Kinda interesting biological computing

[angupesa0611]

Hey,

 

I'm a physics undergrad and I'm quite passionate about my field of study, systems biology.

 

It really isn't that popular of a branch of physics at all, but being a tech nerd I like the prospects it might have in the future and wanted to share it with others who are just curious of another type of computing.

 

As you know we currently have two paradigms for computing, binary computing (the one you are most likely using rn) and quantum computing (the qbits one). Quantum computing has mathematical advantages in processing data to to the sheer amount of informationa qbit can manage, but it's quite flimsy and kind of a jerk because it require veeeery especific conditions to be avoid decoherence. Where does sytems biology come in? Well I'm working on a thesis and research project that may create a third alternative. Using genes you can create something like a bit! and even logic gates! Let's say some condition make a certain gene in a DNA pool get translated into RNA then into a protein. Lets say this condition is a voltage. That created protein, may prevent the translation of itself, that's a NOT gate. What if we have three genes, A, B and C, that make their respective proteins, and A and B have some arbitrary condition to start translation. You can tailor gene C so that it will only be translated (it's condition for translation, in other words) is the presence of protein A and B. That's an AND gate. What if on this trigene system we make it that so that gene C can be translated with either the presence of gene A or B, well clearly that's an OR gate. With these three fundamental gates you can create all other gates, so even a computer (in theory).

 

This all just sounds like speculation, but its very posible and not even new. Look up promotors and inhibitors of molecular biology and you will see. The real issue is that it's not all practically inmediate (like electromagnetic based systems). A generated protein needs to reach the place were it will do it's work after being created, but proteins drift in the water environment they are in. This is an stocastic problem then, making this processes and its timing quite random. However, with some professors we noticed that the stocatis nature, on average can be generated so it looks like a quantum wave function. Not one from any known particles, but still similar. Thus, if you operate the wave functions, not the end binary results, you would be doing uantum computing. The conditions of the space to make this happen? well, just a livable environment for protein genesis and one that doesn't denaturalizes them. Wich means... a human livable environment pretty much! quantum computing without low kelvin temperatures or high energy systems! Sure it sounds cool, but its limitations? well, protein drift, even on average, is quite slow compared to electromagnetic phenomena (I mean, of course), but if this slow speed can still beat on computing power a regular binary machine due to its quantum nature it might be worth it. Another limitation are the genes themselves. For this to work you would have to need to model the gene, its ARN counter part and protein counter part quite well. But currently we use already in nature proteins which are long and really complex, making the modeling quite hard. IF, that's a big if, we could make a couple of simple genes (one creting a promotor and another and inhibitor) that could be slightly altered so different proteins don't overlap on jobs, we would have better chances at modeling.

 

Currently I'm working practically in seeing if the second limitation is even possible to solve and also doing a theoretical job on the first one supposing it is possible to see if its even worth it.

 

Just wanted to share with you this possibly (hopefully) new way of computing, let me know your thoughts :)