James Hamlin

How much of the human genome is considered "junk" or unused information? There must be some, correct? I believe I read something about this in a textbook a while ago.

I feel that this would be an interesting argument against creation.

Doubting Didymus

It depends largely on how you define junk. There are a couyple of different definitions, though the amount is staggering every which way.

Talkorigins gives me a couple of figures: 90%, 95% and 97%

My Campbell Biology, 6th ed. tells me that 97% of human DNA is [i]non-coding[i]. That isn't to say that it has no effect at all: some genetic disorders arise from non-coding DNA.

You are right that this is an argument against creationism. It's not exactly the soul of efficiency.

James Hamlin

There we go, that was my textbook, Campbell Biology. I knew the number was high. Thanks for the help, Doubting Didymus.

pz

The variability in figures arises because a) there is still some flux in precisely how many genes there actually are (the range was 25-40 thousand a few years ago, although it is fast converging on about 30 thousand), and b) whether the author is reporting coding sequences only or are also including regulatory regions, which roughly doubles the functional size.

90% is a conservative estimate of the amount of nonfunctional junk. Heck, 50% consists of various repetitive transposable sequences.

Oolon Colluphid

Hmmm. I had this thrown at me by a creationist recently:

Faye Flam, Science 266:1320, 25 Nov 1994

(That is the ‘stable URL’ for citation, and apparently one may copy it into other web pages. Hope this means it’s okay here.)

(Edited: bugger, doesn't seem to work. Well it basically refers to some research indicating that the patterns in junk DNA are rather similar to human language in terms of Shannon-esque information, or something (have only scanned it, I prefer to read stuff on paper, and it's too old, so a TIF, so can't copy & paste any of it. I can send it to anyone interested.)

So junk DNA is a language all its own... though no-one knows what it says. This looks fairly speculative, but do others have any thoughts on it?

Cheers, DT

Wounded King

I looked at a related article

Mantegna RN, Buldyrev SV, Goldberger AL, Havlin S, Peng CK, Simons M, Stanley HE.
Linguistic features of noncoding DNA sequences.
Phys Rev Lett. 1994 Dec 5;73(23):3169-72.

They say
Does a sequence which goes CGCGCG....etc or suchlike constitute one with lower entropy, the very regular pattern of the repeats, and larger redundancy, the high frequency of the repeats?

Actually having looked at the sequences they used they seem to be looking at non coding regions associated with genes rather than simply intergenic regions.

Then again my CG repeat sequences do still show up within gene containing regions, Im somewhat confused now.

A later paper suggests that this doesnt really mean much biologically

Chatzidimitriou-Dreismann CA, Streffer RM, Larhammar D.
Nucleic Acids Res 1996 May 1;24(9):1676-81
Lack of biological significance in the 'linguistic features' of noncoding DNA--aquantitative analysis.

But it doesnt say whether or not they think it contains hidden messages from god, how remiss of them.