Dr Page's best example of common descent explained from the GUToB.

Hi Scott,I just wanted to add a little bit to this: For the benefit of others trying to follow along I'd like to reinforce what Scott is saying here. For large population sizes, the confidence level of a statistical measure is purely a function of sample size, not of the population size. In other words, the confidence level of a sample size of 2000 is the same regardless of whether the population is a million or a billion (if memory serves me correctly, a sample size around 1785 gives a confidence level of 95% that the result is accurate within a standard deviation.Of course, this is only true for randomly selected samples, and most of the effort has to go in to making sure the selection scheme is random. For instance, instant telephone polls after a presidential speech or press conference are notoriously unreliable because the sample isn't random. It's a selection of people who tend to be home in the evening, aren't already on the phone, and agree to answer polls.I suspect Scott's sample isn't random, either, though I have no concrete information to go on. But I suspect that the sample consisted of those areas of the chimp genome that happened, for one reason or another, to have already been studied, or which were known because they were easy to study (the "looking under the lampost" effect), and so forth.If you suspect your selection method isn't random then you can negate that problem by increasing the sample size. If Scott's sample size is substantially above a few thousand then the confidence level might be very, very high.--Percy

This file: introduction to molecular phylogeny (pdf) gives a good description of the various techniques used to produce phylogenetic trees, how to interpret trees etc.

Is there any way that through this knowledge we can have an idea of what the perfect human would be like and his/her capacities? as to there being no degeneration what so ever in this beings coding?

quote:

---

Percy:\I suspect Scott's sample isn't random, either, though I have no concrete information to go on. But I suspect that the sample consisted of those areas of the chimp genome that happened, for one reason or another, to have already been studied, or which were known because they were easy to study (the "looking under the lampost" effect), and so forth.

---

That is why I put "effectively" in there. Frankly, I don't know how one would go about 'randomly' sampling genomic data. Previous discussions on such topics have yielded the predictable responses - links to websites showing that 'random' isn't really 'random', claims that any 'choice' being involved at all - such as which species to compare or which locus to seqeunce - invalidates the results, etc. etc.The 'randomness' of genetic analyses are limited by a number of factors - among them, which loci you will actually be able to sequence, which species you will be able to get genomic DNA from, etc.So, avoiding the semantics quibbles over what is and is not truly random, the point was that as you allude (and I also had mentioned) is that samples from all over the genome have been analyzed. I could certainly understand the skepticism of conclusions based on, say, data from only one chromosome, or only from protein coding sequence. I would be skeptical myself - that is why I am always skeptical of the claims of someone having 'refuted' evolution by, for instance, looking at one small 350 base pair partial exon from a mitochondial genome - individual loci - especially small ones - can produce artifactual errors. Better to base conclusions on larger, more diverse datasets, where any potential "outliers" can ber identified and dealt with.

Hi Scott!If I wasn't clear, I was only trying add a little background to what you had said, and repeat some of it in different words. I don't think we disagree, and any uncertainly I expressed about the statistical reliability of your results is based on my lack of familiarity with it and nothing else.But unfortunately we live in a math-challenged country, perhaps world even, and so even this undetailed statistical discussion is inaccessible to many, particularly those who followed a curriculum focused on the arts (English, history, art, music, philosophy, etc).--Percy

One of my main GUToBjections is that PB seems to
be restating the 'common descent cannot be distinguished
from common design' reasoning that many creationists
get to.Evidence for GUToB is the same as for a more mainstream
ToE interpretation ... so even if the data really matched
GUToB very well it doesn't refute an alternative theory!!One theory does not refute an equally well supported one.The claim that PB has disprooved ToE is contentious and mis-leading.If he is simply putting forward an alternative, fine, that's
what is being discussed and elaborated ... but to then say that
it disproves neo-darwinian evolution is just wrong.

Hi Peter,In response to:Admin asked me to repost this here as it was more relevant
to this discussion ... I'll delete it from the
GUTOB thread.
The following is how Peter Borger describes NRM:quote:
--------------------------------------------------------------------------------
1) Non-Random Mutations type 1 (NRM1).
This type of NRM introduces mutations on the same positions and the
position where they are introduced depends on the DNA regions where it is part of.
NRM1 are also known as ‘positional NRM’.
NRM1 has been described for T4 virus.
NRM1 is likely to be present in the ZFY region and in mtDNA.
--------------------------------------------------------------------------------Some locations in DNA sequences are more likely than others
to suffer copy errors.That is statistically more likely.Decsribing this as 'non-random' is spurious at best. Randomness
is takne, generally to mean, unpredictable ... which is what is
seen. Sometimes the location appears to have changed, sometimes
not. There is no known causative link.PB: We've discussed this before.quote:
--------------------------------------------------------------------------------
Semi-Random positional mutations are a special form of NRM, i.e. semi-random with respect to nucleotide, but
non-random with respect to position (=when mutations occur in the region they are always introduced at the same spot).
--------------------------------------------------------------------------------Peter: 'Semi-random' does not make sense at all.PB: Well, how does 'quasi-random mutation (QRM)' sound? It holds that the position of the mutaion is always the same, while the nucleotide varies.Peter (cont): The changes seen are either caused by a mechanism or are
copy errors.The above observation would tend to undermine the idea of a
directed mechanism involved in copy errors.PB: What matters is that a mechanism is involved that introduced these mutations. Whether it is copy errors or another mechanism doesn't matter. For the 1G5 in Drosophila gene I think it is protein/RNA driven. As long as they are introduced on the same spot --irrespective the mechanism-- they will give the illusion of common descent (as demonstated for the ZFY region, mtDNA, 1G5 gene, etcetera)quote:
--------------------------------------------------------------------------------
Implications: NRM1 will line up and give the illusion of common descent in phylogenetic analysis. Since phylo-geneticists cannot exclude NRM1, this type of mutations question the evolutionary conclusions that the alignment of genes and shared mutations is proof for common descent.
--------------------------------------------------------------------------------Peter: That mutations often occur at a particular locus does not
undermine the common descent hypothesis.PB: As long as you do not the origin (random or non-random) of the mutation common descent based on sequence analysis is conclusion jumping.Peter: Perhaps common descent gives the illusion of directed evolution.PB: So, we have another uncertainty?Peter: The existence of alignments is not considered proof of common
descent ... it is one evidence for it.PB: No, it is not as long as you cannot exclude NRM.quote:
--------------------------------------------------------------------------------2) Non-Random Mutations type 2 (NRM2).
This type of NRM is mediated by protein and/or RNA driven
mechanism that translocate preexisting DNA elements, or vary nucleotides in genes in a similar fashion as observed for immunoglobulins.
NRM2 also plays a pivotal role in parasite-host interactions, and are likely to be abundant in other interactions between organisms where ‘evolutionary armsrace’ is ongoing.They have been demonstrated in the 1G5 gene in Drosophila, and in cone snail toxin genes.Implications: NMR2 may give alignment of mutations (‘shared mutations’) in related MPGs.Variation is limited and preexistent.
--------------------------------------------------------------------------------Peter: First, the last line in the above quoted section is, surely,
a starting assumption and not an implication of the NRM2.PB: Since variation required an initial source to vary upon, variation can only be limited. As long as the DNA elements are limited the variation is limited as well.Peter: Mechanisms for mediating copy errors are compatible with standard evolutionary theory too. They aid in preventing unviable organisms.PB: Elaborate please.Some questions for Peter Borger::1) Why do some MPG's translocate sequences and otehrs not when
there is a mechanism at work? Does the mechanism do different
things on the same inputs?PB: Why? How do I know? Probbly depending on sveral factors like DNA microenvironment, proteinmodification, extrenal factors? There are many things possible in biology. Expect the unexpected.2) Since we only have current genomes to examine, how can you tell
that some locii only ever bear certain nucleotides ... maybe ancient
versions of the same genome had different nucleotides in those
locations ... we cannot tell from modern genomes.PB: There are only 4 possibilites per position, some more likely than others. It depends on the codon position what happens.3) If you are marking four assigments and find that they each have
the same opening paragraph but different concluding paragraphs
do you assume they are independent or based upon the same
source?PB: depends on how the first paragraph came into being: random or non-random.Best wishes,
Peter

Peter: Decsribing this as 'non-random' is spurious at best. Randomness
is takne, generally to mean, unpredictable ... which is what is
seen. Sometimes the location appears to have changed, sometimes
not. There is no known causative link.PB: We've discussed this before.M: And you have failed in all cases to make a point. If there is non-random mutation (or deterministic mutation since this is what you are actually proposing) then you have to provide an experimental or natural example where you can a priori without error show where the next mutation will occur in a given DNA sequence. If you cannot do this (which you cannot and have not) then your argument is false. Your argument has been basically that you "knew all along" that the mutation would occur where it did after the fact which is not determinsitic and shows no predictive capacity on your part which is a requirment if it is a deterministic mechanism. But you do have recourse to experimentation.... so get to it hotshot

Peter: Decsribing this as 'non-random' is spurious at best. Randomness
is takne, generally to mean, unpredictable ... which is what is
seen. Sometimes the location appears to have changed, sometimes
not. There is no known causative link.
PB: We've discussed this before.M: And you have failed in all cases to make a point. If there is non-random mutation (or deterministic mutation since this is what you are actually proposing) then you have to provide an experimental or natural example where you can a priori without error show where the next mutation will occur in a given DNA sequence. If you cannot do this (which you cannot and have not) then your argument is false. Your argument has been basically that you "knew all along" that the mutation would occur where it did after the fact which is not determinsitic and shows no predictive capacity on your part which is a requirment if it is a deterministic mechanism. But you do have recourse to experimentation.... so get to it hotshotJ:
Is random the best word to describe the mutations described by Lynn Caporale and Lynn Ripley?
Statistically it appears that some events are "favoured". If they are strongly favoured then calling them random may be misleading in as much as random can mean no favouring at all.

The short answer is "yes", it is the best word. Remember that simply because there is a statistically higher probability of a given mutation occurring in a given spot, it is still random in the sense that there is no way to predict when or even if the mutation will occur. In a biological sense, it is random - a shorthand term used by biologists. I admit that the common usage of the term has a different connotation. The problem with "non-random" as PB uses it, for example, implies deterministic or directed mutations. The thing we've all been trying to get PB to show is that, if non-random applies, the implication is that you can unequivocally state when and where the mutation happens. This is not the case, and Dr. Caporale makes it very clear in her writing. The problem you appear to be articulating is the scientific definition of the term is very specific, whereas the common use of the term can mean something else. "Theory" is another example.Hope this helps.

Thanks Quetzal!quetzal:
The short answer is "yes", it is the best word. Remember that simply because there is a statistically higher probability of a given mutation occurring in a given spot, it is still random in the sense that there is no way to predict when or even if the mutation will occur. In a biological sense, it is random - a shorthand term used by biologists. I admit that the common usage of the term has a different connotation. The problem with "non-random" as PB uses it, for example, implies deterministic or directed mutations. The thing we've all been trying to get PB to show is that, if non-random applies, the implication is that you can unequivocally state when and where the mutation happens.Judge:
I'll let PB speak for himself on this, but for my own determinism seems irreconcilible with creationism (and i doubt that PB would argue for pure determinsim), and from my perspective there are probably important implications for science itself that stem from this.
If these mutations were in fact determined then what is to stop it all being a big "machine"? If these things are determined then we are faced with profound philosophical problems that have plagued "non theistic" systems of thought fro all recorded history.If this is determined then what becomes our our own freedom? Are our deepest thoughts and emotions determined? Is our own freedom swallowed up?
The tension between chance and fate or determinism and freedom can never be resolved.
And if I might be so bold and provocative, I believe the only way it can be dealt with is by viewing the cosmos as personally governed by a transcendent lawgiver. Only then can we have order and yet that order not be determined.[This message has been edited by judge, 03-24-2003][This message has been edited by judge, 03-24-2003]

Hi Judge,
Here is a way to think about it. If you are driving in fast moving heavy traffic in the rain, the probability that you will be involved in a car accident is higher than if you are driving on an empty road in good weather in a 30 MPH zone. However, if you have an accident in either case, it is still a random accident. Or in an area with earthquakes, every house has a higher probability of collapsing but which house collapses will be random.The higher mutation rates for specific DNA bases or certain sequences is affected by multiple chemical variables which can make mutation liklihood higher but it is still random. What Peter Borger is claiming as non-random is actually pre-determined mutations. If this were true, he would be able to unfailingly tell me where in the next cell division each non-random mutation will occur. He has never been able to do this for a single mutation in any conversation he has had with anyone on this subject.Hope this helps..if not let me know and I will try to clarify.cheers,
M

Hi mammuthus, good to make your aquaaintance.M:
What Peter Borger is claiming as non-random is actually pre-determined mutations. If this were true, he would be able to unfailingly tell me where in the next cell division each non-random mutation will occur. He has never been able to do this for a single mutation in any conversation he has had with anyone on this subject.Hope this helps..if not let me know and I will try to clarify.J:
I have tried to read what I could of Peter Borgers posts but did not get the idea that he was saying that any mutations were determined as such. perhaps PB could comment?
I did see that he perhaps suggested IIRC that the concept of Random mutations had been falsified, but this argument seemed to pivot on a certain definition of Random. If we take random to mean unpredictable then we must admit that the mutations are still unpredictable, though perhaps less at some times than others (and strictly speaking still random).
None the less it does seem to be very important in the creo/evo debateI went to the trouble of buying Dr caporales book and paased on a reference in there to some work by Lynn Ripley, Lynn Ripley "Predictability of Mutant Sequences" in Lynn Helena Caporale. ed, Molecular strategies in Biological Evolution, Annals of the New York Academy of Sciences Annals of the NY Academy of Science,870 (1999) 159-172 to an aquaintance who works with molecular phylogenies...here are his comments to me after ordering the paper 'subito'..Mrs. Ripley describes here specific patterns of natural mutations (i.e. mutations that were not induced by mutagens).
She found that mutations are not equally distributed among the genome (that was already known to me). But she also found that mutations are not equally distributed within the mutational "hot spots" (that was new to me). She found that several specific sequence motifs are more prone to naturally occuring mutations than others and she correlated them e.g. to weaknesses in the proof-reading abilities of DNA-polymerases or to certain secondary structures (like "quasi-palindromic sequences" or stem-loop structures) that are likely to increase the error rate of polymerases. She was even able to create new mutational hot spots by insertion of such motifs into E. coli plasmids.

​

In conclusion, it is possible that the exact same mutations arise independently in two organismal lineages. This means that we must carefully investigate the replication- and error-properties of DNA polymerases, but it does not refute common descent.

​

Thanks for this intersting snippet of information, judge! I appreciate it, escpecially because I wasn't aware of this. It is an area of research that has to be monitored, especially while working with molecular phylogenies (as I do).It should be obvious that on the surface this is of extreme interest to creationists with the obvious question. How do we work this revelation into molecular phylogenies.What do you think?[This message has been edited by judge, 03-24-2003]

It should be obvious that on the surface this is of extreme interest to creationists with the obvious question. How do we work this revelation into molecular phylogenies.What do you think?M: Hi judge, Glad you are interested in the topic. Actually, all of this has been known for quite some time. Almost all algorithms used in constructing phylogeny's take variable mutation rates into account (or are able to) and their are statistical tests for determining for example when your sequences are undergoing accelerated evolution within one species.What it ultimately means for the field of phylogenetics is that deriving phylogenies from a single locus can lead to errors. But again, this has been known for a long time.What Borger is suggesting (he won't comment until next week as he has been suspended for one week) is that there is no identity by descent and that any similarity between two organisms is due to pre-determined mutation though he allows for some static background mutation that is random. This is falisified by about 50 years of research on mutation. But more importantly, it is falsified by genetics. The reason genetic diseases run in families is because a mutation occurs and then is passed on to the offspring i.e. you are related to your parents by common descent. Borger would have you believe that a family in which muscular dystrophy runs is because each individual has a de novo mutation that appears to have been inherited by common descent...of course the implication of this is that we are not related to our parents.But I will grant you one problem that you observed. Given that their are four bases and a limited number of combinations one can have in a sequence, it is possible to make an error where for a given base position in a sequence, the sequences have arrived at the same base independently and not by common descent. This problem increases with the time of separation between the two species. The solution at this point has been to choose either highly conserved sequences that do not mutate much even over long periods of time (for comparing distantly related organisms) or using increasing amounts of sequence i.e. more data. Of course, other data is usually included in deriving phylogenies such as morphology. When the various data sources are in sync with one another, the confidence in the phylogeny increases.SLPx can probably explain phylogenetics much more comprehensively than I can but feel free to ask away if anything is not clear.Cheers,
M

Hi judge: PB is essentially arguing that 1) all adaptations are present in the genome from the "beginning", and 2) that mutations are predetermined in some way (i.e., non-random). In this sense, he's advocating determinism at the genetic level as a capability of an organism to perferentially mutate in response to its environment. I'm not quite sure how you got from non-random genetics to behavioral determinism. There's a difference between "determinism" as PB is using it in the context of non-random mutation, and behavioral determinism (i.e., genes drive behavior). In the latter case, I'd say you're probably right that it would have implications for the whole metaphysics of free will, etc.Interestingly enough, that has been a major argument for quite some time in biology. It's one of the reasons that sociobiology, from its quite solid foundation, has been relegated in a lot of places to "crank" science. However, most modern evo psych and evo bio thought is that, whereas genes provide some of the framework, it's environment and culture that determine behavior. I just started reading what is turning out to be a great book, Paul Ehrlich's "Human Natures". I highly recommend it as a response to the genetic determinist argument. Yeah, I knew you thought that.