Why can’t you run a simulation with the environment as a feedback mechanism and find function in a changing sequence? Start with an exceedingly strong feedback mechanism.
I have no idea what you mean. If what you’re saying is that you want to create a sort of simulation of selection towards english sentences without particular specific target sentences, you should probably just test for readability by people who can read english.
That would require someone to read the sequences to see how much sense they make and then “score” them according to how sensible they find them. Higher scores means more sensible. Then make the algorithm so that higher scores → more likely to be passed on to the next generation.
This is an interesting idea. Not exactly an environmental test but it will show if some functionality is found then selection occurs. This is not to a specific target but can you generate new meaning starting from a 100 bit functional sequence and can you create 100 bits of functional meaning from a random set of characters within 100k generations or so.
CLAVDIVS
(Tiberius Claudius Caesar Augustus Germanicus)
23
I think going back to @Rumraket’s original question in this topic, can you (or anyone else for that matter) think up a functional protein sequence de novo? If so, it would be interesting to see it. If not, why is that? Do we lack enough knowledge of the chemistry involved? Does it require iterative processes? In that case what was the role of the mind? Is to to design the process that designs the protein?
I’m going to sound like a broken record here but I think you need an actual mechanism or else it’s easy to lose any persuasiveness of your argument from analogy.
Once the KRAB domain was fused to the tetracycline repressor TetR, the TetR-KRAB fusion proteins were the first engineered drug-inducible repressor that worked in mammalian cells.
We have websites that if one enters a DNA sequence, it will try to construct a Zinc-Finger sequence that will bind to it. This is in order to make man-made transcription factors. I played with the tools for a couple hours once. They don’t work so well as the God-made zinc fingers.
There are two ways the tools work. One puts zinc finger array amino acid sequences, and it computes the DNA target. Well, I put in a known zinc finger array defined by amino acid sequences, and the computed target didn’t look at all like the actual DNA binding target determined in the lab.
Then one can do the reverse and put in a proposed DNA binding target, and the tool will try to concoct at protein zinc finger sequence. Well, reports are that it doesn’t work as well as they’d like, for sure.
The reason the attempts have some hope is the zinc finger domain fold is simple because of the metal binding residues. So for some limited applications we have some hope of trying to concoct a functional protein to do something de-novo.
I don’t think we’re any where close to building proteins that form de-novo complexes like the chromatin modifiers. That would be some serious engineering.
spent waves sang
beating pointed time and
measureless
with cold knowledge
revealing
one dream is I perhaps love you
i say that sometimes
on these long talkative animals
are laid fists of huger silence.
i have seen all the silence
filled with vivid noiseless boys
It’s also possible to write an evolutionary algorithm and use a spelling and grammar checker as the selection process. Or to modify ‘weasel’ so that it doesn’t have a fixed target, but uses comparison against any substring from any available work.
That’s nice but not really what I suggested. I imagine a process that basically starts with random gibberish sequences containing english letters, spaces, and symbols. Then the “reader” scores it for making sense. Of course to begin with it won’t make much sense at all, perhaps there will just be a few letters that make out a single misspelled word in a larger nonsense sequence. Well, that’s better than if there was no comprehensible word at all. Now the question is if subsequent mutations to that sequence can bring it further towards a comprehensible sentence. If it can, those can be selected to seed the next generation.
If the algorithm starts out with correctly spelled english words and correctly puts spaces between them I could see how that is to some extend stacking the deck. The question is, can sense evolve from pure gibberish? Not that I think this would prove much except the power of cumulative selection. But in the end, biochemistry isn’t written english.
That’s possible too. As I said, it’s possible to write an evolutionary algorithm and use a spelling and grammar checker as the selection process.
I tried it once, several decades ago, but gave up because I couldn’t produce a workable fitness function for grammar using the tools I had available at the time. I did discover the hard way how evolutionary algorithms pervert selection criteria, as my attempts at a ranking process favoured
(i) long sequences of letters with no spaces (only one misspelt word!)
(ii) all spaces, no letters (no misspelt words!)
(iii) “a a a a a a a…” (lots of correctly spelt words!)
IIRC I eventually persuaded the algorithm to generate strings of non-trivial valid words by giving every word of length n a score of n^2.
Against our agreed upon definition of mechanism I think this is done.
In the science of biology, a mechanism is a system of causally interacting parts and processes that produce one or more effects. Scientists explain phenomena by describing mechanisms that could produce the phenomena.
A mind fits this definition as it is a known cause of functional sequences and purposely arranged parts. This is the theory at this point.
How the mind does this is an interesting question but we know very little at this point to make this a very productive discussion. We can understand how and why proteins bind (charge and shape) but how they work together to form a very complex function is a more difficult question.
But you don’t have processes. You have a causal agent and end result, I’m still not seeing a mechanism.
The reason this is important is so we can distinguish between multiple possible mechanisms. If the end result is the same, and the only distinguishing thing is the causal agent, we have no way to tell the difference.
@colewd does not have a causal agent, either. We do not know of any minds that exist independent of DNA and protein-based organisms. So a mind cannot be posited as a cause of DNA and protein. This is using ID creationists’ own argument that we must directly observe mutations, drift and natural selection create a functional protein from scratch if we are to accept evolution. The ID creationists have a serious credibility problem if they are going to use such a blatant double standard.
No, it doesn’t. Saying “a mind is the mechanism” is as scientifically vacuous as saying “design is a mechanism”. Both are empty placeholders until ID does the work and identifies the forces harnessed by the Designer to manipulate matter.
I was meaning in the sense of what was being proposed. I wasn’t assessing the truth of the causal agent, just that he was proposing one.
What is somewhat frustrating about ID to me, is that I actually agree that there was a mind, a designer, who is the ultimate cause of all life, but ID’s seeming insistence that it (mind/design) is a scientific proposition distracts from real dialogue between scientists and people of faith.
It isn’t enough to think something through. You have to actually do some work. Even in abstract mathematics, we need to write down our proof – which is a physical action.
Without that – we are very clever at fooling ourselves.
Why does it not meet the posted definition? To be a mechanism it simply has to be interacting parts (brain cells etc) causing an effect…information. A mind is clearly a causal mechanism.
Mechanism and process are different things. A process is the cause and effect of several mechanisms. We can represent a process diagram by several boxes where Box A is the cause of Box B and so on. After DNA is available to the cell then several processes are enabled. These processes are understood however the origin of the genetic information in DNA is not. This is where mind is a cause that is known to create the effect of long complex sequences or genetic information.
Einsteins theory of general relativity has causal mechanism as mass/ energy to explain space time curvature. There is some more detail in the stress-energy tensor but it is the same idea.
Joshua suggested in the discussion with Gpuccio to have negative controls. This is a great suggestion and is a way to eliminate other potential causes like random variation.