I saw on the science instagram that Harvard scientists were able to encode a video of a horse running onto DNA that was put inside a living bacteria. Then they retrieved said video at 90% accuracy. What does this mean and how is this possible?
In: Biology
Data is just a series of 1s and 0s, on and offs. The human genetic sequence is like, and referred to often, as our genetic code. It can be read, manipulated and i guess now…. They can write to it.
Your brain and nervous system is a series of signals to communicate. Like a computer.
I’m not a professional scientist, but it probably wouldn’t be hard to encode dna in a specific binary sequence. Especially since, at least with human DNA, there is a lot of garbage genetic material in the genomethat doesn’t do anything because it’s redundant or remnants of viral DNA. Find a good stretch of that crap, splice in what you want and VOILA! Of course, it will change slightly because of mutations, but you’ll have self replicating code.
Computer code, at its most basic, is a line of 1s and 0s. DNA, at its most basic, is a line of bases (As, Ts, Cs, and Gs).
The scientists decided on a conversion (in a simple conversion, maybe A/T = 1 and C/G = 0), then converted the computer code over to DNA code. They then put this DNA into a living bacteria. Later, they pulled it back out and coverted the DNA code back into computer code. This code was still about 90% correct afterward.
As far as i know, there’s no pressing need for this technology (though it’s not my area of interest, so I could easily have missed something), but being able to do this reliably could allow us to store information as DNA, or it could help if we ever get to the point in genetic engineering where we want to create long custom strings of DNA instead of splicing a bunch of strands together.
DNA is made up of 4 chemicals (cytosine [C], guanine [G], adenine [A] or thymine [T]) and a bit of other stuff. A strand of human DNA will contain many millions (or hundreds of millions) of these.
Since DNA can be represented in each position by one of four things, each is the equivalent of two binary bits. For example, you could say that C=00, G=01, A=10, and T=11. So you can encode a binary string by breaking it up into pairs of bits and encoding each as one of the four chemicals in DNA.
For example, the binary string 0110100001100101011011000110110001101111 is the ASCII representation for “hello”. In my example, you’d parse that as
01 10 10 00 01 10 01 01 01 10 11 00 01 10 11 00 01 10 11 11. Using the encoding I described, that would be GAACGAGGGATCGATCATT (if I didn’t screw that up).
So you create a DNA strand with that sequence and it contains the information for “hello” if you decode it using the same scheme as the one you used to encode it.
We can make DNA cheaply and at large scale.
We can sequence DNA cheaply and at large scale.
All they did was follow those two steps. DNA can be thought of as information, like information you would find in a computer file, so they just copied it.
What this means is just scientists showing off. Maybe can be useful for long-term, large scale data storage.
Will this lead to something where we can store petabytes of data cheaply by using DNA? Who knows, maybe, but we’re nowhere near that point, likely that would be decades away if even economically viable at all.
Latest Answers