FFS, Mung. Just tell us what you mean.
We search with sequences we make up all the time–most often PCR primers, because a match of only a few bases at the 3’ end can cause nonspecific amplification. However, we’re not doing any sneaking.
Well, you appear to think that sneaking would be required to search with artificial sequences.
I think your questions would be perceived as more friendly if they did not include weird, prejudicial terms like “sneak.”
2 Likes
It’s already there, up thread, John. To me you’re just coming off as obtuse.
No, they are not, for the reasons I gave, which had nothing to do with scientists faking data. And you acknowledged the language you used was sloppy.
The answer, of course, is yes. So as I stated:
See this post:
That’s all i was getting at. nothing more, nothing less. Why people had to try to turn it into more than what it was is beyond me.
1 Like
No, I don’t see that. If I made up random sequences and input them the programs would not output trees with low p values. Known sequences, with few exceptions, produce trees that have very tiny p values.
You’re not seeing that the tree produced from the data is not THE phylogenetic tree. It is a mathematical representation of the sequence data. Every tree output has a p value relative to no mathematical relationships, and each tree is specific to the input data.
2 Likes
Bingo. That is a excellent, very pithy description! It concisely says almost all that need be said.
Nevertheless, I’ll risk saying more in order to perhaps help emphasize the point.
No. Not at all. You are not understanding this at all. With “sequences that you just made up”, it wouldn’t suggest Common Descent at all.
Instead of being a “just made up” scenario, a person would have to work very very very very hard to carefully manufacture “made up random sequences” (that is, to “make up” huge numbers of random sequences but then only select the extremely rare, appropriate random sequences) which would just happen to smoothly fit into tree outputs with impressive p-values. Then and only then would it be possible to claim that the Mung’s Fictional Values, aka “made up sequences”, would have any chance whatsoever of looking anything like Common Descent.
Mung, as a former Young Earth Creationist denier of evolution who had to dig his way out the hard way, let me tell you what I discovered long ago: The evidence for Common Descent is truly overwhelming. And then when many years later I had the opportunity to review these kinds of genomic data, this yet-another massive pile of evidence for Common Descent was an absolute slam dunk! Masterfully so.
From reading your posts thus far, it appears that you aren’t at all grasping the weight of John Mercer’s pithy summary of the phylogenetic trees. As the saying goes, “The numbers don’t lie.”
Indeed, it would take extremely brainy, careful, and hard-working gremlins to produce the misleading phylogenetic trees you are talking about, Mung.
@Mung would you like to learn how to use one of the programs, so you can see for yourself what happens when you put random sequences into them?
4 Likes
Oh my. Have I said somewhere that I was ever a young earth creationist denier of evolution?
No. I never claimed that you did. I’m describing my own pedigree so that you understand that I came from a particularly defiant strain of evolution-deniers, the Morris-Whitcomb-Gish “creation science” movement of the 1950’s and 1960’s. I’m suggesting that I was probably a much stronger denier of Common Descent back then than you are today----yet the massive evidence won me over.
I recommend that you take up Dr. Swamidass on his try-it-yourself software offer.
2 Likes
How easy it is to be misunderstood! My bad. I should have read further.
I am not a denier of common descent. 
Nevertheless, you are posting the very same kinds of arguments that deniers of Common Descent like to emphasize.
As the popular saying appears to suggest: If you quack like a duck, lots of people will assume that you are a duck.
I am not talking about phylogenetic trees being misleading. You are in fact making the same point that i was making, perhaps unwittingly. Just any old phylogenetic tree that can be constructed is insufficient on its own to serve as evidence for common descent. Something more is required, whether consideration of p values or something else. That is the ONLY point I have been trying to make.
Sure. It would be an interesting exercise from which I might learn something I didn’t know before.
But keep in mind that I am not denying common descent. I am denying that producing a phylogenetic tree is prima facie evidence for common descent. John and others at least appear to agree that it is not. And that is ALL that I was disputing.
Not without further qualification and expansion.
You’re still apparently missing the point. The trees we have are graphical representations of the mathematical evidence. They come with p values.
I strongly recommend that you take Swamidass up on his generous offer. You’ll learn a lot and overcome these misunderstandings. You can even make up sequences as controls and see how they are handled.
1 Like
And that denial is wrong. We aren’t talking about some arbitrary “phylogenetic tree” that a non-scientist amateur might make. The phylogenetic trees that scientists make are based on impressive p-values. Always. So they are indeed such prima facie evidence for common descent.
1 Like
I have a question. If the p-values weren’t so impressive, wouldn’t those resulting graphical structures be more like confused, tangled networks (as in M:M networks)—and not anything much like trees at all?
Does the way that I’m looking at this make sense?
Indeed, wouldn’t Mung’s imagined “trees” actually be tangled spaghetti? (They’d have various random clumping, no doubt, but still just a mess. Right?)
In a previous post, @John_Harshman wrote:
Perhaps you missed that part?
You can force any old set of features into a tree. The question is if it has a high p-value (i.e. is it statistically significant).
1 Like
Thanks
So if I am reading you correctly, i could not make up some imaginary taxa, allocate some sequence data to them, and get a phylogenetic tree? Or that I could do that but … what?
Could you briefly describe the role of the p value? I could create a tree in the above hypothetical scenario and it would come with a p value, and that p value would indicate … what?
Keep in mind that I am probably not the only layperson reading this.
2 Likes
I didn’t miss it. The software constructs different trees and tries to pick the “best” tree where best is defined by the p-value?
Many different trees can be constructed from the same data. Is that correct?
The software tries to pick “the best” tree from the possible trees. Is that correct?
And the software uses the p value to choose which tree is best?
Don’t tell me it’s not rocket science, I am pretty sure it isn’t. 
1 Like
Depends in the method, if my very small understanding of computational phylogenetics is correct. Bootstrapping is one of the methods that one can use to determine if the tree you have built is dissimilar from randomly chosen data.
The method does something similar to what you are asking. It randomizes the data and compares it to the hypothesized tree. If the tree you are testing isn’t significantly different from random data then it casts doubt on the tree being the product of common descent.
2 Likes
I’m not convinced. But hopefully we are working our way through it. I’m not arguing against common descent. I’m arguing for clarification on how we decide that a phylogenetic tree is evidence for common descent. It appears to have something to do with “statistical significance.”
Would I be too far off base if I thought that meant that something other than “chance” was at play? Not chance, therefore common descent? Or, we can rule out chance, and we can rule out gremlins, therefore common descent?