Jump to content

Help (or viability) on translating lineages into cladograms


Recommended Posts

Macrophyseter

When it comes to evolutionary lineages that tend to be represented by flow charts, would it be viable to also represent/reorganize them into hypothetical cladograms? Considering that flow charts continue to be used for certain lineages (i.e. Lamnidae), I am feeling the possibility that there might be something that makes the interpretation of such charts incompatible with cladograms.

 

For example, below are evolutionary lineages for Isurus and Carcharodon per Heim (1996) and Canevet & Lebrun (2018) (left and right respectively) that I translated into possible cladograms. If this is a viable thing to do, I'm also curious about how anacladogenesis can be represented.

 

image.png.12b6cf1a6fd8a1ec41245c5b3a0532ae.png          image.png.0fb07111e39832b4e87fd39b5cd9de6d.png  

 

image.png.38078fe3cef957ae7adeea231fd6de46.pngimage.png.8a003d0d9b3cec7977bced4c35d2e840.png

Edited by Macrophyseter
  • I found this Informative 1
Link to post
Share on other sites

question:  when you say you translated them into cladograms, how did you do that?  Cladgrams are derived from amassing huge lists of charcteristics for all the animals involved and comparing them all to each other and deciding which are the ancestral condition and which are shared derived characters.  With more than a few characters in the mix, it needs to be done with a computer.

  • I found this Informative 2
Link to post
Share on other sites
Macrophyseter
13 minutes ago, jpc said:

question:  when you say you translated them into cladograms, how did you do that?  Cladgrams are derived from amassing huge lists of charcteristics for all the animals involved and comparing them all to each other and deciding which are the ancestral condition and which are shared derived characters.  With more than a few characters in the mix, it needs to be done with a computer.

I failed to specify that they would be hypothetical cladograms, as in it would be the cladogram you would predict to get based on the current lineage data if you were to build and analyze a character matrix. In this case, the assumption would be that the progression of each species in the lineage flow charts are somehow representable in a cladogram under the thinking that two taxa can be sister groups and "share" a most recent common ancestor if that common ancestor is one of the two taxa.

 

Of course if I were to want to do something beyond hypothetical I would need to collect data in a study way beyond my ability, although I do have the computer programs to analyze them.

Link to post
Share on other sites

This is an interesting question. We've generally abandoned linnean taxonomy for larger clades, but teaching the basics of cladistics to the public is... really aggravating. To most folks the trees don't really mean anything, and even less so if they're not time-calibrated.

 

The point of all this is... it may not be worth the effort. Though, within paleoichthyology, I'd always love to see something like that.

  • I found this Informative 2
  • I Agree 1
Link to post
Share on other sites
Macrophyseter
3 hours ago, Boesse said:

This is an interesting question. We've generally abandoned linnean taxonomy for larger clades, but teaching the basics of cladistics to the public is... really aggravating. To most folks the trees don't really mean anything, and even less so if they're not time-calibrated.

 

The point of all this is... it may not be worth the effort. Though, within paleoichthyology, I'd always love to see something like that.

 

I guess you could say that lower-level cladistics is always going to be pretty esoteric, but I guess that's a reason why some people would take an interest in them. As a hobby, at its least significance.

  • I found this Informative 1
Link to post
Share on other sites
Macrophyseter
Posted (edited)
3 hours ago, Boesse said:

We've generally abandoned linnean taxonomy for larger clades

Is Linnaean taxonomy on a general trend towards total obscurity, and in the next ten or more years scientists may as well abandon the concept entirely (except for perhaps the binomial system)? A bit intriguing to consider the biology classes of tomorrow no longer teaching about reptiles but instead synapsids.

Edited by Macrophyseter
Link to post
Share on other sites

So the answer is....sort of, but you lose information.

 

The common ancestor (CO) of the clade containing Taxon A and Taxon B should form a basal polytomy within that clade, because A, B, and CO share a common ancestor, which happens to be CO. The key here that distinguishes ancestors and descendants is branch length ,i.e. the amount of inferred/estimated evolution in each lineage. In this hypothetical example, A and B both have positive branchlengths, but CO has a branch length of zero, as no evolution has occurred between it and itself.

 

When you convert that to a strict cladogram, you lose branchlength information, which means it is impossible to know which of those three is the common ancestor from the cladogram alone.

 

  • I found this Informative 3
Link to post
Share on other sites
On 4/11/2021 at 5:36 PM, Macrophyseter said:

Is Linnaean taxonomy on a general trend towards total obscurity, and in the next ten or more years scientists may as well abandon the concept entirely (except for perhaps the binomial system)? A bit intriguing to consider the biology classes of tomorrow no longer teaching about reptiles but instead synapsids.

 

It's complex. Remember that Linnean taxonomy was devised and developed at a time when the taxonomy of life was thought to reflect a divine plan of organization, which is why many levels of taxonomy were originally grouped in fives. Now that we understand evolution and common descent we understand that there is no insight into the divine plan that comes from understanding a taxon to sit at one level or another.

 

However, aspects of the Linnean system have been retained in our modern bookkeeping of taxonomic diversity, specifically the levels of species, genus, and family. We are unlikely to eliminate these because that bookkeeping and nomenclature is still very useful. However, most workers don't view these things as more than just useful bookkeeping tools.

  • I found this Informative 3
Link to post
Share on other sites
On 4/11/2021 at 11:33 AM, Macrophyseter said:

I failed to specify that they would be hypothetical cladograms, as in it would be the cladogram you would predict to get based on the current lineage data if you were to build and analyze a character matrix. In this case, the assumption would be that the progression of each species in the lineage flow charts are somehow representable in a cladogram under the thinking that two taxa can be sister groups and "share" a most recent common ancestor if that common ancestor is one of the two taxa.

 

Of course if I were to want to do something beyond hypothetical I would need to collect data in a study way beyond my ability, although I do have the computer programs to analyze them.

 

 

In general....yes. There are actually a suite of tools, broadly called stratocladistics or stratophylogenetics, which allow inference of ancestor-descendent relationships (i.e. anagenesis) in the modern phylogenetic framework of approaches. I can point you in the direction of some specific papers and tools if you're interested.

  • I found this Informative 2
Link to post
Share on other sites
Macrophyseter
2 minutes ago, jdp said:

In general....yes. There are actually a suite of tools, broadly called stratocladistics or stratophylogenetics, which allow inference of ancestor-descendent relationships (i.e. anagenesis) in the modern phylogenetic framework of approaches. I can point you in the direction of some specific papers and tools if you're interested.

 

I would sure love to take a lead on it.

 

Is there anything exclusively unique about stratophylogenetics compared to time-explicit clocks like Bayesian (or would it be considered that)?

Link to post
Share on other sites

So tip-dated explicit clocks in Bayesian frameworks are not quiiite stratophylogenetics. The reason for this is that these methods still do not directly establish ancestor-descendent relationships. That has a few impacts on the resulting tree, specifically that you have inferred ghost lineages in the tip-dated Bayesian tree which would be eliminated in a stratophylogenetic framework. You will never quite get anagenesis in most Bayesian approaches.

 

That said, people still use the Bayesian tools because they are relatively easy to access and are useful for a lot of different things. This is sort of standard in science where less than perfect tools are used over and over because they're the tool someone has learned.

 

In terms of stratocladistics, I've been out of that world for a while now, but a good starting point would be Marcot & Fox 2008, which reports their program StrataPhy, which was an attempt to create a program for running stratocladistic analyses:

 

https://palaeo-electronica.org/2008_1/142/index.html

 

There are some good citations in there as well.

 

I can dig some other references up for you if you're interested. I published in this little niche corner of the field some time ago, but it's really been a long while and I'd have to dust off that part of my brain.

  • I found this Informative 4
Link to post
Share on other sites
pachy-pleuro-whatnot-odon

While I can in no way match the excellent answers already given in this thread, I thought I'd just dumb this down a bit to my understanding of the discussion. For I think the original question showed some confusion about data versus presentation. For while a cladogram is typically presented as a binary tree, the data presented in it doesn't relate to ancestry, but rather to semblance in a predetermined set of features. This is beautifully illustrated in Zietlow's (2020) article on the analysis of ontogeny, sexual demorphism and possible anagenesis in American Tylosaurus species. Here cladograms are used to explore whether the known population of a species can be subdivided into binary groups representing different sexes, analyse whether two tylosaurine species may be equated as one representing juveniles of the other, and test the hypothesis that one Tylosaurus species evolved into the other by way of anagenesis.

 

And while ancestry often also leads to commonality of characters, this is not necessarily the case. A good example of the latter are the cladograms drawn up for nothosauridae, where certain chronostrategraphically late (if not terminal) species of nothosaur are grouped together with early ones based on their shared characteristics, rather than ancestry being applied.

 

In conclusion, of course you can use a tree diagram to represent evolutionary development. Just be careful not to confuse it with, or call it a cladogram.

  • I found this Informative 2
Link to post
Share on other sites

Yes, with caveats. The main caveat is that the methodology contains explicit character transformation models so the implication of any cladistic analysis is that the tree itself will reflect patterns of shared ancestry. This creates a few problems for the whole using-phylogeny-to-look-for-ontogeny approach that people like Carr and, here, Zeitlow, have used, and personally I would not recommend this approach, as both ontogeny and sexual dimorphism do not actually follow the assumptions of the toolset. Personally I recommend use of ordination methods rather than phylogenetic toolsets to analyze these sorts of datasets. This falls into my above point about people using the tools they know rather than the tools tat are appropriate for the job at hand.

 

In the context of nothosaur phylogeny, I don't know that part of the tree very well, but some likely-approximately-true phylogenies have really poor stratigraphic consistency. There are biological explanations for this in many cases. It is also possible that there are known biases within the datasets used to infer nothosaur phylogeny, but I really don't know enough about that group to speak to the details there. Where computational phylogenetics disagrees with historical hypotheses of relationships and lineages that were assembled by hand, typically we favor the computational approaches these days, as they are reproducible in ways that by-hand hypotheses are not.

  • I found this Informative 1
Link to post
Share on other sites
siteseer
On 4/11/2021 at 12:45 PM, Boesse said:

This is an interesting question. We've generally abandoned linnean taxonomy for larger clades, but teaching the basics of cladistics to the public is... really aggravating. To most folks the trees don't really mean anything, and even less so if they're not time-calibrated.

 

The point of all this is... it may not be worth the effort. Though, within paleoichthyology, I'd always love to see something like that.

 

Hi Bobby,

 

This is a great discussion.  For people of my generation or of a certain level of science-mindedness, we quickly get lost in the edges of the weeds of cladistics.  Oh no, I don't want math mixed in with my sharks!  When I talk about cladistics, I start at the most basic point and that is it is all about establishing ancestors and descendants - not groups that seem to be related to each other.  At that point, the question is, "Isn't that the same thing because when you work out degrees of relatedness, you figure out ancestors and descendants?"  At that point I talk about reptiles. 

 

When I was a kid, we learned that turtles, lizards, snakes, crocodilians, and the tuatara are the living groups of reptiles and that there are many more extinct groups.  If it had feet and hard scales, it was a reptile.  Even then, however, I was confused because there was still something in the books I read that made it apparent that turtles were different from the other groups.  Also, when I read about the genus Seymouria, it was often called a "link" or an "intermediate" between amphibians and reptiles even though it lived at the same time as Dimetrodon, Edaphosaurus, and other named reptiles.

 

I think people in their 30's and younger understand this better because they grew up with the term, "non-avian dinosaurs."  That is a scientific term but the kids accept it quickly and you teach cladistics starting there.  I understand it only a little better because I read "Evolution of Insects" by Grimaldi and Engel (Cambridge University Press, 2005).  That book gave me more of an idea of how paleontologists see the "tree of life."  Mantises only appeared to be the predatory relatives of grasshoppers .  They share particular modifications of anatomical characters with roaches and termites that don't appear to be just similarity nor coincidence.  They're indicators of common ancestry not seen in grasshoppers.  It took a lot of time and study along numerous lineages since the 60's to get to that level of understanding.  We thought Aristotle did a lot to try to classify animals in his time but a lot of us also thought it was pretty simple and that we actually had it down in the 70's.  Now, these kids today are showing us we were also just starting to play with blocks.

 

Jess 

  • Learned something new 1
Link to post
Share on other sites

@siteseer I'm a bit of a nihilist on this because some former colleagues [who FORTUNATELY resigned their posts] designed and wrote a new lab that had students fill in parts of a cladistic tree of vertebrates, with emphasis on different dinosaur groups for our intro geology/earth history labs. It was a total nightmare. Granted, these are often business/communications majors with very little interest in the natural world. Students can understand an ancestor-descendant relationship but get all bent out of shape when they start trying to think about what features group different species together. Now, I'm also coming from a perspective of teaching young adults who may not necessarily know what salamanders are or marsupials, or really even what the term 'mammal' means. Maybe 1 in 10 students knows what a platypus is. Point is, the folks who wrote this cursed lab left and we ended up just ditching the cladistics part because it was kind of hopeless and we weren't sure what the students were actually getting out of it. Half the time I would sit down with individual groups and have to walk them through it so I could leave lab on time.

 

So, while cladistics remains the major method we use to interpret evolutionary history, it is so nuanced and difficult to teach that you end up wasting a lot of time teaching students how to interpret a cladogram that it often takes away from the point of the course - learning earth history and the change in lithosphere and biosphere through time - that it's best, IMO, to completely sidestep it for general audiences. If there's an easier and quicker way to teach cladistics to adults indifferent to the natural world, we haven't found it yet. I may also be overly pessimistic right now because it's final exam season here...

 

@jdp It took me *way* too long to realize that was you hahaha

Link to post
Share on other sites
LabRatKing
On 4/11/2021 at 2:45 PM, Boesse said:

This is an interesting question. We've generally abandoned linnean taxonomy for larger clades, but teaching the basics of cladistics to the public is... really aggravating. To most folks the trees don't really mean anything, and even less so if they're not time-calibrated.

 

The point of all this is... it may not be worth the effort. Though, within paleoichthyology, I'd always love to see something like that.

 I battle with platyhelminth cladograms  on a regular basis. The traditional linean trees are become useless once the results from transcriptomes come in.

 

I try to avoid them all together.

Link to post
Share on other sites
2 hours ago, Boesse said:

@siteseer I'm a bit of a nihilist on this because some former colleagues [who FORTUNATELY resigned their posts] designed and wrote a new lab that had students fill in parts of a cladistic tree of vertebrates, with emphasis on different dinosaur groups for our intro geology/earth history labs. It was a total nightmare. Granted, these are often business/communications majors with very little interest in the natural world. Students can understand an ancestor-descendant relationship but get all bent out of shape when they start trying to think about what features group different species together. Now, I'm also coming from a perspective of teaching young adults who may not necessarily know what salamanders are or marsupials, or really even what the term 'mammal' means. Maybe 1 in 10 students knows what a platypus is. Point is, the folks who wrote this cursed lab left and we ended up just ditching the cladistics part because it was kind of hopeless and we weren't sure what the students were actually getting out of it. Half the time I would sit down with individual groups and have to walk them through it so I could leave lab on time.

 

So, while cladistics remains the major method we use to interpret evolutionary history, it is so nuanced and difficult to teach that you end up wasting a lot of time teaching students how to interpret a cladogram that it often takes away from the point of the course - learning earth history and the change in lithosphere and biosphere through time - that it's best, IMO, to completely sidestep it for general audiences. If there's an easier and quicker way to teach cladistics to adults indifferent to the natural world, we haven't found it yet. I may also be overly pessimistic right now because it's final exam season here...

 

100%. I think part of this is because some paleontologists and paleo-adjacent bloggers put undue importance on a few features of cladistic taxonomy (particularly "omg birds are living dinosaurs") and because some of the tools for analyzing and displaying phylogenies are available for free. As a result, there is a sense that phylogenetics is the most important thing we are doing (not true) and also that it is the place where amateurs can make the most important and lasting contributions to the field with minimal effort. I think this is unfortunate because the best phylogenetic work is an outcome of substantive specimen-based work (collection of new fossils, description of these fossils, etc.) rather than a goal in and of itself. I say this as someone whose most-cited and, likely, most impactful papers are both heavily phylogenetic in nature.

 

 In terms of communicating cladistic thinking, the way I normally explain it in the classroom is by having people think explicitly about common ancestors. I don't even use trees for this; I normally use Venn diagrams because that heads off the tendency of cladograms to guide people towards thinking about directionality and advancement. Then I show how a cladogram is a representation of the Venn diagram. I normally start this off at the level of family trees where shared common ancestors are intuitive and have names (i.e. my brother and I have a common ancestor, who is my mom, and we share a common ancestor with my cousin, who is my grandmother....etc). In my experience, that helps.

 

2 hours ago, Boesse said:

 

@jdp It took me *way* too long to realize that was you hahaha

 

Ahaha you mean the fact that I use the same pic here and one twitter didn't immediately tip you off?

  • I found this Informative 1
Link to post
Share on other sites

THANK YOU for all that.  Also, yeah I had to double check the picture... "hmm that looks familiar and so do those initials..."

Link to post
Share on other sites

Yeah no problem. Tree-thinking is a really hard skill to teach so I've given it a LOT of thought. I haven't used that in years though, as I mostly just teach anatomy these days.

Link to post
Share on other sites
FossilDAWG
10 hours ago, Boesse said:

I'm a bit of a nihilist on this because some former colleagues [who FORTUNATELY resigned their posts] designed and wrote a new lab that had students fill in parts of a cladistic tree of vertebrates, with emphasis on different dinosaur groups for our intro geology/earth history labs. It was a total nightmare. ...

Career termination for putting together a poorly thought out lab seems a bit harsh.  No room for error I guess?

 

Don

Link to post
Share on other sites

having taught someone else's badly-thought-out labs before, career termination is a light sentence

Link to post
Share on other sites
FossilDAWG
1 hour ago, jdp said:

having taught someone else's badly-thought-out labs before, career termination is a light sentence

Seriously?  Work your way through undergrad, MSc, PhD, two postdocs, finally get a faculty position, train a dozen masters and PhD students, bring in millions in external funding, publish a ton of papers, teach more than anyone else in the department, but one crappy lab exercise and you're done?  Out the door with a kick in the pants?  Is that really how it works in paleontology?  I thought you had to something serious to have your tenure revoked and be shown the door, something like sexual harassment.  If I had known this when I was an undergrad I would probably have been able to get at least 50% of my professors fired.  I wonder what punishment an entire course of boring lectures would merit.

 

Don

Link to post
Share on other sites

Hah, @jdp was kidding. I won't go into specifics but they voluntarily left, for unrelated reasons, and that was extremely fortunate.

Link to post
Share on other sites

Same, but I will say that I have taught some very poorly-designed labs as a TA, and the worst designed of those were very, very badly designed, to the point where one (a group presentation project) literally ended up promoting eugenics. In just about every case it was very obvious that the lab activity was likely to go off the rails but TAs seldom have the authority to revise these sorts of projects but end up having to deal with the fallout.

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...