Eye on Innovation: The Nano-brain

How was the Nano-brain created?

We were trying to grow a monolayer of the organic molecule duroquinone, and it was basically an accidental discovery when I saw that assembly for the fi rst time on a gold surface. I saw one circular ring, and whenever I tried to beam it with the [scanning tunneling microscope (STM)] tip—a sharp needle with a nanowire tip—it moved as a whole. What you normally see with the STM characterization process is a class of molecules that move apart when you scan them. In this case, they always moved together on the surface as a stand-alone system. Knowing this, we tried to control the central molecule, and we found that all the surrounding molecules changed simultaneously. Later, I developed a way to build it by inducing a self-assembly process, as described in the Proceedings of the National Academy of Sciences paper.

How does the Nano-brain work?

For medical and other applications, you need to hide the architecture inside an enzyme or other organism that is active in the biological systems. Th ere are several biological systems in our body that can supply multiple electrons to the system. If any of these are added to this particular assembly, then in principle, they can work as a stand-alone system. Also, that is essential to avoid antibodies that attack any foreign bodies. When we started working, our initiative was to only oversee computational aspects of this, but later we found it had more immediate applications in medical science. Th e current structure is suffi cient to attach to an enzyme and study by injecting it into the body of a rat, or other animal. So, we are discussing testing this kind of medical application in the near future. We are also now building a much-more complicated structure; our recently published work only demonstrated the computational advancement. Later, we will demonstrate the commercial version of this structure.

What applications does the Nano-brain have in medicine?

If we want to show the scientifi c community real evidence that this can work, we need a diskette structure, but we can easily create a three-dimensional, nearly spherical assembly that is a generalized version of the disk-shaped assembly. For the medical applications that we are planning to execute, all are in the 3D architecture, because a three-dimensional assembly is much easier to study in the human body. We first had the problem of hiding the structure from antibodies, because when we inject it into the body, the antibodies attack this foreign system, and it eventually loses its functionality. We have taken care of this using multiple, redox-active biological organisms.

This molecular system sparked a lot of other molecular machines invented by other scientists, because prior to us, there were several molecular machines that could do the specifi c job of a factory of microscopic doctors, but they didn’t have the brain. We just basically supplied the brain. Our brain as a stand-alone is nothing without their machines. So, when we connect the two, they should go to the right place [in the body] and carry out a series of multiple instructions, such as those to destroy a tumor. One set of instructions for the system could be encoded before injecting it into a body. But that is insuffi cient for carrying out a complete operation.

A concern of ours is that we have no control after [the assembled molecule] is inserted inside the body. Sending wave signals to guide a remote body is not a practical answer, though it may appear exciting. The assembly should have a sequence of logical operational capabilities encoded inside it. We’re working to incorporate multiple logical assignments into a 3D-version of this assembly. We will send one instruction to the assembly, and we will observe it continuously and see that it is doing a defi ned set of work for a preset time—5 seconds, 10 seconds—and then another set of instructions will automatically generate and execute, and then automatically generate another set of instructions to be executed. If we succeed in doing this, the Nanobrain will be ready for real testing, because then it will be able to carry out a real set of operations.

When will the Nano-brain be commercially available? We currently have a 16-bit assembly, with the ability to carry out 16 times more operations than a normal computer transistor. Using this molecular architecture, we can multiply this assembly by a particular protocol, and we can increase by the power of two (16, 32, 64, 128). Presently, we are working toward an assembly of 1,024 duroquinones, with the ability to carry out 1,024 instructions simultaneously. By late 2009, we will be able to finish 1,024. Then it will take three more years to reach 10,000. It’s timeconsuming, because one small group cannot see all aspects of the work; we need feedback from the scientific community.