Freethought & Rationalism Archive

ps418 · 05-08-2003, 07:02 AM

There is an interesting paper in today's Nature. One of the authors many of you will recognize: Robert Pennock. I am sure alot of people wil recognize Lenski's name as well. The article is about a computer simulation (using the program Avida) of evolutionary change, particularly the evolution of "complex functions" by mutation and selection. Looks interesting.

The press release:

Artificial Life Experiments Show How Complex Functions Can Evolve

The abstact and ref:

Quote:

A long-standing challenge to evolutionary theory has been whether it can explain the origin of complex organismal features. We examined this issue using digital organisms�computer programs that self-replicate, mutate, compete and evolve. Populations of digital organisms often evolved the ability to perform complex logic functions requiring the coordinated execution of many genomic instructions. Complex functions evolved by building on simpler functions that had evolved earlier, provided that these were also selectively favoured. However, no particular intermediate stage was essential for evolving complex functions. The first genotypes able to perform complex functions differed from their non-performing parents by only one or two mutations, but differed from the ancestor by many mutations that were also crucial to the new functions. In some cases, mutations that were deleterious when they appeared served as stepping-stones in the evolution of complex features. These findings show how complex functions can originate by random mutation and natural selection.

Lenski et al, The evolutionary origin of complex features. Nature 423, 139 - 144 (2003).

Patrick

RufusAtticus · 05-08-2003, 07:25 PM

Thanx for the heads up, I'm downloading the paper now.

RBH · 05-08-2003, 08:42 PM

For those who don't have a subscription to Nature, the full text and supplementary materials are here.

RBH

GunnerJ · 05-09-2003, 06:13 AM

Creationist: It's just a computer program!

(Biologists show how living things in nature evolve complex fetures)

Creationist: It isn't science if it isn't done in a lab!

(Biologists preform an experiment theough induced mutations over many years that causes the evolution of novel complex features)

Creationist: It isn't evolution if it doesn't occur in nature!

(Biologists shoot themselves in despair for the fate of the human race)

Principia · 05-09-2003, 07:36 AM

Or an IDiot:

Quote:

Dawkins' "METHINKS" program this IS...

As a friend of mine from Europe pointed out, all this simulation does is show that genetic algorithms can perform a simple hill-climbing exercise. A smooth pathway towards a desired target. Not very awe inspiring from the computer science side of things.

"simple hill-climbing exercise" LOL.

Peez · 05-09-2003, 08:57 AM

Quote:

Principia:
"simple hill-climbing exercise" LOL.

Climbing hill improbable.

Peez

Dr.GH · 05-09-2003, 09:41 AM

Thanks for the link, Patrick. And, yeah, Gunner, you got it right.

Gary

Non-praying Mantis · 05-09-2003, 12:17 PM

Quote:

Originally posted by GunnerJ
(Biologists shoot themselves in despair for the fate of the human race)

Of course, the more level-headed biologists would just shoot the creationists, instead! - adapted from Douglas Adams

NPM

RBH · 05-10-2003, 01:40 AM

I've now read the paper half a dozen times and looked at some of the Supplemental Information supplied. No matter how much ducking, bobbing, and weaving IDists do, the central finding of the paper is unmistakeable, regardless of naturalistic assumptions, Darwinian zealotry, or funding source:

1. Start with a population of competing replicators.
2. Add random mutations of biologically known kinds occurring at rates comparable to those that occur in biological populations; a fitness function; and selection against the fitness function.
3. Let the population evolve by itself (with no "intelligent agency" intermittently intervening) for 100,000 cycles.

Results: In nearly 50% (23/50) of the evolutionary runs, each starting from a different randomization, an assembly language computer program that meets the technical requirements of Behe's definition of "irreducible complexity" evolves. And the entire lineages of those irreducibly complex programs can be traced all the way back to the ancestral population, showing that they evolve by regular Darwinian incremental step-by-step evolution, using such exotic concepts as point, insertion, and deletion mutations, and cooption. Even epistatic interactions emerge step by incremental step, just like in biology. And just like those benighted Darwinians said it could happen. And it ain't even rare!

There are other interesting findings. For example, the irreducibly complex programs that emerged from the evolutionary runs had different lineages from run to run - lots of different evolutionary pathways got the systems to the same phenotypic outcome. And the role of mildly deleterious mutations in 'setting the stage' for later fitness gains is real interesting. Contrary to Leo's [on ARN] naive belief, natural selection doesn't instantly remove each and every deleterious mutation the instant it appears. I'd suggest Leo learn some evolutionary biology rather than guessing at it from the simplistic misrepresentations of Wells, Behe, Dembski, et alia.

Irreducible complexity is dead on the ground, no longer even twitching feebly. It's time to give the mousetrap Last Rites and bury it beside phlogiston, N-Rays, and cold fusion.

RIP: IC, CSI, and SC.

RBH

JonF · 05-10-2003, 02:21 PM

Naw ID's still alive and viable.

There was an intelligent entity within 1,000 miles of the computers on which this work was performed, right?

Therefore CSI was smuggled in!

05-10-2003, 01:40 AM	#9
RBH Contributor Join Date: Aug 2002 Location: Ohio Posts: 15,407	Cross Post from ARN I've now read the paper half a dozen times and looked at some of the Supplemental Information supplied. No matter how much ducking, bobbing, and weaving IDists do, the central finding of the paper is unmistakeable, regardless of naturalistic assumptions, Darwinian zealotry, or funding source: 1. Start with a population of competing replicators. 2. Add random mutations of biologically known kinds occurring at rates comparable to those that occur in biological populations; a fitness function; and selection against the fitness function. 3. Let the population evolve by itself (with no "intelligent agency" intermittently intervening) for 100,000 cycles. Results: In nearly 50% (23/50) of the evolutionary runs, each starting from a different randomization, an assembly language computer program that meets the technical requirements of Behe's definition of "irreducible complexity" evolves. And the entire lineages of those irreducibly complex programs can be traced all the way back to the ancestral population, showing that they evolve by regular Darwinian incremental step-by-step evolution, using such exotic concepts as point, insertion, and deletion mutations, and cooption. Even epistatic interactions emerge step by incremental step, just like in biology. And just like those benighted Darwinians said it could happen. And it ain't even rare! There are other interesting findings. For example, the irreducibly complex programs that emerged from the evolutionary runs had different lineages from run to run - lots of different evolutionary pathways got the systems to the same phenotypic outcome. And the role of mildly deleterious mutations in 'setting the stage' for later fitness gains is real interesting. Contrary to Leo's [on ARN] naive belief, natural selection doesn't instantly remove each and every deleterious mutation the instant it appears. I'd suggest Leo learn some evolutionary biology rather than guessing at it from the simplistic misrepresentations of Wells, Behe, Dembski, et alia. Irreducible complexity is dead on the ground, no longer even twitching feebly. It's time to give the mousetrap Last Rites and bury it beside phlogiston, N-Rays, and cold fusion. RIP: IC, CSI, and SC. RBH

Thread Tools	Search this Thread
Show Printable Version	Search this Thread: Advanced Search

05-08-2003, 07:25 PM	#2
RufusAtticus Veteran Member Join Date: Nov 2001 Location: NCSU Posts: 5,853	Thanx for the heads up, I'm downloading the paper now.

05-08-2003, 08:42 PM	#3
RBH Contributor Join Date: Aug 2002 Location: Ohio Posts: 15,407	For those who don't have a subscription to Nature, the full text and supplementary materials are here. RBH

05-09-2003, 06:13 AM	#4
GunnerJ Banned Join Date: Sep 2001 Location: a place where i can list whatever location i want Posts: 4,871	Creationist: It's just a computer program! (Biologists show how living things in nature evolve complex fetures) Creationist: It isn't science if it isn't done in a lab! (Biologists preform an experiment theough induced mutations over many years that causes the evolution of novel complex features) Creationist: It isn't evolution if it doesn't occur in nature! (Biologists shoot themselves in despair for the fate of the human race)

05-09-2003, 09:41 AM	#7
Dr.GH Veteran Member Join Date: Jul 2001 Location: Dana Point, Ca, USA Posts: 2,115	Thanks for the link, Patrick. And, yeah, Gunner, you got it right. Gary

05-10-2003, 02:21 PM	#10
JonF Veteran Member Join Date: Sep 2002 Location: Acton, MA USA Posts: 1,230	Naw ID's still alive and viable. There was an intelligent entity within 1,000 miles of the computers on which this work was performed, right? Therefore CSI was smuggled in!

Freethought & Rationalism Archive

The archives are read only.