Branching Trees - A question for evolutionists

[Ophiolite]

Your claims about cars and trucks were demonstrated to be empirically false with respect to constructing phylogenetic trees.

So either:
A) You don't understand enough about phylogenetic trees to know why the above claim was shown to be wrong;
B) You can't remember any of those past discussions; or,
C) You just don't care.

Which is it?

There is a fourth alternative: he is employing the technique laid out in the quote attributed to Goebbels, "If you tell a big lie often enough and keep repeating it, people will eventually come to believe it."

 

[xianghua]

Cars and trucks accumulate change and pass it on? How does that work?

irreleant since we are talking about classification. and we can do that with designed objects too.

What criteria are employed to make that 'tree'?
In actual science, shared derived characters are employed.

level of general similarity.

 

[pitabread]

irreleant since we are talking about classification. and we can do that with designed objects too.

It's not just classification. Phylogenetic trees are by definition based on relationships defined by inheritance. That is their whole purpose.

This is why phylogenetic trees are not used for designed objects, because designed objects lack that inheritance.

 

[xianghua]

Why don't you group the vehicles by engine type? Or manufacturer? Or number of axles? Or load capacity? Or colour? Or function? Or place of manufacture? Or design year? Or . . . . . . ?

because if we will do that we can get many different trees since we we are dealing with a single trait. but it might be possible to get the same tree in most cases. for instance: most cars use basically the same kind of wheels. so if we will make a tree out of wheels kind i think that we may get the same tree in most cases.

 

[xianghua]

Because your example is worthless. All you're going is "that thing looks like a car, so it goes in the car group. That thing looks like a truck, so it goes in the truck group."
The cladograms in the evolutionary 'tree' show actual progression in species from basal to modern, and also showing where each group branches off.
Your 'idea' falls at the first hurdle which is that bikes came before cars. And chariots came before bikes or cars. And carts came before bikes or cars. And TRAINS came before either cars or bikes but after chariots and carts.

first: we also find such "out of place" fossils in evolution. so i see no difference. second: we can solve it by convergent designed. on the same base of convergent evolution.

 

[xianghua]

Except they don't.

We've been over this several times in the past. Your claims about cars and trucks were demonstrated to be empirically false with respect to constructing phylogenetic trees.

so a tipical car in general isnt more similar to other car than to a truck?

This is why phylogenetic trees are not used for designed objects, because designed objects lack that inheritance.

so whats that?:

Modeling with Nonliving Objects to Enhance Understanding of Phylogenetic Tree Construction

 

[pitabread]

so a tipical car in general isnt more similar to other car than to a truck?

Not when it comes to constructing a phylogenetic tree, no.

We've been through this before, remember? I tested your claim by creating a matrix of characteristics for various cars and trucks, and generated phylogenetic trees. The results were that trucks and cars were mixed up with no clear deliniation between them.

And the reason makes perfect sense: most of the physical characteristics of cars and trucks are not used for vehicle classification.

so whats that?:

Modeling with Nonliving Objects to Enhance Understanding of Phylogenetic Tree Construction

This is simply an example used for teaching the concepts of phylogenetic trees. It's easier to use artificial objects for the purpose of teaching because students are more likely to be familiar with said characteristics.

However, the resulting tree is meaningless because phylogenetic trees are based on the principles of inherited traits.

I would not recommend confusing the above example for the purpose of the teaching with the actual phylogenetic trees used in biology and the intended use of showing relationships based on inheritence.

 

[pitabread]

because if we will do that we can get many different trees since we we are dealing with a single trait. but it might be possible to get the same tree in most cases.

It's not. Remember, I empirically tested this and it turns out using different subsets of characteristics for vehicles results in completely different trees.

We've gone over all of this several times now in the past year.

 

[SLP]

FWIW, when xianghua first made this claim on this forum, I did an experiment constructing trees based on character matrices for cars and trucks (using Mesquite).

It turns out that cars and trucks don't necessarily fall into the neat groupings he thinks they do. And when constructing trees using different subsets of characteristics, I didn't get statistically convergent trees.

In my experience, creationists do not generally understand what goes into making phylogenetic trees.
Do you like Mesquite? Had a computer upgrade and PAUP is not working correctly, and am looking for another go-to program.

 

[SLP]

irreleant since we are talking about classification. and we can do that with designed objects too.

And yet, cladistics relies not on "general similarity" but on extent of shared derived characters.

If actual biologists adopted your naive criterion, then surely whales would be cousins of large fish.

level of general similarity.

Define what you mean by "similarity" and provide a means of quantifying it please.

 

[SLP]

first: we also find such "out of place" fossils in evolution.

Then surely you can provide examples and accompany them with a coherent analysis?

so i see no difference. second: we can solve it by convergent designed. on the same base of convergent evolution.

So you have constrained your deity to a few basic forms.

 

[SLP]

so a tipical car in general isnt more similar to other car than to a truck?

So I guess this must be a transitional:

so whats that?:

Modeling with Nonliving Objects to Enhance Understanding of Phylogenetic Tree Construction

A bad overly simplistic example/analogy to convey a basic concept to simpletons.

 

[SLP]

first: we also find such "out of place" fossils in evolution. so i see no difference. second: we can solve it by convergent designed. on the same base of convergent evolution.

Right - I forgot that molecular phylogenetics has been falsified by this simple analogy -

Starting sequence:

ABCDE

mutation rate = 1 letter per generation

BBCDE

BCCDE

BCDDE

BCDEE

BCDEF

WHAAAAAA??????

The sequence is TOTALLY different after only FIVE generations!!!!

Cladistics and therefore evolution is totally wrong!


I actually had a creationist present that to me as an argument about 15 years ago. He was so very certain that he had disproven a major evolutionary bit of evidence, even after I explained that for his claim to have merit, real mutation rates would have to be increased by about 10 orders of magnitude and this would kill all living things.
Nope...

He was right.

Can't fix stupid, as the bumper sticker on the wannabe monster truck says.

 

[Warden_of_the_Storm]

first: we also find such "out of place" fossils in evolution. so i see no difference. second: we can solve it by convergent designed. on the same base of convergent evolution.

But your idea isn't about 'out of place fossils', or convergent design or evolution.
It is you just spouting the same nonsense time and again, after being told repeatedly how it doesn't work, even I've shown you how it doesn't work, but you keep repeating it.

STOP IT!

 

[pitabread]

In my experience, creationists do not generally understand what goes into making phylogenetic trees.

That is definitely the case here.

Do you like Mesquite? Had a computer upgrade and PAUP is not working correctly, and am looking for another go-to program.

It seemed functional, but buggy. I found it would stop responding after awhile, so I learned to save often.

 

[SLP]

A phylogenetic tree is simply the idea of branches of taxa. You need not impose the religion of Evolution onto it, but you can if you want.

By the way:

Addition to the Statement of Purpose

Christianity cannot be called a myth, and science cannot be called a religion or made up. Threads started, or responses made, to simply disparage science will be considered off topic to the forum. ​

 

[xianghua]

If actual biologists adopted your naive criterion, then surely whales would be cousins of large fish.

no. since in general a whale is more similar to a land mammal than to a fish. when is say "general" i refer to the whole creature, including internal morphology and not just external.

 

[xianghua]

Not when it comes to constructing a phylogenetic tree, no.

We've been through this before, remember? I tested your claim by creating a matrix of characteristics for various cars and trucks, and generated phylogenetic trees. The results were that trucks and cars were mixed up with no clear deliniation between them.

And the reason makes perfect sense: most of the physical characteristics of cars and trucks are not used for vehicle classification.

yes i remember. but you need to do so by testing most of their parts and not just few of them. otherwise i can say the same about these cases that we can find in living things too:

Gene Study Shows Turtles Are Next Of Kin To Crocodiles And Alligators

43 genes support the lungfish-coelacanth grouping related to the closest living relative of tetrapods with the Bayesian method under the coalescence model

so when the number of traits is so low, we can get wrong results.

 

[pitabread]

yes i remember. but you need to do so by testing most of their parts and not just few of them.

You claimed this previously and my response is the same as last time:

1) Statistically speaking you don't need the majority of components since like everything in statistics, you start running into diminishing returns. Now if you want to claim that I didn't use enough components, then you'll have to demonstrate that. I await your calculations.

2) The classification of vehicles is based on only a few characteristics. The majority of characteristics in cars and trucks have nothing to do with their classification. Thus, we have no reason to expect that cars and trucks would sort themselves into perfect categories even if we used every single component in them.

Regardless, we're back to the same issue as last time: you are making testable claims but you are not testing them.

So if you think that using the majority of vehicle characteristics will result in phylogenetic trees with cars and trucks nicely sorting themselves into completely different groupings, then show us. Download the software and test your claims.

Until you do, all you have is a bunch of unsupported claims that when tested turned out not be true.

edited to add:

This also contradicts your earlier claim that "a car in general is more similar to other car than to a truck." If this were true then even a subset of characteristics would result in discrete groupings.

 

[xianghua]

Now if you want to claim that I didn't use enough components, then you'll have to demonstrate that. I await your calculations.

you used something like 10-15 traits. i showed that even by using more than 20 traits (genes) in living things we can get different result. thus using just few traits should be problematic.

2) The classification of vehicles is based on only a few characteristics. The majority of characteristics in cars and trucks have nothing to do with their classification

how do you get to that conclusion? does your car has a truck engine? doest your car has a truck wheels? doest your car has a truck size? im almost sure that the answer is "no" since most cars dont have trucks traits.

Regardless, we're back to the same issue as last time: you are making testable claims but you are not testing them.

i do think that we can do that. lets start with simpler comparison- bicycle and a car. do you think that in general a regular bicycle is more similar to other bicycle than to a car?