Better Living through DNA Encryption

Peter Alexander Merel pete at cssc-syd.tansu.com.au
Sun Jul 19 09:30:03 EST 1992

I have had what I think is a pretty wild idea, and I expect someone to shoot 
it down. If no one can shoot it then I'd like a Nobel Prize please.

In order to understand the idea you must understand a little about DNA
transcription, and a little about Public Key encryption. If you already
understand a little about these things then please skip over the next
two paragraphs.  If you understand a lot about these things then please
bear with me, because I only understand a little.

DNA trancription:

DNA transcription is a cell's way of making useful machines, called
proteins, from the genetic information in its nucleus. The way that it
works is quite involved, but the basics are: 1) a bunch of enzymes (a
sort of protein) inside the nucleus transcribe meaningful sections of
the DNA (called genes) into RNA; 2) A bunch of enzymes (called a
ribosome) outside the nucleus transcribe the RNA into proteins; 3) these
proteins fold up into useful shapes, and everything in the cell except for
the RNA and DNA molecules is made from them.

Public Key encryption:

Public Key encryption is the most secure form of data encryption
known.  The way it works is quite involved, but the basics are: 1) two
keys are generated by a clever algorithm with a random seed; each key
is a function that transforms a stream of bits into another stream of
bits; 2) if one function is run on the output of the other (the
ciphertext) then the original stream of bits (the cleartext) is
restored - that is, these two functions are inverses of eachother; 3)
it is computationally impractical to generate one key from the other,
no matter how much cleartext (input) and cyphertext (output) one has
for a given key; this is to say that it would take the fastest computer
imaginable longer than the expected lifetime of the universe to do so;
4) one of these functions is called the public key, and the other is
called the secret key; it doesn't matter which is which so long as one
is secret and the other public.

My idea:

My idea is to use Public Key encryption techniques to protect human DNA
from the effects of virus infection (AIDS, Hepatitis, Herpes, the
common cold ...) and random and induced mutation (Cancer, Ageing, ...).
There are two processes required to effect this encryption process,
which I will deal with in reverse order.

Let us say that a secret key has been used to encrypt all of your DNA.
Without the public key this would render the DNA useless - it could
still replicate when your cell divided, but no useful proteins could be
constructed from it.  So, let us also say that the enzymes that build
the RNA from the DNA have the public key built into them. Rather than
just transcripting the DNA directly into RNA, they decrypt the DNA with
the public key and then transcribe it. Because the DNA has been
previously encrypted with the secret key the end result is that the
proteins it codes for are constructed as they should be.

How does this stop viruses? Assume that a virus invades your cell. It
does this by patching your DNA with some of its own DNA, and then
waiting for your enzymes to transcribe that DNA into machinery that
makes more viruses. If we can assume that the virus has no access to
your secret key, then there is no way that it can contain DNA that will
code for anything in particular when decrypted by your public key, and
it will always be decrypted by your public key before your cell
attempts to make protein from it. Therefore no virus can use your cell
to replicate itself.

How does this stop cancer?  Let us tag each gene in the unencrypted DNA
with a distinctive signature - say, a cyclic redundancy check - before
we encrypt it. Let us now introduce a mechanism such that if a gene is
decrypted with a bad CRC then the cell self-destructs.  Maybe we could
get this mechanism to try the check twice before destruction, but it is
probably safest to do it just once. Now when a mutation occurs within a
gene, the gene will decrypt with a bad CRC, and the cell will die.
Therefore there will be no successful mutations, and so no cancer.

How does this stop ageing? It might not, but one school of thought says
that ageing (as opposed to maturation) is chiefly the result of
accumulated transcription errors in genetic material; if this is so
then no mutations means no ageing.

Sex and the Secret Key:

Obviously we'll want to protect the secret key pretty well. The best
way to do this is to destroy it immediately after using it. After all,
we only need to use it once to encrypt the DNA and generate the public
key for the nucleic transcription enzymes; this must be done at
conception. After that it would only be a security liability to keep the
secret key around.

I figure that key generation would work a little like this: the sperm
and the egg contain only unencrypted DNA. We could arrange for them to
carry cipher DNA and a copy of their public key, but that doesn't seem
to me to improve security since we're going to generate a new secret
key anyway.

Upon fertilization the unencrypted DNA from the sperm and the egg is
combined in the usual way to form an aggregation of unencrypted genes
sufficient to make a human grow and function. At this point a new
secret/public key pair is generated by the public key generation
algorithm using a random process for its seed (lots of those about at
this scale). The secret key is used to encrypt the DNA of the new
human, and the public key is used to generate appropriate nucleic
enzymes (DNA decryptase?) for this DNA.  The secret key is now
zealously destroyed and the resultant embryo develops pretty much as
per usual.

The result of this process is that every human on the planet will have a 
different encryption key for their DNA, so even in the extremely unlikely
event that some warped genius manages to design a virus that can successfully 
patch the DNA of one human, it won't be able to affect anyone else.

My Nobel Prize:

Oh, just plonk it in the mail. No, wait on, there's money there, right? I
better pick it up in person. On the other hand ... it'll be thirty years or
more before the above becomes feasible. In thirty years I won't care about the
money - I might get a kick out of refusing to accept the Nobel. Do they offer
it to you if you've put it about that you might not accept it? Okay, tell them
I'll definitely accept it. No, really.

I should hasten to consider that if no one shoots it down the above
could be worth a whole lot of money one day. Well, to hell with it. I hereby
place the above ideas in the public domain, with the proviso that they be 
known by a silly name like "Pete's Panacea" or "Merel's Medicine" or some
such. You get the idea.

Internet: pete at cssc-syd.tansu.com.au   UUCP: {uunet,mcvax}!munnari!cssc-syd!pete
Snail: 1/18-20 Orion Road, Lane Cove NSW 2066 Australia    Phone: +61 2 911 3130
Consistency and Pedancy make for Genius. Um, Pedantry.

More information about the Ageing mailing list

Send comments to us at biosci-help [At] net.bio.net