One Scientist’s Bid to Build Cancer-Killing Nanorobots

We already have this - they are called bispecific monoclonal antibodies [0]. If you gave me the job of curing Cancer (and the resource to do it) then this is what I would spend your money on.

[0] http://en.wikipedia.org/wiki/Bispecific_monoclonal_antibody

That turns out to be the focus of a lot of research right now[1][2]. The computational aspect of that has a lot to do with

    1. faster, more accurate analysis of genetic code
    2. faster, more accurate simulation of interactions 
       between candidate peptides, binding sites on cancers, 
       and the cells of the immune system

[1]: "Designing immunogenic peptides" Darren R Flower

[2]: "Cancer immunotherapy comes of age" Ira Mellman et al

Precisely this formulation of "nanorobot" is mentioned in one of the Venus Prime books. Basically, they're very complex enzymes that can be regulated by chemicals and each other, mediated by the protagonist solving systems of differential equations with a supercomputer she screws on the end of the kitchen faucet. (So it can dissipate heat by producing gobs of steam.)

It's pretty much the very edge of modern synthetic biology. Repurposing biological materials and tools to entirely designed uses. A much more realistic approach to atomic-resolution engineering than shrinking our metallic macro world.

I highly recommend Regenesis by George M. Church: http://www.regenesisthebook.com/

He's one of the leaders of this field and the book is a great intro to what's happening, future developments, pitfalls, etc.

What parts would you like to learn more about? Shawn's construction is a pretty straightforward (and brilliant) combination of 3D DNA origami and DNA aptamers. The paper can be found at https://dl.dropboxusercontent.com/u/5290041/publications/223... ; the best way to learn about something like this is often to look at the paper, and then work backwards for the references. If you don't have journal access, lab websites usually have copies of their papers on them; I know our lab (dna.caltech.edu) has our papers there, including Paul's original DNA origami papers.

This paper is actually a bit annoying in its focus and tone regarding the field. DNA origami is actually surprising simple; the design is not at all "as complicated as it sounds." Short, synthesized strands of DNA that are complementary to various bits of a long scaffold strand of DNA (often single-stranded plasmid DNA, like M13mp18) in a manner similar to the way one pushes different faces of paper together with paper origami. Details of design can be tricky, but it really is something that can be done by hand. CADnano is great from a convenience level, not from some fundamental necessity level.

And the major challenges facing this particular technology really aren't, in my opinion, the cost of synthetic DNA. Instead, as many people have pointed out since this was first presented several years ago, the major problem with in vivo use is that nucleases are everywhere, and it's really expected that things like this will get entirely ripped apart in, for example, mammals. Even if it isn't ripped apart, it needs to not end up with stuff stuck around to it to the extent that it won't operate. Without some way of protecting the structure, it's not going to work well.

Still, this has long been the example I've used of an application of our field.

then we are at our current solution, chemo.

hey why the two points less!! the biochemistry that takes place in our cells is kind of a hack no? So if we invest more in the cannabis investigation maybe we could check why the CDB in the sativa strains helps in the apoptosis process which in turns destroys the bad cells. This is more natural and cheap than a lot of robots inside ourselves. or I'm wrong?