Journal club time again 🙂
I like this paper: Functional information and the emergence of biocomplexity by Hazen et al, 2007 in PNAS, and which I hadn’t been aware of.
I’ve only had time to skim it so far, but as it seems to be an interesting treatment of the concepts variously referred to by ID proponents as CSI, dFCSI, etc, I thought it might be useful. It is also written with reference to AVIDA. Here is the abstract:
Complex emergent systems of many interacting components, including complex biological systems, have the potential to perform quantifiable functions. Accordingly, we define “functional information,” I(Ex ), as a measure of system complexity. For a given system and function, x (e.g., a folded RNA sequence that binds to GTP), and degree of function, Ex (e.g., the RNA–GTP binding energy), I(Ex ) = −log2[F(E x)], where F(Ex ) is the fraction of all possible configurations of the system that possess a degree of function ≥ Ex . Functional information, which we illustrate with letter sequences, artificial life, and biopolymers, thus represents the probability that an arbitrary configuration of a system will achieve a specific function to a specified degree. In each case we observe evidence for several distinct solutions with different maximum degrees of function, features that lead to steps in plots of information versus degree of function.
I thought it would be interesting to look at following the thread on Abel’s paper. I’d certainly be interested in hearing what our ID contributors make of it 🙂
Or, I can respond to whatever I feel like responding to whenever I feel like responding to it, whether or not you consider it an “acceptable” alternative.
Very true. This is not UD and Elizabeth is not Barry, after all.
Helloooo guano!
You continue to try imposing goals and targets on the process.
Outside the laboratory, selection simply happens, It isn’t a deliberately structured algorithm. At some level, all causes are descriptions of before and after states. Gravity is a fine example of a cause that could be called tautological.
WJM:
You saying God’s designs are getting worse?
All that a bacterium can produce is two bacteria. That is its entire ‘fecundity function’. Its 4-billion-year success at that is indeed testament to the power of that way of doing things. Still, ultimately, it hits limits to growth, and the steady state average is limited to one surviving ‘offspring’ per ‘parent’.
But multicellular organisms – by virtue of precisely that feature, multicellularity – can produce billions upon billions of ‘offspring’ – ie, gametes. That excess is filtered by the need to find a partner in a sexual species, and ultimately, brought back to an average of a half-share in each of two offspring per organism by environmental limits in a steady state population. Which is the same as the bacterium – after all that messing about, all the gametes in a present population end up generating an average of one copy each in the next (in a steady state population).
The haploid genomes are neither more nor less fecund – they just generate intermediates that are one hell of a lot bigger and more complex. And one significant reason for that is sex.
Evolution is running-to-stand-still. When a new allele arises that raises fitness (offspring production), it spreads until the entire population has it. The playing field has been levelled again – until another allele arises that raises (immediate) fitness. It is a mischaracterisation of the process to think that it means that modern organisms should be more fecund on the average than ancient ones – there are external limits, not least a world full of fecund organisms all playing the same game. A reading of Malthus might help there. Fecundity promotes variants within the population; it does not indefinitely increase the average fecundity of the population, in the presence of other populations.
Gene frequency?
You are starting with what needs to be explained in the first place.
Geez it appears that you don’t even understand what ID says.
LoL!
No one is refuting natural selection. We are saying it doesn’t do anything.
Huge difference.
You continue to fundamentally misunderstand the difference between (1) claiming a target is prescribed by a process and (2) pointing at what a process is claimed to have achieved and using the term “target” to refer that structure and others categorically like it.
Of course 🙂 I just said I’d be equally interested in a concession as in a rebuttal.
You are more than welcome, of course to supply something completely different, or nothing at all.
But “nothing at all” of course would be less interesting.
So let’s just say I’d be interested in any response you care to offer 🙂
But unfortunately I won’t be back to read it until Sunday. Have a good time, guys 🙂
(And Joe G, please not a whole pile of guano for me to shift when I get back – my garden is grateful, but my back hurts ….)
Joe G,
William Dembski doesn’t understand it either? His proofs use a space of genotypes (which, as anyone who has taken my theoretical population genetics course must know, means that any population of genotypes from that space has gene frequencies).
No, organisms are not “less fit”, in any sense that is relevant, although perhaps in senses that you consider tautological.
Populations today are (to those of us who accept common descent) those that have an unbroken lineage to our earliest common ancestral population, and consist of enough individuals with sufficient fitness to pass on their genetic inheritance to their offspring that the population remains extant (not extinct).
They are the populations, in other words, with genomes that have been supplied with variance, and honed by natural selection, in such a manner that they have fitted their bearers to environments that range from primordial seas to a vast range of alternatives, including environments consisting of other living populations.
Our own population now numbers 7 billion, and has even ventured into environments beyond the earth itself.
If your definition of “fitness” excludes that diversity of capacity to thrive as “fitness”, I respectfully submit that your definition may be too narrow 🙂
Joe,
ID is all about origins- how living organisms originated directly affects any subsequent evolution.
Living organisms were designed then they were also designed to evolve/ evolved by design.
and how would we tell that from regular NDE?
I think that, given the statements here by ID supporters, I should make such a post. I’ll draft it and send it to you Sunday so it will be here when you get back. I hope serious commenters like William J. Murray will be available for that discussion. There are assertions being made here that natural selection cannot put Functional Information (or Specified Information) into the genome. These are mistaken. This forum seems a good place to get everyone to agree. If your email address has changed recently please email me.
Thanks for the kind words about my paper. Many of its arguments are due to others but I worked long and hard on it, and read William Dembski’s works very carefully. I think it is the clearest explanation out there of why (the effectiveness of) natural selection has not been refuted. It is now almost 5 years old, yet pro-ID commenters keep repeating arguments that were shown wrong in that paper. So we need another go-round, I guess.
Thanks! Joe F. If you register here, I can set your permissions to post under your own name. Off to Wales now, see you all Sunday 🙂
Joe:
If that evolutionary capacity, and this mysterious GA, were so poor that the designer had to keep coming back and undetectably fiddle with it, it’s not really ‘designed to evolve’, is it?
Natural selection doesn’t put anything into the genome. It only removes stuff that has already appeared there.
NDE doesn’t do anything- it can’t even be tested- no testable hypotheses.
Nice strawman…
I, for one, look forward to Joe’s paper that demonstrates natural selection can actually do something…
It’s not too hard to see why natural selection can increase the value of this mathematical expression whether it is called “functional information” or something else.
If the “function” that a subset of assembly rearrangements is to perform has to be performed at a specified level or better (those specifications are set by the current environment, of course), then there will be a subset of that subset of assemblies that have a higher level of performance.
All natural selection has to do is “up the ante” by raising the bar on performance. That will either reduce the size of the subset of assemblies that can do the job or it can simply set the bar so high that there will not be any rearrangements of the assembly that can get the job done at this level.
Thus the negative of the logarithm of that fraction of the assemblies gets larger as the fraction gets smaller. That is simply a result of the way this mathematical expression is defined. It’s the name “functional information” that appears to be confusing ID/creationists.
Note that this does not imply that the subsets of assemblies that do the job well are necessarily extremely complicated (however one wishes to quantify that). Complexity may be an inhibitor in doing a particular function that a simpler system can do much better.
In fact, the best system for the job may be intermediate between extreme simplicity and extreme complexity. So natural selection does not necessarily select for some “awesome complexity.” The result may not even give the impression of being designed
Here, let me help you.
That would only be true if information was made of DNA. So are you saying that information is made of molecules?
Umm I have already read that paper- no evidence, just rhetoric.
Except natural selection isn’t selection it is just a result. And a result that apparently cannot be differentiated from genetic drift.
WJM:
I’m not sure where you get that from, but it is not true. Perhaps you are confusing what NS does to populations with what happens in individuals. Remember that “removing stuff that has already appeared there” can include getting rid of an ancestral sequence in the face of a new variant with elevated fitness. This is a process of “putting in something new” – concentration of one is automatically dilution of the other, in either direction.
One needs to be clear about what is meant by ‘the genome’. Joe F was talking in terms of the ‘consensus’ genome of an entire species – cf The Human Genome [sense 1]. But another way of defining the genome is the entire hereditary information in a single cell [sense 2]. Evolution happens by a change in sense 2 being inherited such that there is an overall change in sense 1.
Natural selection neither adds to nor subtracts from the genome sense 2 (individual cell). That task is performed by mutation and recombination. DNA can be added and removed, activated or disabled; ‘function’ can be changed for the better or worse (or not at all). NS isn’t doing it, but that does not affect the fact that the process can both ‘put in’ and ‘remove stuff’.
Once a change has happened in one cell (genome sense 2), then to register a change in the species (genome sense 1), it has to become fixed in the population. Note that the change has already happened in the very first cell it arose in. It takes over the population by leaving descendants.
The process of fixation in the species can include Natural Selection. That means that the new variant causes the fitness of its bearers to be greater on average than ‘wild-type’. In that sense, Natural Selection adds to/subtracts from the genome sense 2 (population) that which was added to/subtracted from the genome sense 1 (first bearer).
JoeG: “Umm I have already read that paper- no evidence, just rhetoric.”
Boy, THERE’S an exhaustive critique, if I’ve ever seen one!
Perhaps JoeG is complaining about the lack of empirical data from actual organisms? Yeah, Joe Felsenstein wrote a merely theoretical refutation of Dembski’s completely theoretical argument… we’re missing the lab experiments, the field experiments, and the experimental replication of the entire history of life tracking every mutation nucleotide by nucleotide (and that experiment would have been designed, therefore proving ID). And balloon animals!
What is there to critique?
Please point me to the EVIDENCE which DEMONSTRATES that natural selection can construct things.
Don’t blame me because your position lacks supportingg data.
joe g said:
“Except natural selection isn’t selection it is just a result.”
A result of what? And can there be a “result” if nothing happens?
joe g said:
“Please point me to the EVIDENCE which DEMONSTRATES that natural selection can construct things.”
I don’t recall anyone saying that NS “can construct things”. You’re arguing with real scientists, joe. You should be more specific, and at least try to sound more sciency.
Hey joe, please point me to the EVIDENCE which DEMONSTRATES that your imaginary god allah can construct things.
Joe Felsenstein,
Hello ID/Creationist here, I have just starting reading your paper and I believe I have stumbled across a possible glitch in your premise that would lead to an invalid conclusion. I’ll wait for Sunday to view the contents of your summary before posting my 2 cents.
We can do that here too. The post I hope to make soon will mostly set up a specific genetic model and show that natural selection does put Functional Information into the genome. If you have found a glitch in the argument of my 2007 paper you might better mention it here unless it concerns that model.
Thanks for being willing to actually discuss the argument of my paper. It is a refreshing contrast to absurd demands that I change the topic to the Origin Of Life, or to some sort of empirical evidence.
I have now registered. Enjoy Wales and its friendly people, who Flanders and Swann memorably said “sing much too loud, much too long, and too flat”.
Well if you don’t have any empirical evidence you don’t have anything wrt science.
BTW ID is all about the OoL. As I said if living organisms were designed then they were also designed to evolve.
Joe:
Well … I can’t resist – why does this mean that there is anything wrong – anything at all – with evolutionary theory? That is, a process of successive mutation-fixations taking place since, giving rise to the pattern of Common descent that arose from their differential operation across reproductive barriers, plus IC, CSI, OMGJIAFA etc, while the designer suns himself on Omega 6 with a good book? Why did the designer have to do anything further, if life was ‘designed to evolve’ … ?
Well ID is not anti-evolution- and if the evolution is directed then it ain’t blind watchmaker evolution.
And again if your position had any positive evidence we wouldn’t be having this discussion
Jesus Christ, Joe, how many times have you posted that? Don’t you get bored?
You say that ID is not anti-evolution, and then refuse to accept Word 1 from evolutionary theory. And positive evidence? Give me a break! Show me one of your freakin’ designing aliens and we wouldn’t be having this – ahem – ‘discussion’.
But either way – evolution including Darwinian mechanisms or something unspecified arising from your ‘GA’ – I’ll simplify my question to this: do you think the designer continued to tinker with life after programming it at the OOL, or do you think the ‘GA’ was sufficient?
By the same token, Dembski doesn’t have anything with respect to science.
Yes, ID is only anti-evolutionary theory.
Where’s your empirical evidence of allah, joe?
Where’s your empirical evidence of ID?
Where’s your empirical evidence of “CSI”?
Where’s your empirical evidence of your assertions about genetic algorithms?
Where’s your empirical evidence of your assertions about origins/creation?
Where is the ‘science’ in anything that you assert?
Where’s your empirical evidence of allah, joe?
Where’s your empirical evidence of ID?
Where’s your empirical evidence of “CSI”?
Where’s your empirical evidence of your assertions about genetic algorithms?
Where’s your empirical evidence of your assertions about origins/creation?
Where is the ‘science’ in anything that you assert?
Joe Felsenstein,
It pertains to your example of NS and csi. But I think your post will be directed at this.
Exactly- and that only applies to the current ToE
Except ID has plenty of empirical evidence that agencies can produce CSI and IC.
Allan- don’t blame me for your ignorance.
Also I see you need proof of ID and that proves that you ain’t interested in science.
GA was sufficient- as I said it all depends on those starting conditions.
As dawkins and others have said- if living organisms were designed we would be looking at a totally different kind of biology- just as saying Stonehenge was designed turned a possible geological investigation into an archaeoloical investigation- totally different.
Given that an algorithm may be defined as a list of instructions to perform a sequence of operations, some of which may be dependent upon the results of preceding operations,; whereabouts in the cell, Joe, is your GA – the list of instructions – stored?
In the genome or elsewhere?,. If not in the genome, where exactly?
In the cell- that is all I know at this time. I doubt it is in the DNA although the DNA can be a storge medium for instructions.
Joe, think that’s an incorrect use of the word “know”
When are you coming over to the UK, Joe?. I could pick you up at Heathrow, we could be in Cambridge in just over the hour, and I’ll try and line up some top-flight geneticists and cell physiologists to talk to, and labs to visit – Cambridge is stuffed with ’em both within and outside the University.
I’ll even try to line up some IDists, though they’re a bit thin on the ground in these parts
That paper at the beginning of this thread might well be interesting without any slant on ID.
Could we have a sensible discussion about what it implies without any ID types interfering?