Background
For the past month or so I’ve been investigating the claim that the phylogenetic signal is evidence that a dataset shares common descent.
Supposedly, the phylogenetic signal is one of, if not the, strongest pieces of evidence for common descent. It is one of the first of the 29+ evidences for evolution offered over at Talk Origins (TO).
Quoting from the article:
The degree to which a given phylogeny displays a unique, well-supported, objective nested hierarchy can be rigorously quantified. Several different statistical tests have been developed for determining whether a phylogeny has a subjective or objective nested hierarchy, or whether a given nested hierarchy could have been generated by a chance process instead of a genealogical process (Swofford 1996, p. 504). These tests measure the degree of “cladistic hierarchical structure” (also known as the “phylogenetic signal”) in a phylogeny, and phylogenies based upon true genealogical processes give high values of hierarchical structure, whereas subjective phylogenies that have only apparent hierarchical structure (like a phylogeny of cars, for example) give low values (Archie 1989; Faith and Cranston 1991; Farris 1989; Felsenstein 1985; Hillis 1991; Hillis and Huelsenbeck 1992; Huelsenbeck et al. 2001; Klassen et al. 1991).
http://www.talkorigins.org/faqs/comdesc/section1.html#nested_hierarchy
I’ve been skeptical of this claim. A tree is just one kind of directed acyclic graph (DAG), and my hunch is many kinds of DAGs will also score highly on metrics for phylogenetic signal. I picked one metric, the consistency index (CI), which according to Klassen 1991 is the most widely used metric. It also is the featured metric in the above TO article. Plus, it is very simple to calculate. So, I’ve focused my efforts on the CI metric.
Result
What I have found is that my hunch is correct. It is simple to create a DAG that scores highly in CI, well within the range of published CI scores for real datasets.
Consequently, it is incorrect to say the phylogenetic signal is strong evidence for evolution. In particular, this claim is provably false (as I have proven here):
Phylogenies based upon true genealogical processes give high values of hierarchical structure, whereas subjective phylogenies that have only apparent hierarchical structure give low values.
http://www.talkorigins.org/faqs/comdesc/section1.html#nested_hierarchy
How have I proven it false? I generate DNA sequences from directed acyclic graphs, and the trees derived from these sequences using well established methods produce CI scores well within published ranges. Here are two such experiments plotted on the chart from Klassen 1991:
This is a phylogeny with very high value of hierarchical structure not generated from a true genealogical process.
Methods
To reproduce my results you can run the DAG dataset generator here: https://repl.it/@EricHolloway/Phylogenetic-Signal-Fallacy-Nucleotide-Level
Take the generated DNA sequences, which are in FASTA format, and paste them into the ClustalW online tool.
Take the results of the ClustalW tool, and use the PAUP software to generate trees and measure CI scores. You’ll need to fiddle with the NEX file format, so to save you the trouble, I’ve included an already created NEX file that I’ve generated from the aforementioned process, which you can pop into PAUP.
Once you load a NEX file into PAUP, here are the steps to generate trees, and then measure CI.
- Press “Generate Trees” in the “Trees” menu.
- Press the “OK” button.
- Press “Describe Trees” in the “Trees” menu.
- Press the “Describe” button.
- You will see something like the following:
You will find the 27 taxa in the file will generate CI scores in the range of 0.48-0.53. If you look at the Klassen 1991 chart, you will see this is well within published scores for that number of taxa.
Conclusion
So, what is my takeaway from this?
Basically, highly statistically significant CI scores do not indicate common descent. They can just as easily be generated by a DAG. Therefore, we cannot infer common descent from high CI.
Furthermore, insofar as CI is representative of the state of phylogenetic signal measurement, my result undermines the more general claim that phylogenetic signal indicates common descent.
As such, the Talk Origin’s claim that the nested hierarchy of species is well attested by the data is highly questionable if not outright false, and should be retracted as evidence for evolution until such time as a much more rigorous analysis with DAG eliminating controls is established.
Addendum
To visually illustrate what I mean by a DAG generating the DNA sequences, here is a graph of one such DAG. Each colored/numbered box represents a gene, which is replaced by a unique, randomly generated (uniform over ‘GATC’) DNA sequence of 20-30 letters long in post processing. Arrows indicate when ‘ancestor’ gene sets are combined into larger gene sets. If you look closely, you will see each gene set contains the union of all incoming gene sets, plus one new gene. As you can see, this looks nothing at all like an evolutionary process, yet it produces very high phylogenetic signal as measured by the consistency index (CI) metric.
EricMH,
Dude, there is no real match between your data matrix and your DAG. The matrix has 16 taxa but the DAG has only 15. Some of your taxa in the matrix have over 30 “present” states but the largest terminal taxon in the DAG has only 15. I don’t think you know what a presence/absence character is or how to derive them from your DAG. The very notion that there could be gaps in a presence/absence matrix also shows that you don’t know what you’re doing.
Luckily, you’re off on another obsession, so this is all probably over.
EricMH,
This is scientific Creationism in a nutshell.
Ooops: when I wrote “between 49.5% and 50.5%”, I should have written “between 49.95% and 50.05%”. My bad. The 3-sig-fig vote shares come from Edison.
Well, if you look at the maps of which counties carried the Dem vote, it is usually just a very small number of counties, located around a major city, which happens to be controlled by democrats. Whereas in the rest of the state the vast majority of the counties voted for Trump.
Additionally, you have this from Diebold Election Systems, now known as Dominion Election Systems.
https://www.cs.cmu.edu/~matth/lists/support.w3archive/200110/msg00122.html
“Its a tough question, and it has a lot to do with perception. Of course everyone knows perception is reality.
Right now you can open GEMS’ .mdb file with MS-Access, and alter its contents. That includes the audit log. This isn’t anything new. In VTS, you can open the database with progress and do the same. The same would go for anyone else’s system using whatever database they are using. Hard drives are read-write entities. You can change their contents.
Now, where the perception comes in is that its right now very *easy* to change the contents. Double click the .mdb file. Even technical wizards at Metamor (or Ciber, or whatever) can figure that one out.
It is possible to put a secret password on the .mdb file to prevent Metamor from opening it with Access. I’ve threatened to put a password on the .mdb before when dealers/customers/support have done stupid things with the GEMS database structure using Access. Being able to end-run the database has admittedly got people out of a bind though. Jane (I think it was Jane) did some fancy footwork on the .mdb file in Gaston recently. I know our dealers do it. King County is famous for it. That’s why we’ve never put a password on the file before.”
Forbes research on use of Dominion machines:
https://www.forbes.com/sites/adamandrzejewski/2020/12/08/dominion-voting-systems-received-120-million-from-19-states-and-133-local-governments-to-provide-election-services-2017-2019/
So, the notion of a stolen vote does not strain credibility too much, at least to me.
What do you think it is displaying?
I’ve been trying to get the original timeseries datasets from the NYT. I contact their data supplier Edison Research, but no response so far.
I miswrote 15 taxa. If you count the taxa in the graph there are 16, and in the NEX file there are 16 taxa + a randomly generate outlier. A person at Biologos told me for some reason that outlier was essential, so I added it. But, I can drop the outlier and the results are unchanged.
Those boxes represent multiple binary characteristics, which is why the cardinality does not match what you see in the NEX file.
Anyways, the graph and NEX are all generated from the same datastructure internally, so they cannot diverge.
I really should write an article explaining in detail so there is no more confusion.
I grew a small vegetable garden this year. What should I have observed that would substantiate evolution?
What do you consider to be the most solid piece of evidence for evolution?
Like I said:
Eric,
I no longer regret asking you what you studied in college.
Unfortunately, your fellow conspiracy theorist left off the neither-Trump-nor-Biden data, so we cannot actually re-create the fallacy. But we can make powerful predictions about how the dataset will behave, and test those predictions.
FOR EXAMPLE:
Let’s plot the changes in percentage Trump against the changes in total vote tally.
If the tallies represent real data with fraudulent vote dumps, then the changes in percentage Trump should be correlated with the size of the changes in total tally, except for the fraudulent dumps, which should stand out as having a large percentage change for a small tally change. Most importantly, the non-fraudulent data points should show a continuous distribution.
If the tallies are being back-calculated from percentages with 3digit precision, then the re-calculated percentages will jump by multiples of 0.1% whenever there is a Biden <—> Trump transfer of share, plus a multiple of 0.05% on the rarer occasions when there is a major candidate <—> remainder transfer.
The changes in percentage will be clumped around these values.
Let’s take a look:
Then you have no idea what presence/absence characters are or what you’re modeling.
I count 15. We’re counting taxa that have no descendants, right? Your “outlier” is more properly called an outgroup, used to root the tree. A random outgroup is senseless, and it’s just more evidence that you have no idea what you’re doing.
If by “evolution” you mean common descent, I’d say that nested hierarchy is that evidence. But common descent doesn’t have to rely on just one piece of evidence. The fossil record is another. What’s your alternative model that accounts for all the data?
That’s like asking what is the best piece in a 1000 piece jigsaw puzzle.
The consilience of all the evidence from dozens of different and independent scientific disciplines – biology, genetics, geology, paleontology, radiometric dating, etc.- is the most solid feature.
@DNA_Jock
I think I see what you are saying. Someone has taken the official series of percentages and final vote count, and is creating a fake dataset purporting to show fraudulent vote dumps. That would explain the massive series of increments at the end that keep the ratio even, as the person is using null events to bring the votes back up to parity with the official record.
And you are saying because the person only has 3 dp, then that could explain why we get these strangely regular deltas for Biden, since they have to be in 0.1% increments.
It just so happens that when we divide these weird deltas by the total we get 0.1%.
4800/(2459249 2469225)=0.00097
Very plausible, but why don’t we see the same deltas on Trump’s side? An alternate explanation is the 3 dp resolution is the reason for the oddly specific fraudulent vote dumps.
No, there are 15 taxa in the graph. I counted them very carefully. And there are 16 taxa in your PAUP file, one of which might be your random outgroup.
Well, that one seems questionable. E.g. the cambrian explosion. Yes, I know you’ll have some explanation why cambrian explosion is not a problem for evolution, which may be well true. But it is not obviously true.
In other sciences I know, the basic results can be proven in a way that is entirely non controversial. I have never seen anything like this in evolutionary science. All the textbook examples: Haeckel embryos, moth melanin, tree of life, etc. are taught in a manner that makes them appear obviously true. But, when I dig into the actual science it is anything but obviously true, and tend to be controversial within the arena of well established biologists (I’m not talking about the IDists here). For example, just look at the Wikipedia page for recapitulation or the moth experiments. Or read the actual moth experiments for yourself.
The conscilience of evidence for evolution is one big soggy mess that whenever I peer too closely at any one support it crumbles and desparately screams out ‘conscilience of evidence’ to cover up its shortcomings. If you have a structure built on crumbling supports, it doesn’t matter how many crumbling supports you have, the whole structure is unsound.
Ok then it is 15 taxa + 1 random outgroup.
There’s some pure unadulterated Creationist horseshit straight from the horse’s mouth. Still angling for a paying job as an anti-science propagandist for the DI? You sure have the unwarranted arrogance and profound scientific ignorance requirements covered.
Ok I get it, and you already answered my last question. *All* the deltas are falling on these 0.1% increments. I’m convinced, this is a manufactured dataset where someone backed out the percentage increments to raw vote counts. Nicely done!
I will eat all the crow you have available.
Your problem here seems to be that you read only creationist literature, which keeps you in a bubble. It’s rather like your Trump/Qanon bubble in that respect. If you would read the actual scientific literature you would see a different picture, but of course you won’t.
Your inability to count the taxa in your own data does not inspire confidence in your powers of judgment.
EricMH,
Oddly, I have the opposite experience. It was pretty noncontroversial already in the 1970’s, when I was in school. Getting to university, with lectures going into much more detail, it was apparent how rich the evidence was – and that was in the days when the only sequence data we had were a couple of proteins. Since then, sequences are being added to databases at an accelerating rate, and throughout these publicly searchable databases, the signal of common descent is metaphorically deafening, both in simple pairwise comparisons and in multigene, multi-species trees. That a low-level, non-morphological gene – succinate dehydrogenase, say – should so closely parallel the near-300-year-old Linnaean hierarchy based upon morphology is quite remarkable. Special pleading – we don’t know Gene X isn’t involved in morphology – soon loses traction, if every last gene’s consilience with the hierarchy has to be excused in that way. There must be some genes not involved in morphology, and indeed they appear to be in the majority.
What would you recommend? I have heard Stephen Jay Gould is pretty good.
Which gets to another problem. Some piece of literature is said to be really good. So I read it. I find some flaws. And then I’m told everyone knows about those flaws and now I should read something else. Again, a bit soggy.
Is there some really solid piece of writing on the Cambrian explosion that will adequately debunk all the criticisms and won’t turn out to be soggy?
Do you have some datasets I can analyze?
I don’t get this impression when looking at genetic data. E.g. if common descent is true then all triple comparisons of species based on some particular molecule should obey the ultrametric property. I tried this with mitochondrial DNA, and almost half of the comparisons did not obey the ultrametric.
And a straightforward reading of the CI scores in the Klaussen chart shows that we get something very non-tree like when comparing large number of taxa, which is why all my DAG results are corroborating.
You didn’t need to do anything. Just watch as living organisms repopulate the space. Life is a constant struggle for space and opportunity.
I don’t know what criticisms you mean, but for a good picture of the Cambrian explosion, I recommend two sources: The Cambrian Explosion by Doug Erwin and Jim Valentine, and a great review paper, Budd G.E., Jensen S. A critical reappraisal of the fossil record of the bilaterian phyla. Biological Reviews 2000; 75:253-295.
EricMH,
You’d have to be more specific about which mitochondrial sequences you looked at, and in which species. It’s odd that you can’t make it work at all, yet systematists routinely use it to resolve problem nodes.
Not true. Comparisons should obey the ultrametric property only it all lineages evolve at an equal rate, i.e. if there’s a perfect molecular clock. If evolutionary rates vary (and they do), ultrametricity is not expected.
You can download all sorts of data matrices, including the one from my croc paper, at treebase.org.
This is quite funny. A creationist expects the scientific literature to be dedicated to debunking creationist mythology about the Cambrian explosion! Oh, and all in one single article, please.
Sure, let me write a scientific article explaining that there’s no tigers among the Cambrian fossils, followed by another explaining that humans are not present among those fossils either … wait, no, just one that starts by explaining that there’s no humans, that there’s no tigers, no trees, no Adam and Eve there either, that there’s previous fossils of multicellular life, only not as dramatic, many with soft bodies (which left prints, rather than fossilized materials), that the “explosion” lasted millions of years, rather than one second, and let us not forget whatever new myths the creationists will imagine …
🤣
On reflection, you’ve given me the idea for an OP.
The specific criticism is that the cambrian explosion shows a great diversity of organisms showing up all of a sudden, contrary to the incrementalist approach necessitated by evolutionary theory.
Gould came up with punctuated equilibrium to explain this issue, saying sometimes we have very fast incremental evolution, and the species wander around, which gives the appearance of a sudden jump. I guess the idea is the ancestors to the cambrian explosion wandered off, evolved really quickly into a great diversity, and then all came back to the same spot.
Others say the predecessors were soft bodied, and so were not preserved in fossils.
If I recall, Stephen Meyer argues these are unsatisfactory resolutions to the problem, since with really fast evolution you don’t get much diversity, and there are plenty of preserved soft bodied organism.
So, if you have an article that debunks these specific arguments by Meyer, that’d be helpful.
Ok, that’s fine. I had read in Dawkins Blind Watchmaker this constant molecular clock hypothesis was the strongest piece of evidence for common descent that he knew of. So I guess that’s a soggy mess as well.
Basically, I’m looking for a claim like “X can only occur if evolution is true, here is a dataset Y proving X is true.” and then I can go out, get Y and prove X for myself, just like I can in every other rigorous science.
Dawkins’ claim got me excited, because it is exactly this format, but it turns out to be false.
Insufficient to generate the great variety of highly specified complex arangements of parts we see in the biological world.
Here’s the short list from my research we need for evolution to work:
1. protein function space very dense, so it is easy to hop from one functional protein to another with a flip or two of a nucleotide
2. very small increment pathways from one piece of complex proteins pieced together for a purpose to another
3. many, many such pathways
4. pathways well matched to environmental necessities for survival
The combinatorics here are incredible, and I’ve seen nothing indicating somehow these coditions hold for the real world. There are a few examples thrown out on these forums, and they are laughably inadequate.
If evolution were true, then cartoons like mutant ninja turtles and x-men are actually realistic, since it should be very easy to mutate organisms into more advance functionality that is well matched to their environment, just like these cartoons hypothesize.
Then there is the whole question of where genes come from in the first place. People say duplicate, random variation and genetic drift: poof you’ve got a new gene. There are so many problems with this it’s hard to take this as serious science.
EricMH,
Unless you can model and/or test this event based on real Biology the claim is at best speculation. Evolution can model changes in allele frequency in populations, however this starts with functional living or viral populations.
Right, it presupposes the very thing to be explained.
Well, not too bad, Bill. But evolution needs self-sustaining self-replicators to work. Viruses are parasites. They can’t have been the first replicators.
LOL! Baby ID-Creationist ErichMH and “a disembodied mind did it!” babbling Bill Cole giving each other hand jobs. 😀 Between the two I bet they regurgitate half the standard ID-Creationist bullshit talking points they’ve read from professional liars Meyer, Wells, Behe, etc. We’ve already had peppered moths, Cambrian explosion, too much “specified complexity”, yadda yadda yadda. Can’t these clowns ever think of anything new?
What do you mean ‘fed’? Were you taught that at school or university?
Also, research continues: https://advances.sciencemag.org/content/6/30/eabb4685
quite interesting really.
Alan Fox,
Viral populations evolve as the sequence changes over time. Again, population genetics starts with fully functional populations as physics starts with fully functional atoms.
So? Population genetics is only a tiny fraction of the huge amount of amassed positive evidence for evolution. You do need your security blanket to desperately cling to it seems.
It turns out that this great diversity doesn’t actually show up all of a sudden. Read Erwin & Valentine.
No, in fact he did not. He came up with it to explain apparent sudden transitions between closely related species. The Cambrian explosion has nothing to do with it. Please stop reading exclusively creationist literature; in particular, do not confuse Darwin’s Doubt with a work of science.
Page numbers and/or direct quote, please. It doesn’t sound like something Dawkins would say, and it’s a silly notion.
Adapa,
This is a nice sales pitch. It runs into trouble when you start to define your terms such as “amassed positive evidence” and “evolution”.
Population genetics is more than a sales pitch. It has a biological/mathmatical model that you can test. It’s a great theory.
.
So is the huge amount of positive evidence for evolution. We already know you’re too ignorant / dishonest / both to deal with any actual physical evidence. You’ve demonstrated that in spades across plenty of C/E discussion sites. I’m sure we’ll get more of your “a disembodied mind did it!” woo now, right?
Let’s take it by baby steps, what exactly do you mean by “all of a sudden”? After that, please explain why that would be “contrary an incrementalist approach necessitated by evolutionary theory.” That looks problematic in several levels, from mistaking the theory for the phenomena that the theory explains, to some hidden assumptions I cannot put my finger into. But start with “all of a sudden.”
If you’re willing to have a conversation, we can deal with the rest of your misinformation later.
Wikipedia says the “Cambrian explosion” occurred over a period of probably tens of millions of years. Other sources have told me that the phyla that apparently originated during this period probably had tens of millions of years to develop for which (after 550 million years or so) little evidence remains. Yet other sources say that the “new” phyla from this period were in many ways very similar, and are classified as different phyla today because, in hindsight, we know how lineages diverged since.
Taken together, I don’t see how one could regard the Cambrian explosion as particularly distinct, unusual, or in violation of any current understandings. Unless, of course, one is highly selective in their source materials.
Adapa,
I agree there is a lot of positive evidence for evolution depending how you define evolution. Population genetics is part of modern evolutionary theory. You have not shown capability yet to precisely define your terms.
Not sure what your point is, Bill, but evolution can only begin when there are self-sustaining self-replicators. Do you disagree?
There is, as yet, no single well-evidenced broadly accepted theory for how those first replicators came into existence but when it happened and what happened since is well documented and supported by swathes of consilient evidence.
A real live Gish Gallop, as I live and breathe!
Such as reading a sophist like Meyer.
Alan Fox,
Evolution begins with fully functional populations of different animal types. Any other claim is speculation that is without a model and/or a test.
Considering that you are unwilling to admit that any two species in the same genus are related by descent, what would you consider a test to be? Further, you should know that animals are not the beginning of evolution. Animals are opisthokonts, eukaryotes, and some combination of eubacteria and archaeans. The origin of life is deeper still.
Is that your latest cowardly excuse Bill? “Terms not precisely defined”? I suppose it’s a step up from your usual Creationist cowardly excuse “bad assumptions!”
I doubt that anybody would consider a constant molecular clock, if there was such a thing, the strongest evidence for common descent, let alone the strongest they knew about. Outside of molecular biology, I’d imagine that some people could think that there’s such expectation, a constant molecular clock, but even then I very much doubt they’d claim it to be the strongest evidence of common descent they knew about.
That’s one of the foundational problems with your OP. That’s not rigorous science. The scientific proposal would be, if evolution happens in way-Y, then X, we check and find X, then it’s possible that evolution works way-Y. That doesn’t mean that evolution works absolutely in way-Y, nor that no other process could produce such a result. It just means that if we expected to falsify evolution’s way-Y by testing for X, then we failed. After all, we cannot claim to know all the ways of nature. We’re just us, tiny human beings.
ETA: Of course, once we have tested several expectations from common descent, it does become “perverse” to deny it.
ETA: To fully appreciate why we don’t make such kinds of claims in science (or would be quickly sorry if we did), a simple exercise of the imagination should suffice: you believe in an all-powerful magical being. A being like that could, by definition, make everything look exactly the way expected from evolutionary scenarios. Given that, there’s nothing anybody can propose that would only be true if evolution happened. Whether that would conflict with the doctrines around your more specific magical being is a different story, and besides the point.