IUBio Biosequences .. Software .. Molbio soft .. Network News .. FTP

Junk DNA

Shane McKee shane at reservoir.win-uk.net
Sat Aug 5 06:52:13 EST 1995


 
In article <3voejj$400 at ixnews5.ix.netcom.com>,
 Frank Francone (francone at ix.netcom.com) writes:

>I have been involved in evolutionary computation for the last couple of
>years.  As in biology, variable length EC genomes accumulate Junk Code.
>
>Rencently, my colleagues and I looked at this phenomenon in EC and found
>some strong indications that the "Junk" code in EC genomes plays an
>important role in artificial evolution.
>
>In our model, an example of junk code would be IF 2 < 1 then . . .
>
>Our working hypothesis is that the "Junk" code plays a structural role
>in allowing the genome to direct the attention of the crossover
>operator to locations where high fitness blocks of code are not
>disrupted.  So far, ou
>r results are consistent with this hypothesis. (The work is available
>as a TR out of Univ Dortmund and will be availible in "Advances in
>Genetic Programming II" due out sometime this year.)
>
>I am quite interested in a biologist's perspective on this subject. Is
>there any evidence that the junk code in DNA plays any kind of a
>structural role in crossover, mutation?  Is there any evidence, however
>indirect, that the junk code plays any role whatsover?

It would be a good idea to read Richard Dawkins' Selfish Gene. The
basic principle is the distinction between Replicator and Vehicle.
The Vehicle is the organism in its entirety, and is recognised by
its phenotype, and on the basis of this, subject to selection. The
Replicator is the genotype, and this is analogous to your EC code.
To say that Junk DNA necessarily has a purpose 'for the benefit of
the organism' is incorrect. Junk DNA will arise by standard
mutational processes, and as long as it has no gene-competitors
that have a better chance of making it through into the next round
of replication, it will be favoured by natural selection.

I agree with Dawkins that it is time to get away from the concept
of the individual being the unit of selection, and move more
towards a gene's eye view. This should be the case in evolutionary
computing too.
 


Shane McKee (JHO, RVH, Belfast)  | /      Art becomes science when
Shane at reservoir.win-uk.net     --O--    you start trying to figure
AGACTGCGCTTGCTTTACACATTTCTTCTC / |  out what the heck you're doing



More information about the Mol-evol mailing list

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