Nope, that's no longer possible on such a brain. What you do is you inject special tracer substances (while the mouse is still alive). These substances will stain neurons at the site of injection and then cross the synapse to connected cells, either in the direction of the information flow or opposed to it, depending on the system used. These tracers are then imaged using the method that this article is about. To further aid you, you can do different stainings to see what type/subtype of neurons you are looking at. By combining this information with known functional properties of the found neuronal types, you can try to infer what is actually happening in the area you examine.

True, but the level of transparency wasn't that impressive with that method, it only worked up to 1-3mm of depth. BABB based protocols were a lot better in that regards.

Hell yes, that's the big one here. Plus, expressed fluorescent proteins in the tissue don't get degraded as much as with BABB et al. Definitely give it a shot, you probably have all the ingredients around the lab anyway. The clearing is done with PFA, acrylamide, bis-acrylamide, VA044 and PBS. The slices should then be immersed in glycerol, so nothing special there as far as I can see it. You only need to build a custom electrophoresis chamber to stain the brain, but even that shouldn't be too hard.

No, I was thinking of something still lethal but less freaky: cancer. Plus, even if one of the patients goes insane for whatever strange one-in-a-billion chance, it's not really infectious unless he's still capable of drilling a hole into your skull and injecting a tiny amount of purified virus into precisely the correct area of the brain (think micrometer precision). So no zombie apocalypse there, sorry.

True, but I would say that a few lost muscle cells are less problematic than a few neurons lost in the wrong part of the brain. AFAIK there is no reliable way to control the site of lentiviral integration. Plus, purifying lenti properly is nasty, the stuff can be either quite neurotoxic or not infectious at all if something goes wrong during that step.
Recombinant Adeno Associated Virus is much less problematic, it's dead simple to manufacture and only the potential protein overload problem remains (and in mice we're using them a lot without any apparent problems). However, in an adult brain, the effect of rAAV is only temporary since it doesn't integrate and gets degraded over time.

Disclaimer: I work in neuroscience and have used viral transfection quite a lot.

Myth? It's not trivial to get the infectous titer and purity of the virus right and it's even harder (read: almost impossible) to predict the exact expression levels that the virus will cause in an actual brain. Much less if such a potential overexpression of a non-native protein will mess up regular cell trafficking/function. Even if the protein is thought to be harmless (as is the case with Channelrhodopsin or Halorhodopsin), the sheer fact that the cell now has to produce, store and process large numbers of something it usually doesn't have can cause problems and take resources away from the normal function. Plus any virus that will stably integrate into the genome can cause all kinds of fuck up down the road since you don't know WHERE it will integrate and what other function it might overwrite.

Don't get me wrong, it is interesting, it is potentially very beneficial but I'd still be cautious when applying it in the brain (as opposed to applying it in muscle or skin cells) since adult neurogenesis isn't really happening much...

As interesting as ever:
Adam Curtis - The Power of Nightmares
It gives an interesting view of how this whole situation with evil states and global terrorist networks spiralled out of control.
If you have three hours of time that is... But it's definitely fascinating imho.
http://www.youtube.com/watch?v=xGo1DqmfHjY
http://www.youtube.com/watch?v=o0kNNqZk3mg
http://www.youtube.com/watch?v=qATc5jRbVOA

It is indeed possible to print IN stainless steel, titanium etc.... Using a technique called selective laser sintering, fine metal powder is selectively melted/fused by a high power laser, allowing you to directly print custom parts from metal.

Simple solution: They shouldn't try to fool people into thinking they are actually BUYING the game. Rename it to say "license the game" or "rent for an unlimited time" or whatever. I'm fine with their non-transferable model as long as they do not try to tell me I'm actually buying the game. Because if I buy something I expect to actually own it and be able to give it away anytime I want.

What is the situation with the source/GPL?
"Any code touching the user interface created within this endeavor will be licensed under the GNU General Public License Version 2 or later (GPLv2+), possibly with an exception for the Windows Store if needed."
I remember vaguely that there once was a VLC for iOS around before some internal debate about whether or not this sort of port was acceptable with the GPL caused apple to remove it. Exception for Windows Store? How should that work out then?

Germany has them as well for landline, it's actually the standard option
for mobile, most german providers will slow you down to GPRS speed after a certain amount of data consumed per month (anywhere between 100MB and 30GB depending on the contract)

Mod parent up, it's spot on! The actual patching is not really hard, especially since it only seems to do "blind" patching (just guided by the electrical response).

The only things to add are
- Fighting with your pipette puller to deliver constant results, because either you have your own crappy little puller that is hideously unreliable or you have a fancy one which is shared, meaning there is a high risk of someone messing with the filament and/or settings. And if your pipettes are crap, neither you nor the robot are going to succeed...
- Finding out why you suddenly have massive noise in your recordings
- Not freaking out after matlab, igor et al crash halfway through your recording session taking all your data with it.

Anyone wants to bet how long it'll take for "Pay 5000$ now or you'll hit that nice tree over there" style rootkits to appear? Scareware indeed...

5 months ago? Vodafone Germany will START to sell them "in a couple of weeks". I'm sure it will be really popular here...

It doesn't make a difference, anyone who wants his/her paper published will give the 1000 dollars (because they have to). And either way, I don't think science and nature will run out of people wanting to publish there...
But I fully agree that open reviews is a bad idea. In a field where almost all players know each other it would really hamper objectivity if you knew the person whose (bad) work you just shredded might come back to bite you.

But then, in some ways they still have a classic view:
"Flame is described as enormously powerful and large, containing some 250,000 lines of code, making it far larger than other such cyberweapons." [...] "But this new weapon is twenty times the size of earlier cyberbombs and far more powerful, making it practically an army on its own, said Roel Schouwenberg, a senior security researcher with Kaspersky Labs."

Wait until they finish the TerraBomb, with it you can overload many a computer simply by copying the WEAPON to the HD. It doesn't even need to be launched (though you might need to copy it once or twice on machines with a larger HD...). Only downside is that the attack takes a considerable amount of time to carry out.