Neohapsis is currently accepting applications for employment. For more information, please visit our website www.neohapsis.com or email hr@neohapsis.com
Crypto Archives: DNA computing & secure envelopes

DNA computing & secure envelopes

Jim Choate (ravageeinstein.ssz.com)
Mon, 4 Oct 1999 22:41:40 -0500 (CDT)

----- Forwarded message from David Honig -----

Date: Wed, 29 Sep 1999 12:28:06 -0700
From: David Honig <honigsprynet.com>
Subject: Re: [honigsprynet.com: Re: secure time-stamped envelopes]

At 11:07 AM 9/29/99 -0500, Jim Choate wrote:
>> Travelling salesman problem. The DNA computer described
>> in _Science_ about a year ago solved a toy TSP problem.
>Which is related to the cracking of DES via a DNA computer or the recent
>work in factoring (via the URL I supplied) exactly how?

You said:
Subject: Flourescent DNA computing [sciencedaily.com]
    From: Jim Choate <ravageeinstein.ssz.com>
    Date: Sat, 25 Sep 1999 08:25:41 -0500 (CDT)


Duke Chemist Describes "Fluorescent" Method For
Computing With DNA

Both Adleman's TSP solver and this chemist's SAT solver
are parallel string matchers. The chemist is combining
the fluorescent probes, DNA string matching, and combinatoric
chemistry to generate the possible solutions.

As such, they need at least one molecule per possible

Any real (vs. toy) problem will have more than one Mole of
searchable strings. Much more. A Mole of DNA weights (hundreds of)
thousands of grams (depending on the DNA length = molecular
weight). And the DNA is not reconfigurable for a different

----- End of forwarded message from David Honig -----

Sorry for the delay, I didn't forget - been very busy.

First off, go back and re-read what I posted. I did *NOT* imply in any way
that what was required to make a time stamped envelope system using DNA
was the implimentation of a crypto system in DNA. Though one could
certainly do that. What I said was that a very fixed rate and reliable TIMER
could be done in DNA. In addition that time could at the time of expiration
be programmed to do all kinds of things such as kick out two particular
protein chains, only one of which you know about up front and would be the
taggant that would be used on the detector to indicate a valid result. One
could then do a spectrograph of the other protein, convert it back to the
DNA that produced it, then convert that DNA sequence back to numerics and
wallah, you've got your key to the external, non-DNA, crypto system that
would unlock the 'envelope'.

Now let's look at your numbers regarding the DNA [1],

The typical base pair is 6.10 g/mol (28g/avoirdupois ounce, 16 avoirdupoise
ounces/ lb.) or 6.10 daltons.

So, let's look at DNA sequence of 10,000 base pairs, that would be,

10E3 pair/mol * 6.10 g/mol = 61E3 g


136.2 lbs., heavy yes. Is this something you're going to do for some trivial
issue? Not likely. Is it doable, you bettcha. Is purchasing this amount
economicaly problematic, not likely from the pricing I see on the webpages I
checked. Hell, it wouldn't be that problematic to use something like yeast and
the apoptosis system. Not only would this be cheaper but it would provide a
longer and more accurate (e.g bamboo flowering periods and global synchronicity
of same) time reference standard (e.g. circadian rhythm).

As to being reconfigurable, so what. Not relevant in this application. We
only need the timer to work once.

In a related vein, the SciAm site has an interesting article on the ribosome
and RNA.

           Error of opinion may be tolerated where reason is left
           free to combat it.

                                            Thomas Jefferson

       The Armadillo Group ,::////;::-. James Choate
       Austin, Tx /:'///// ``::>/|/ ravagessz.com
       www.ssz.com .', |||| `/( e\ 512-451-7087
                           -====~~mm-'`-```-mm --'-

[1] http://www.cbs.dtu.dk/nikob/science/nuc_table.thml

This archive was generated by hypermail 2.0b3 on Tue Oct 05 1999 - 00:43:01 CDT