Sure. But that’s because the graph was created to fit the data, and it contains enough nodes and connections to fit the data perfectly. It was made, not discovered.
4 Likes
The astute reader will realize that a dependency graph model has an advantage over common descent in fitting the data because it can postulate modules to explain otherwise inexplicably distributed gene families. Consequently, it may seem that the dependency graph model will have a better fit to the data for almost any possible dataset.
It appears he is taking your objection into account. This is lifted from page 6.
And the current state of gene annotation for the various genomes.
1 Like
That’s not “taking into account”. That’s “mentioning”.
5 Likes
… are those two cited sources @Meerkat_SK5’s actual sources, or is he once again cutting and pasting from somewhere else that he hasn’t cited?
Let’s have another look at that last paragraph, and compare it with an article by Hugh Ross:
@Meerkat_SK5 is a terrible misciter. He keeps getting caught.
2 Likes
On page 7 he shows how he takes the added complexity into account.
Model selection theory provides the tools to answer this question. The fundamental idea is to penalize a model’s degree of fit based on the complexity of the model [44]. A more complex model is considered less fit to make up for its lack of parsimony. In the case of Bayesian analysis, this is done by treating each pa- rameter as a random variable. Instead of evaluating the goodness-of-fit of the biological data to a particular dependency graph, we evaluate the goodness-of-fit to a random dependency graph. Generally, this means that we compute the probability of the data using the law of total probability. For example, in the case of the dependency graph:
Pr[Gˆ = g] Pr[D|Gˆ = g] (1) g∈G
where G is the set of all possible dependency graphs, g is a particular graph, D is that data being explained, and Gˆ is the actual dependency graph as a random variable. This gives us the fit of the probabilistic average graph instead of a particular graph, thus avoiding postulating a particular complex graph. In practice, this summation is often intractable and will be bounded or estimated. Effectively, this leaves us with a new parameterless model which can then be compared to other parameterless mod- els.
Once the parameters have been resolved, we are left with two parameterless models that can be directly com- pared. This is done by computing the log Bayes factor:
log Pr[D|A] (2) 2 Pr[D|B]
where D is the data, A is one model, and B is the other model. This is the log-ratio of the likelihoods of the data under both models after removing all parameters by use of the law of total probability. The log-ratio will be positive if the data fits model A better than model B, and negative if the data fits model B better than model A. The numbers are on a logarithmic scale and, according to a widely cited table [45, p. 432], 6.6 bits is considered decisive.
It is important to note that when a Bayes factor favors one model over the other, it does not necessarily mean that the favored model is more likely to be correct or is the better explanation. It means that the particular evidence under consideration fits one model better than the other, but the other model may still be more likely to be correct if we take into account other information. Here we simply wish to test predictions about which model better fits different datasets. The argument for the dependency graph hypothesis rests on fulfilled predictions rather than Bayesian inference.
Not to mention actual taxon sampling and genome sequencing.
2 Likes
I fail to see it’s predictions. What does it predict, really?
1 Like
The hypothesis is predicting the data will follow model 3 from the images below where genes families exist as a “tool box” and are shared among species.
That doesn’t answer the question.
1 Like
Yes. Read his paper?
Yes. What is it that you think is his attempted explanation? Does it qualify as an explanation?
1 Like
Yes, I agree. Let me start over.
Why Universal design and descent are fundamentally different?
According to Fuz Rana:
"Closely related to the phenomenon of repeatable evolution is convergence. Convergence refers to the widespread tendency in nature of unrelated organisms to possess nearly identical anatomical and physiological characteristics. The wings of birds and bats is one textbook example of convergence. Birds and bats are unrelated organisms, with birds belonging to the class Aves and bats to the class Mammalia. Though superficially similar, the wing structures of birds and bats are fundamentally different. Another common example of convergence—one in which the fundamental structural differences are not so obvious—is the remarkable anatomical similarity shared by the modern placental wolf and the extinct Tasmanian wolf.
Both the creation and evolutionary paradigms offer an explanation for convergence. Creationists view convergence as the intelligent activity of a single Creator who employs a common set of solutions to address a common set of problems facing unrelated organisms in their quest for survival. Evolutionists assert that convergence results when unrelated organisms encounter nearly identical selection forces (environmental, competitive, and predatory pressures). Natural selection then channels the random variations believed to be responsible for evolutionary change along similar pathways to produce similar features in unrelated organisms."
As Fuz Rana pointed out, examples of convergent evolution can be explained by both models in regards to past events. However, due to the nature of random mutations, we model would expect a rare occurrence of them while the other model we would expect widespread examples if we assume mutations were directed.
For example, “A study by biologists now provides evidence that, at the molecular level, evolution is both unpredictable and irreversible. The study focuses exclusively on the type of evolution known as purifying selection, which favors mutations that have no or only a small effect in a fixed environment. This is in contrast to adaptation, in which mutations are selected if they increase an organism’s fitness in a new environment. Purifying selection is by far the more common type of selection.”
Evolution is unpredictable and irreversible, biologists show – ScienceDaily
This means that the inability of evolutionary processes to retrace the same path makes it highly unlikely that the same biological and biochemical designs would repeatedly appear throughout nature among unrelated organisms. On the other hand, evolutionary biologists have noted that evolutionary processes frequently seem to independently converge on identical anatomical, physiological, behavioral, and biochemical systems.
Universal Descent and Design are mutually exclusive
According to Gert Korthof in his book Why Intelligent Design Fails ,
"Common descent of life means that all life on Earth is physically, historically, and genetically connected. Common descent of life means that life is one unbroken chain of ancestors and descendants. Common descent of life means that every organism inherited all its genes from the previous generation (with slight modifications). And that includes irreducibly complex systems. Every supernatural intervention is a violation of common descent, because it means that a new irreducibly complex system in the first individual showing it was not inherited from its parents. It would be unjustified to say, ‘I inherited all my chromosomes from my parents, except an irreducibly complex system on chromosome X, which has a supernatural origin.’
Common Descent: It’s All or Nothing (updated chapter) (wasdarwinwrong.com)
Furthermore, as I argued previously, when researchers assumed that mutations were only additive and the effect of each mutation is done singly in experiments, the striking result of this design is that the simple additivity assumption was validated. The success of the method implied that additive mutations is big enough for engineering potent changes in activity. In other words, only directed mutations by a conscious agent has been shown to produce meaningful changes. On the other hand, the evidence showing that random mutations can produce the same potent changes has yet to be determined.
This means that universal common design and common descent are mutually exclusive. God cannot guide an entirely unguided process. Another reason why they are mutually exclusive involves how the first life forms came about.
For example, RNA viruses are irreducibly complex and cannot propagate without a host or intelligence. For this reason, it would require the designer to continuously create viruses to have them evolve into bacteria from within deep-sea vents AND force those bacteria to evolve further from other created viruses. As a result, this would require multiple origins, which happens to be consistent with current observations while common descent has no model for the first life.
Universal Common Design Theory
A Universal common designer re-used a DNA blueprint, RNA viruses, and chemical constituents to separately construct plants, birds, insects, herbivore mammals, carnivorous mammals, sea mammals, fish, crawling animals, and humans species.in different locations around the globe.
These original created species listed above would not be static but dynamic. They would be able to adapt to changing environments and diversify into kinds within a species over long epochs of time. This would involve the designer employing many familiar mechanisms ,such as gene drift, and a common set of solutions to address a common set of problems facing unrelated organisms in their quest for survival.
In other words, this model would also expect nested hierarchies within clades, but these hierarchies would be independent (unrelated organisms) where each stems from an organism created by fiat and filled out through secondary cause-and-effect reproduction. However, this means that created kinds within a species would need to be determined through observations and comparisons.
How do we determined this?
Finding function within vestigial features and/or looking for different environments that might show how a functional feature is fully optimized in comparison to another similarly constructed organism.
For example, contrary to what evolutionists have previously expressed about the "bad design, " of the giant panda’s thumb, A study analyzed it and showed that the "way in which the giant panda, Ailuropoda melanoleuca , uses the radial sesamoid bone - its `pseudo-thumb’ - for grasping makes it one of the most extraordinary manipulation systems in mammalian evolution… “The radial sesamoid bone and the accessory carpal bone form a double pincer-like apparatus in the medial and lateral sides of the hand, respectively, enabling the panda to manipulate objects with great dexterity.”
Following this publication, another study found that the giant panda and the red panda were not related even though both species possess the false thumb. The false thumb of the giant panda was intended to manipulate bamboo and the false thumb of the red panda was designed as an aid for arboreal locomotion, With the red panda secondarily developing its ability for item manipulation. [Just ask for references]
I am afraid it does. Please read this source for more:
A Common Design View of ERVs Encourages Scientific Investigation - Reasons to Believe
Because it is the primary mechanism God used to create diversity for advanced life on a biochemical level dating back to the first life form, which would be viruses.
I would agree with you that whether natural selection or common descent was universal or not is a separate question. But, as I explained before, random mutations and common descent are INseparable. If you insist otherwise, then please explain why and how common descent does not depend on random variation to produce the kind of diversity we would expect.
Yes, that kind of convergence – at the anatomical or functional level – is fairly common.
A fact that has nothing whatever to do with the frequency with which convergence occurs at the anatomical level. True convergence at the molecular level – converging on an identical protein from different starting points, for example – is extremely rare or nonexistent.
So yes, you have come up with a test for distinguishing common design from common descent. Common design fails that test completely.
3 Likes
My emphasis.
The likelihood of convergence at the molecular level is not equivalent to its likelihood at the other levels you mention there.
1 Like
Real replies don’t contain unmarked and uncited plagiarised paragraphs (from Rana and Ross, “What Darwin didn’t know”).
Start over once more.
1 Like
You’ve been corrected on this before, just 6 weeks ago. You’re implying the word “extraordinary” there means something like “good” or “most efficient” (“they showed it was… extraordinary”), when it actually just means “unusual”. The study says nothing about the thumb being “good” or “bad” design.
5 Likes
dont be sure about that:
https://www.nature.com/articles/nature.2013.13679
" The new study, published today in Nature, hints that evolution may be finding the same genetic solutions to a problem more often than previously thought."
in addition, what is the chance to get two similar creatures if we know that the sequence space is basically infinite?
1 Like
Yeah … no.
See Determining the Null Model for Detecting Adaptive Convergence from Genomic Data: A Case Study using Echolocating Mammals - PMC and No genome-wide protein sequence convergence for echolocation - PubMed for refutations of that study. Which is not to say that there is molecular convergence in genes related to echo-location, just that there isn’t very much of it.
More importantly, this is not convergent evolution by the definition I stated, and which I stated because definitions vary. I would describe these changes in hearing-related genes as parallel evolution rather than convergent, since they represent parallel changes to the same genes. That’s why I specified that the changes arise from different starting points.
An irrelevant question since we’re not dealing with infinite sequence space. For bat and bird wings, for example, we’re dealing with modest changes to tetrapod genomes.
4 Likes
That’s not what I asked for. I said “Deal with my first reply. Then deal with this one.” You have ignored everything and have launched into a new rant. Why should I bother even reading it, much less replying, when you have shown that you have no desire for any back and forth?
And your new post is full of non sequiturs and word salad.
That’s not agreement. That’s garbling.
Suppose God personally causes every mutation in just the way he likes it. Not random but guided. Isn’t that simple?
2 Likes