Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utgpu!water!watnot!watmath!clyde!rutgers!lll-lcc!ptsfa!well!msudoc!umich!itivax!m-net!michael
From: michael@m-net.UUCP
Newsgroups: sci.bio
Subject: Re: question - DNA's information
Message-ID: <1146@m-net.UUCP>
Date: Mon, 6-Apr-87 02:43:23 EST
Article-I.D.: m-net.1146
Posted: Mon Apr  6 02:43:23 1987
Date-Received: Sat, 11-Apr-87 06:21:53 EST
References: <11189@teknowledge-vaxc.ARPA> <978@aecom.UUCP> <3310@udenva.UUCP> <1534@husc6.UUCP>
Organization: M-NET, Ann Arbor, MI
Lines: 45
Summary: Yet another answer - what did you mean by that question?

In article <11189@teknowledge-vaxc.ARPA>, rburns@teknowledge-vaxc.ARPA (Randy Burns) writes:
> I was wondering roughly how many 'bytes' of information are contained
> within human chromosomes?
and gets large number of replies (some with novel definitions of bits
and bytes).

Some of the replies made the assumption that only the information that was
eventually transcribed into protein was significant, and that repeated
sequences, introns, the third base pair of some three-letter codes, and so
on, could be ignored.  I must take issue with that.

Introns may yet prove to have functions beyond their own excision.
(Negative feedback on protein production is an obvious candidate.) 
Some of the regions near protein-coding sections regulate expression.
The third codon for the hypervariable part of antibodies is significant
(affecting the potential antibodies after DNA editing), and may have
other effects in other genes (i.e. affecting the likelyhood of cancer
by changing the probability that a growth-regulator will mutate into
an "always-grow" form).  Repeating sequences may function in DNA repair
mechanisms.

Because of these and other possible functions of DNA, I would not exclude
any information from consideration, and would answer with the number of
8-bit bytes required to specify the reconstruction of the actual DNA found
in an individual.

At 3*10^9 codons, and two bits per codon, that's 3/4 * 10^9 bytes, or
about 750 Megabytes.  (Specifying cutting points to separate the
chromosomes, and other artifacts like the arbitrary choice of direction
of the chromosomes and their order in the representation, gain or lose
so few extra bits they're lost in the one-significant-digit estimate of
the number of codons).  We're starting to see disk drives about that
big.

But this number really represents the maximum data that the genome
could hold.  Some of the redundancies could be absorbed in data
compressing codings, squeezing down (but not quite eliminating) the
data representing repeating regions, but not affecting third-codon
redundancies.  Sorry, I can't put a number on that.

  "I've got code in my node."	| UUCP:  ...!ihnp4!itivax!node!michael
				| AUDIO: (313) 973-8787
	Michael McClary		| SNAIL: 2091 Chalmers, Ann Arbor MI 48104

(If you want to be sure I see it, MAIL to the address in my SIGNATURE!)