time magazine too, but so far none of them have been able to do it.
As a biologist, I usually avoid reading /. discussions on anything biology-related. Usually it regresses to simple-minded discussions about nanobots from people who don't understand the difference between DNA and RNA. However, I was pleasantly surprised to see a rational response (for the most part) to the golden rice story. As far as social issues go, I think a good analogy is to compare vitamin A deficiency to heart disease in the US. Yes, it's not a genetic problem, it's a problem of overeating, lack of exercise, etc. However, heart medications save lives. Golden rice does not answer the questions of world poverty, but it will save many lives and prevent many many more cases of blindness, which is a huge strain on many third-world economies.
However, the real reason I'm posting is that many people are still misinformed about the technical aspects of genetic engineering. Everyone should should have their own opinions on social issues, but the underlying technical information should not be incorrect. Misinformation stems largely from distribution of information that caters to naturalistic ideas and human arrogance, which love to have us believe that "life" is this thing of complexity too grand for mere mortals to ever understand. Well, to be honest, most biologists (myself included) get a little pissed off when people keep telling us over and over again what we don't understand, how we "tinker" mindlessly, "play god," etc. I'm not saying there are no dangers to GMOs whatsoever. But as biologists, trust me when I say that WE KNOW WHAT WE'RE DOING. Just as it is beyond my comprehension that people can split atoms, make rockets fly to the moon, or build skyscrapers, what we as biologists do is also completely beyond the comprehension of the general public. I truly wish this were not the case, but it is, and everyone must accept this. As a biologist, I have to say that the dangers of GMOs are greatly outweighed by the potential benefits. And the vast vast majority of biological scientists I know agree on this (and I am in nonprofit academia, not in industry).
I can't remember all the dangers of GMOs people usually talk about off the top of my head, but trust me that they are all things that can be controlled. As far as the general ones, here's a link on antibiotic resistance:
http://vm.cfsan.fda.gov/~dms/opa-armg.html
There's also the issue of allergenicity...I don't have a link for that one. It is a potential danger, but it will not be a problem if we do safety testing for it. Plus, we can predict which proteins would be likely to cause allergic responses. Opponents of GMOs love to say that this will be impossible to control for because of fusion proteins, upregulation of endogenous gene products adjacent to the promoter inserted with the transgene, etc., but it is easy to check where the insertions occur if we do them randomly, and more importantly, we do not have to do them randomly.
Those are the only two dangers I can think of that would apply to ALL GMO foods including golden rice. The rest are common sense things like don't plant herbicide resistant oats across the street from wild oats. I'll find a couple other scientifically sound URLs on this and post them later.
I second the idea to use ZIP drives. In the medical school I go to, every computer has a ZIP drive in it, and every student is given a zip disk with a copy of Eudora on it. So everyone carries one zip disk around, with all of their homework, all of their email, etc. all on one disk. This is really nice, as you can't telnet in to the email server, so it probably means less work for the admins not giving us all shell accounts. I mean, it's far from a perfect system, but it bypasses many of the problems I experienced in college with public hard disks, broken floppy drives, etc.
Stimulation of visual cortex for a visual prosthesis for the blind is not a new idea. It was first done back in the 60's (I think) by Brindley and Lewin. There even used to be a neural prosthetics division working on stuff like that over at the NIH, but I guess the consensus was that there needed to be more animal testing done first. Stimulating primary visual cortex is going to be a lot trickier than the retina, though. First of all, it's huge--you would have to get a large portion of it in order to have any real sight, plus it's subdivided on a microscopic level into cells that get input from one eye, ones that get input from both, ones that see certain colors, ones that are tuned for orientation of a contrast boundary, ones tuned for temporal and spatial frequencies, ones for spatial phase, etc. etc. etc. It's not as easy as the retina, where the optics make things a lot more obvious. I'm guessing that within our lifetime there will be prostheses for primary visual cortex that will allow people to see the same way cochlear implants allows people to hear; that is, well enough to function, but not as well as a normal person.
It is highly unlikely that prosthetic eyes will ever be better than real eyes. I'm in grad school for visual neurophysiology, and when you start to learn what the visual system can actually do, you realize there is no man-made device that even comes close. Photoreceptor cells can reliably detect the presence of one photon, and an individual rhodopsin molecule falsely registers a photon approximately once per 500 years at physiological temperatures (I forget the numbers for a whole cell, but it's still on the order of hours, if not days). The eye has a dynamic range orders of magnitude greater than any electronic camera I'm aware of in addition to motor tracking that is very difficult to implement in hardware. The retina also has mechanisms to compensate for motion of rapidly moving objects, which would also be difficult to implement in hardware. But the real challenge would be implementing the feedback mechanisms that your brain uses to tell your eyes exactly where to look, to focus and filter differently, etc. Not to mention depth and motion perception by circuits that integrate things like feedback from head movement with changes in binocular disparity (how different what your two eyes see is), differential movement of objects across the visual field, etc. Even if you could make something small enough that would be optically better than an eye, it would still be lagging far behind the eye in real vision because of the problems of interfacing it with everything else already there. Prosthetic eyes better than natural ones are unlikely to exist in the lifetimes of our grandchildren.
hi, I'm a neuroscience grad student, and I have to comment that basically all of the discussion going on everyone here seriously misunderstanding what this result means. I'm in a hurry, so I can't go into great depth. But to make things simple, the protein they overexpressed in those mice is involved in synaptic modification. So they made mice with more modifiable synapses. The title of the paper is "Enhanced LEARNING after genetic overexpression of a brain growth protein." Learning is not the same thing as intelligence. No one is claiming that anyone made smarter mice except the journalists. The researchers made the brain more modifiable, so the mice could learn a piece of information more quickly. That doesn't mean they could remember it as long, that they could think as well as normal mice, etc. There are a million other factors, but the bottom line is that all they demonstrated was that the transgenic mice were better at a certain task than wild type mice. This could be because they learned faster, or it could be for any number of unrelated side effects (such as ones involving changes in vision). This is like demonstrating that fighter pilots can do some maneuver 95% on cocaine and only 92% normally, and from this extrapolating that it would be beneficial for everyone to install cocaine dispensers that keep them high 24/7. This particular study has no serious implications for evolution or a society of superhumans. The relationship between genotype and brain phenotype is way too complex for anyone to come up with a one-gene solution to creating greater "intelligence." Anything like that will likely involve many genes and large-scale screening of individuals followed by correlations with some measure of the person's intellect, but I won't even start on the problems involved in a study like that...
There are a few misconceptions about evolution that I think need to be cleared up:
First, one can think of evolution as being dependent on a large percentage of the population dying. On earth, this is still true. 40,000 children die every day of preventable illnesses. 16,000 people are infected with HIV every day (which will undoubtedly cause selected populations of people with disproportionate levels of mutations in the CCR5 chemokine receptor that renders them resistant to infection through mucosal membranes). But maybe a more important thing is the idea of unborn children. In the end, the groups having the most children will win out. Evolution still occurs, just not by the same factors as 15,000 years ago.
Another thing to remember is that humans evolve/have evolved as a group, not as individuals. This is because homo sapiens is a pack animal, like ants or wolves. Long fangs and aggressiveness might help out a bear in the wild, but would probably just have gotten a homo erectus clubbed to death by the rest of the group. Traits like kindness and self-sacrifice (or propensity for religious belief) were selected for because they promoted the survival of the race as a whole. So in this sense, "social evolution" is nothing new.
A third thing, men's attraction to large breasts is based on breasts being one of the things that differentiates fertile women from prepubescent girls and other men, not women with large breasts from women with small breasts, as there is no serious correlation of breast size with ability to nourish a child (at least not a more significant correlation than there would be with other obvious markers of health). This is a subtle distinction, but an important one, I think.
As far as I can tell, he just hates Bob Gallo, the guy whose group first discovered that AIDS is caused by HIV (for those who don't know, Mullis and Peter Duesberg's whole thing is that HIV does not cause AIDS). Gallo is the guy played by Alan Alda in that ridiculous inaccurate "And the Band Played On" movie, if anyone's seen that. I've never met Kary Mullis or even looked at his data if he has any; this is just my impression from reading his web pages. I can't remember the URLs, but they're full of quotes like "People are going to know I'm right, as soon as they find out what an asshole Gallo is!" As far as Gallo, I could see why people would dislike him since he seems a little too politician. I've seen him around several times but never met him personally (yet I get gold-leaf xmas cards stamped ROBERT C. GALLO AND STAFF), but my boss and all the other faculty doing HIV stuff in my department know him really well and think highly of him. FWIW, the whole stealing data from the French thing was basically entirely fictionalized for the movie. As far as I can tell, Mullis's motivation seems to be proving that AIDS is caused by the use of amyl nitrite poppers, or ingesting semen, or a complex combination of retroviruses, or whatever they say it is now, as long as it isn't what Gallo and probably some other enemies said it was, HIV.
Sorry, I didn't see the link in the article. That would certainly decrease the cost of it, but I still have to say that the thought of doing PCR without a decent thermal cycler or even pipettes makes me cringe. My real problem with the article wasn't even the cost issue as much as the fact that the person who wrote it obviously didn't know what he was talking about, and I think it would be a lot tougher to people to get this working in their kitchen than he implied. Anyway, I think it would be much more interesting if someone took this one step further and packaged a kit that actually DID something, such as amplify a region of DNA and test it for a restriction fragment length polymorphism ("hey Dad! You're predisposed for heart disease!"). You could sell some polymerase, dNTPs, primers, and a bit of restriction enzyme to test the product afterwards. You could even put a shrink-wrapped agarose gel in there so people wouldn't have to boil ethidium bromide solutions in their microwave.