Nothing is known with absolute certainty. That’s just the best estimate. There are also other ways to estimate that a sequence is evolving neutrally. Does the spectrum of differences between or within species match the mutation spectrum? Is the distribution of differences among sites unbiased by location?
And yes, you should read Larry’s book.
In which conservation is a proxy for function. But with junk DNA, he’s trying to argue the opposite. Amazing.
Exactly, except I would suggest that it is of negative scientific value.
Science itself (hypothesis testing) is a method for testing what one imagines, which is why all ID promoters and followers reject the method, except for their habit of falsely stating empirical predictions of ID hypotheses as facts.
Science is a way to OVERCOME our faulty intuitions.
What, mathematically, is a “perfect similarity,” Gil?
Preach!
And if I may, some of my recent posts on this very topic:
Scientific thinking as the antidote to intuition
Science, intuition and the “strange inversion of reasoning”
What is amazing is that you seem not to be aware of the following :
If « A then B » is true, it doesn’t follow that « if B, then A » is true.
What is amazing is that you seem to not be aware that nobody actually made that argument to begin with, or that you were corrected about that misconception, or that it was explained to you in unambiguous terms how and why what is being said is not that, by yours truly, among others.
Why then John found it amazing that I support Gpuccio’s argument (where conservation is a proxy of function) and at the same time arguing for possible functions of non-conserved sequences?
As I’ve explained before, it’s because you seem incapable of understanding any argument that’s not all or nothing, but is instead about probabilities. Non-conservation is a good argument for lack of function. Even though some non-conserved sequences are likely to be functional (those that have recently become so) and some conserved sequences are likely to be non-functional (again, recently), we can with reasonable confidence estimate that the non-functional fraction of your genome is about 90%. And as has been mentioned by many people, many times, there are other signals of junk, e.g. that a considerable fraction of your genome consists of broken transposons, that the differences among species fit the distribution expected from the frequencies of mutation types and locations, etc.
And if non-conservation doesn’t imply non-function, Gpuccio’s measure of function is indeed useless, since a nonconserved sequence could have as much functional information as a conserved one. If you have no probabilistic argument, you have nothing. It’s your inability to recognize any probabilities other than 1 and 0 that gets in your way.
Why does the fact that a nonconserved sequence could have as much FI as conserved one makes Gpuccio’s measure of function useless, since his use of conservation through deep time only aims at producing true positives for function?
I think the disconnect here may be that while “if a then b” does not mean that “if b then a,” it does mean that “if not b, then not a”
For example: if all dogs are cute, (ie: if it is a dog, then it is cute) then it does not follow that if it is cute then it is a dog. It could be a cute cat, etc.
But it is true that if we accept the above premise as true that if it is not cute, then it is not a dog, because all dogs are cute.
Same thing here.
If conservation is a proxy for function, then if there is conservation, then there is function (probabilistically, @John_Harshman I’m over simplifying. Don’t taze me bro!)
It does not follow that if there is function then there is conservation, but it does follow that if there is not function, then there is not conservation (generally, subject to instances where the function arise recently, and probably other exceptions I don’t understand–we are just taking about the logic part right now, ok?!)
Why do we care? Because rephrased what this means is that all non functional portions of the genome (generally, subject to the above-mentioned exceptions) will be in the unconserved regions of the genome.
So we can say that 90% if the genome might be junk. You have to agree with that, logically.
From there, we have to go from what is possible to what is probable. So if it might be junk, do we have any reason to think it might not be junk other than ignorance or Do we have any evidence it might not be junk? Assuming the answer is neither of the above we have reached the limit of what we can do with logic.
What we can conclude then use that it is possible that this is junk, and we have no reasons to think that it is not junk.
From there, we have to employ a new device of reasoning, a practical one. We conclude, tentatively, and provisionally, that unless and until we have a reason or evidence that it is not junk, that it is junk.
You can disagree with this practical device if you want, and say that it is not strictly logical, and that’s true, it isn’t. But that is because we are at the limit of what is knowable. In other words, we are at the point of “reasonable doubt.” If you continue to doubt this conclusion, it is by definition without a reason.
Did that help clear things up for you at all?
That’s not true, though is it? In general a conserved sequence should have considerably more FI (at least by the measure Gpuccio - incorrectly - claimed to use). And by Gpuccio’s - incorrect - reasoning it must have more.
And the fact that Gpuccio didn’t measure conservation is rather a problem too.
There is plenty of reason to believe it might not be junk as many protein coding genes are not conserved among mammals.
Different mammals share some genes but don’t share others.
This points against the inference of common descent which the junk DNA logic is based on.
How is that a reason to believe that the 90% of the human genome that is unconserved (and which doesn’t fall in to some other known exception such as recently acquired function) is nonetheless functional? Walk me through your logic on this point.
Oh yes, I vaguely remember a paper from several years ago (I think it was brought to attention by Moran on his blog? not sure) about most of human genetic variation consistent with neutral evolution, but I don’t remember what paper and I cannot find it.
Bill couldn’t be carried through logic by a dozen philosophers on amphetamines, so asking him to do the legwork is foolish in the extreme.
Not conserved is an assumption. It could also be simply a different starting point. The gene differences point to different starting points.
Yea, God just picked a nested hierarchy of starting points, consistent with the nested hierarchy of starting points for the functional regions, just so it would look exactly like common descent, but actually it’s all a ruse.
You are assuming common descent is the cause of the nested hierarchy. Common descent cannot explain many of the differences that are a critical aspect of a nested hierarchy.
No, I’m not. It’s just that nested hierarchies are expected of common descent and not expected from processes that are not common descent. Common descent is, therefore, the simplest explanation, for the time being.
Oh, yeah? Like what? Please, name two such differences that are a critical aspect of a nested hierarchy, which common descent provides no adequate account of.