Bill Gates Calls for Crazy Ideas

So I saw a post by Bill Gates on LinkedIn, calling for crazy ideas. It linked to an organization calling for solutions to three new problems and a dozen others. They offered grants, which I have no interest in, nor did I have suggestions to the health problems they suggested.

However, I have some other crazy ideas:

One is for protein folding algorithms and another for possible RNA-based HIV treatments. Probably just pipe dreams though, but I commented on the article anyway. They’re probably not wholly original, nor do I have any means of pursuing them.

(1) “Natural” Protein Folding Algorithms.

Protein folding algorithms that attempt to completely and realistically recreate the protein creation algorithm, including pH, ribosomes, amino acid availability, codon availability, and intracellular brownian motion. I noticed playing FoldIt that many of the puzzles started with proteins that were either a straight chain or partially folded, probably at a bad local minimum.

But, if this process is reconstructed as accurately as possible, wouldn’t the correct protein shapes simply be reconstructed? The process of constructing a protein will introduce forces and other factors which would greatly affect the resulting shape of a protein. So at least, wouldn’t a more natural algorithm significantly limit the outcomes?

The most important question:

If the probability of having these resulting shapes is statistically significant enough to be reliable for life itself, then why do we have such a problem finding them?

(2) Epigenetic & Novel HIV treatments

RNA-based HIV treatments that hack the transcription factors and epigenetic factors that HIV inserts into white blood cells. In other words, could a treatment similar to RNAi be used to prevent RNA polymerase from binding to HIV-specific transcription sites?

Does HIV have a mechanism for “turning up” it’s own genes by altering the attached epigenetic factors? And if so, how does it attach those epigenetic factors, how does it mark them to ramp up their expression, and can we safely manipulate those processes? Or, perhaps we can interfere with the viral RNA that’s emitted from a cell’s nucleus when the it expresses an HIV gene. Or maybe there’s a way to set up “HIV Honeypots,” harmless vacuoles with surface proteins that trick HIV into attaching and neutralizing itself.

Update 2/19/15: From a high level, this HIV Honeypot strategy is very similar to the promising novel HIV treatment just announced this week!

“We took our clue from existing proteins that HIV needs to get into cells. So we took pieces of one of those proteins and we added it to a piece of an antibody and we added it to something that looks very much like a second protein that HIV needs. And the combination works much better than either the components or than any antibody.” - Dr. Farzan on WSJ Video.

It’s important to note …

… That I have no idea what I’m talking about. lulz. These are just some of the problems tumbling around in my head that I like to think about. I wish I had the tools available to answer these questions.

Or the time to complete the relevant classes on Coursera. Ugh, I get so frustrated when I turn in something late on that site and my grade drops to a B/C, especially when it’s a class where I know the material and deserve an A.