Missing Heritability

[john...@gmail.com]

http://www.plosone.org/article/info:doi/10.1371/journal.pone.0029337

Statistical Epistasis and Functional Brain Imaging Support a Role of Voltage-Gated Potassium Channels in Human Memory

Abstract:-
Despite the current progress in high-throughput, dense genome scans, a major portion of complex traits' heritability still remains unexplained, a phenomenon commonly termed �missing heritability.�

JE:-
This missing heritability is rather obviosly, heritable epistasis.

Abstract:-
The negligence of analytical approaches accounting for gene-gene interaction effects, such as statistical epistasis, is probably central to this phenomenon.

JE:-
"Negligence" is just another way of saying, mathemtical model misuse.

Abstract:-
Here we performed a comprehensive two-way SNP interaction analysis of human episodic memory, which is a heritable complex trait, and focused on 120 genes known to show differential, memory-related expression patterns in rat hippocampus. Functional magnetic resonance imaging was also used to capture genotype-dependent differences in memory-related brain activity. A significant, episodic memory-related interaction between two markers located in potassium channel genes (KCNB2 and KCNH5) was observed (Pnominal combined = 0.000001). The epistatic interaction was robust, as it was significant in a screening (Pnominal = 0.0000012) and in a replication sample (Pnominal = 0.01). Finally, we found genotype-dependent activity differences in the parahippocampal gyrus (Pnominal = 0.001) supporting the behavioral genetics finding. Our results demonstrate the importance of analytical approaches that go beyond single marker statistics of complex traits.

JE:-
Indeed. The continued absolute deletion of e within Hamilton's Rule represents a more general and therefore more catastrophic, act of "negligence".

Regards,

John Edser
Independent Researcher

ed...@ozemail.com.au

[William L Hunt]

On Wed, 22 May 2013 11:00:31 -0400 (EDT), john...@gmail.com wrote:

>http://www.plosone.org/article/info:doi/10.1371/journal.pone.0029337
>
>Statistical Epistasis and Functional Brain Imaging Support a Role of Voltage-Gated Potassium Channels in Human Memory
>
>Abstract:-
>Despite the current progress in high-throughput, dense genome scans, a major portion of complex traits' heritability still remains unexplained, a phenomenon commonly termed “missing heritability.”
>
>JE:-
>This missing heritability is rather obviosly, heritable epistasis.
>

You have this upside-down. "Missing heritability" doesn't mean there
is some observed heritability that has not been predicted and needs an
explanation. So your explanation is not needed.
On the contrary, the predicted heritablity from the genetic variance
is greater than what is measured by examining loci. The simple
explanation is that the presumption that all the genetic variance is
additive is incorrect and that some is epistatic (statistical) and
therefore was never heritable per fisher.
The simple explanation is not heritable epistasis; it is
non-heritable epistasis!
William L Hunt

[john...@gmail.com]

http://www.plosone.org/article/info:doi/10.1371/journal.pone.0029337

Statistical Epistasis and Functional Brain Imaging Support a Role of Voltage-Gated Potassium Channels in Human Memory

Abstract:-
Despite the current progress in high-throughput, dense genome scans, a major portion of complex traits' heritability still remains unexplained, a phenomenon commonly termed “missing heritability.”

JE:-
This missing heritability is rather obviosly, heritable epistasis.

WH:-

You have this upside-down. "Missing heritability" doesn't mean there
is some observed heritability that has not been predicted and needs an
explanation.

JE:-
If the observed heritability is greater than what can be predicted OR the reverse i.e. the predicted heritability is greater than what can be observed, in either case, Fisher's mathematics stands empirically non verified. The main point: a falsification of Fisher is not possible simply because mathematics is not a science.

WH:-

So your explanation is not needed.

JE:-
If the predicted heritability is greater than what can be observed in nature, there is more than just the one possible explanation for this! Please suggest at least one other to verify your lack of prejudice.


Unfortunately, your singular, Post Modern, Neo Darwinian explanation that what was epistatic, simply wasn't, is nothing more than a convenient, ad hoc amendment, i.e. a retro fit up of Neo Darwinian theory to the empirical facts via continued mathematical misuse. Since mathematics isn't a science this does not matter to the mathematics but it does matter to any science based on Fisher, i.e. the gene centric Neo Darwinism that you endorse.


snip<

[William L Hunt]

On Wed, 29 May 2013 13:45:51 -0400 (EDT), john...@gmail.com wrote:

>
>http://www.plosone.org/article/info:doi/10.1371/journal.pone.0029337
>
>Statistical Epistasis and Functional Brain Imaging Support a Role of Voltage-Gated Potassium Channels in Human Memory
>
>Abstract:-
>Despite the current progress in high-throughput, dense genome scans, a major portion of complex traits' heritability still remains unexplained, a phenomenon commonly termed “missing heritability.”
>
>JE:-
>This missing heritability is rather obviosly, heritable epistasis.
>
>WH:-
>You have this upside-down. "Missing heritability" doesn't mean there
>is some observed heritability that has not been predicted and needs an
>explanation.
>
>JE:-
>If the observed heritability is greater than what can be predicted OR the reverse i.e. the predicted heritability is greater than what can be observed, in either case, Fisher's mathematics stands empirically non verified. The main point: a falsification of Fisher is not possible simply because mathematics is not a science.
>
>WH:-
>
>So your explanation is not needed.
>
>JE:-
>If the predicted heritability is greater than what can be observed in nature, there is more than just the one possible explanation for this! Please suggest at least one other to verify your lack of prejudice.
>

Another explanation is that the total of the effects of individual
loci is incomplete. The argument is that there are still loci that
have been missed that effect the trait. This may be true but nobody
seems to be able to find them.
But, to repeat, the simple explanation is that the presumption that

all the genetic variance is additive is incorrect and that some is
epistatic (statistical) and therefore was never heritable per fisher.

William L Hunt

[john...@gmail.com]

http://www.plosone.org/article/info:doi/10.1371/journal.pone.0029337

Statistical Epistasis and Functional Brain Imaging Support a Role of Voltage-Gated Potassium Channels in Human Memory

Abstract:-
Despite the current progress in high-throughput, dense genome scans, a major portion of complex traits' heritability still remains unexplained, a phenomenon commonly termed “missing heritability.”

JE:-
This missing heritability is rather obviosly, heritable epistasis.

WH:-

You have this upside-down. "Missing heritability" doesn't mean there
is some observed heritability that has not been predicted and needs an
explanation.

JE:-
If the observed heritability is greater than what can be predicted OR the reverse i.e. the predicted heritability is greater than what can be observed, in either case, Fisher's mathematics stands empirically non verified. The main point: a falsification of Fisher is not possible simply because mathematics is not a science.

WH:-

So your explanation is not needed.

JE:-
If the predicted heritability is greater than what can be observed in nature, there is more than just the one possible explanation for this! Please suggest at least one other to verify your lack of prejudice.

WH:-

Another explanation is that the total of the effects of individual
loci is incomplete. The argument is that there are still loci that
have been missed that effect the trait. This may be true but nobody
seems to be able to find them.

JE:-
Your reply reminds me of a question asked of a Christian: Why can we observe some Christians doing evil things? "That is easy" was the sincere reply, "if a Christian does evil it proves that person was never a Christian". Maybe you prefer the more quaint ad hoc statement that a puddle is always perfectly formed to hold the water it contains? The fun you can have with non falsifiable ad hoc is endless...

Using your Neo Darwinistic Post Modern epistemology, to be able find heritable epistasis becomes an impossibility since whenever a heritable e is found it is easily redefined as additive and therefore, was-never-e-in-the-first-place, thereby maintaining Fisher's dogma for all time that only non epistatic gene associations are heritable. Whenever you resort to ad hoc mathematics as a misused theory of science you insure that you only see what you want to see.

WH:-
But, to repeat, the simple explanation is that the presumption that

all the genetic variance is additive is incorrect and that some is
epistatic (statistical) and therefore was never heritable per fisher.

JE:-
The above is not an explanation in any scientific sense for the simple reasons provided. Fisher's statistics are not a theory in their own right and are misused when employed as such. Under no circumstances can e be removed from Hamilton's Rule, rendering the rule empirically inoperable.

Note: if ALL variance is heritable evolution by natural becomes impossible. This is why deeming e to be partly but not fully heritable makes scientific sense. Defining e as absolutely non heritable as Neo Darwinism continues to do in 2013 is just as nonsensical as defining e absolutely heritable.

Please provide at least two scientifically valid explanations for the observed phenomena and explain what test of nature allows anybody to falsify one in favour of the other.

[William L Hunt]

WLH:- Fisher's model for inheritance of complex traits (numerous
factors, each with very small effect) is used by animal and plant
breeders because it gives predictions better than any other model. If
some other model gave better predictions, then it would be used. But
Fisher's model only makes predictions, it doesn't give a cause why a
trait is inherited the way it is observed to be. Outside of breeders,
most people are left unsatisfied with just a prediction of how the
trait will vary in offspring, they want a causal explanation of why
the trait is different. For that Fisher's statistics are useless.

> Under no circumstances can e be removed from Hamilton's Rule, rendering the rule empirically inoperable.
>
>Note: if ALL variance is heritable evolution by natural becomes impossible.

WLH:- You have this backward. The response to selection increases as
heritability increases. If there is NO heritability, then selection

[john...@gmail.com]

JE:-

Using your Neo Darwinistic Post Modern epistemology, to be able find heritable epistasis becomes an impossibility since whenever a heritable e is found it is easily redefined as additive and therefore, was-never-e-in-the-first-place, thereby maintaining Fisher's dogma for all time that only non epistatic gene associations are heritable. Whenever you resort to ad hoc mathematics as a misused theory of science you insure that you only see what you want to see.

WH:-

But, to repeat, the simple explanation is that the presumption that

all the genetic variance is additive is incorrect and that some is
epistatic (statistical) and therefore was never heritable per fisher.

JE:-
The above is not an explanation in any scientific sense for the simple reasons provided. Fisher's statistics are not a theory in their own right and are misused when employed as such.

WLH:- Fisher's model for inheritance of complex traits (numerous
factors, each with very small effect) is used by animal and plant
breeders because it gives predictions better than any other model. If
some other model gave better predictions, then it would be used. But
Fisher's model only makes predictions, it doesn't give a cause why a
trait is inherited the way it is observed to be.

JE:-
Models are NOT theories so they are either used or misused. Theories cannot be scientifically misused since they simply refute when they are misused.

The practical employment of Fisher's model to predict useful for breeder outcomes represents a correct use of this model. However, the employment of the same model to absolutely delete e from Hamilton's Rule (as you so graphically indicated) represents a gross misuse since basic evolutionary theory causation is now at stake via the invalid employment of Fisher's model.

I have attempted to open free and fair discussion on model use and misuse for over a decade. Neither yourself or other professionals who at some point have posted here will enter into any such discussion. In the past, Prof Joe Felsenstein, Dr Larry Moran and Dr Guy Hoelzler refused discussion. Even Dr John Wilkins who maintains an interest in epistemology would not engage this topic. I can only conclude that the pressing question of model misuse within evolutionary theory and other currently critical sciences such as climate change science, was in the past and continues today, evaded. Indeed, most who post here will not even say if they separate a theory from a modeI! I still have no idea of your position with regards to any of these basic questions.

Outside of breeders,
most people are left unsatisfied with just a prediction of how the
trait will vary in offspring, they want a causal explanation of why
the trait is different. For that Fisher's statistics are useless.

JE:-
Then you agree that employing Fisher to absolutely delete e from Hamilton's rule represents Neo Darwinistic model misuse? Please answer definitively.

I repeat: under no circumstances can e be removed from Hamilton's Rule via employing Fisher, rendering the rule empirically inoperable.

JE:-

Note: if ALL variance is heritable evolution by natural becomes impossible.


WLH:- You have this backward. The response to selection increases as
heritability increases. If there is NO heritability, then selection
becomes impossible.

JE:-
This is the second time you have incorrectly stated I had put something backward. You were incorrect then for the same reason your are incorrect now. I this particular case BOTH situations are prohibited via any bona fide falsifiable theory. IOW, when nothing is proposed heritable and when everything is proposed heritable, evolution by natural selection ceases. If every tiny bit of variation happened to be 100% passed on, the inevitable pilling up of heritable clutter will bring evolution to a halt. This represents the most potent argument against the Lamarkian inheritance of acquired characters.

I repeat:- this is why deeming e to be partly but not fully heritable makes scientific sense. Defining e as absolutely non heritable as Neo Darwinism continues to do in 2013 is just as nonsensical as defining e absolutely heritable.

JE:-
Also I repeat:-

Please provide at least two scientifically valid explanations for the observed phenomena and explain what test of nature allows anybody to falsify one in favour of the other.

[William L Hunt]

On Sat, 8 Jun 2013 13:25:35 -0400 (EDT), john...@gmail.com wrote:

>
>
>JE:-
>Using your Neo Darwinistic Post Modern epistemology, to be able find heritable epistasis becomes an impossibility since whenever a heritable e is found it is easily redefined as additive and therefore, was-never-e-in-the-first-place, thereby maintaining Fisher's dogma for all time that only non epistatic gene associations are heritable. Whenever you resort to ad hoc mathematics as a misused theory of science you insure that you only see what you want to see.
>
>WH:-
>But, to repeat, the simple explanation is that the presumption that
>all the genetic variance is additive is incorrect and that some is
>epistatic (statistical) and therefore was never heritable per fisher.
>
>JE:-
>The above is not an explanation in any scientific sense for the simple reasons provided. Fisher's statistics are not a theory in their own right and are misused when employed as such.
>
>WLH:- Fisher's model for inheritance of complex traits (numerous
>factors, each with very small effect) is used by animal and plant
>breeders because it gives predictions better than any other model. If
>some other model gave better predictions, then it would be used. But
>Fisher's model only makes predictions, it doesn't give a cause why a
>trait is inherited the way it is observed to be.
>
>JE:-
>Models are NOT theories so they are either used or misused. Theories cannot be scientifically misused since they simply refute when they are misused.
>
>The practical employment of Fisher's model to predict useful for breeder outcomes represents a correct use of this model. However, the employment of the same model to absolutely delete e from Hamilton's Rule (as you so graphically indicated) represents a gross misuse since basic evolutionary theory causation is now at stake via the invalid employment of Fisher's model.
>
>I have attempted to open free and fair discussion on model use and misuse for over a decade. Neither yourself or other professionals who at some point have posted here will enter into any such discussion. In the past, Prof Joe Felsenstein, Dr Larry Moran and Dr Guy Hoelzler refused discussion. Even Dr John Wilkins who maintains an interest in epistemology would not engage this topic. I can only conclude that the pressing question of model misuse within evolutionary theory and other currently critical sciences such as climate change science, was in the past and continues today, evaded. Indeed, most who post here will not even say if they separate a theory from a modeI! I still have no idea of your position with regards to any of these basic questions.

I am more concerned with getting a clear and understandable exposition of the concept rather than
whether it is labled a model or theory. Usually it will be clear enough from the context whether is
should be called one or the other but somethimes it is not so clear. I think of when Copernicus
first presented his theory. Out of fear of the church, in his forward he offered it only as a model
to make predictions rather than as a true representation of reality. For some time seafaring
mariners, who may have felt Copernicus's model was a true represention of reality, continued to use
Ptolemy for their calculations because Ptolemy's theory/model gave more accurate predictions.

>
>
>Outside of breeders,
>most people are left unsatisfied with just a prediction of how the
>trait will vary in offspring, they want a causal explanation of why
>the trait is different. For that Fisher's statistics are useless.
>
>JE:-
>Then you agree that employing Fisher to absolutely delete e from Hamilton's rule represents Neo Darwinistic model misuse? Please answer definitively.

This must be a rhetorical question as I have answered NO I do not agree to you at least a dozen
times in previous threads.

>
>I repeat: under no circumstances can e be removed from Hamilton's Rule via employing Fisher, rendering the rule empirically inoperable.
>
>JE:-
>Note: if ALL variance is heritable evolution by natural becomes impossible.
>
>
>WLH:- You have this backward. The response to selection increases as
>heritability increases. If there is NO heritability, then selection
>becomes impossible.
>
>JE:-
>This is the second time you have incorrectly stated I had put something backward. You were incorrect then for the same reason your are incorrect now. I this particular case BOTH situations are prohibited via any bona fide falsifiable theory. IOW, when nothing is proposed heritable and when everything is proposed heritable, evolution by natural selection ceases. If every tiny bit of variation happened to be 100% passed on, the inevitable pilling up of heritable clutter will bring evolution to a halt.

None of this makes sense to me, so I don't know how to respond.

> This represents the most potent argument against the Lamarkian inheritance of acquired characters.

Well there have been actual experiments that demonstrate no Lamarkian inheritance for those who
prefer experimental evidence over argument..

>
> I repeat:- this is why deeming e to be partly but not fully heritable makes scientific sense. Defining e as absolutely non heritable as Neo Darwinism continues to do in 2013 is just as nonsensical as defining e absolutely heritable.
>
>JE:-
>Also I repeat:-
>Please provide at least two scientifically valid explanations for the observed phenomena and explain what test of nature allows anybody to falsify one in favour of the other.

It is not clear but I don't no what the "observed phenomena" can be other than the two-way SNP
interaction and the observed statistical epistasis (with no discussion of heritability) in the
initially mentioned paper. That seems unremarkable so I don't why you would need a "scientifically
valid explanation". Much less two explanations.
William L Hunt

[john...@gmail.com]

JE:-
Models are NOT theories so they are either used or misused. Theories cannot be scientifically misused since they simply refute when they are misused.

The practical employment of Fisher's model to predict useful for breeder outcomes represents a correct use of this model. However, the employment of the same model to absolutely delete e from Hamilton's Rule (as you so graphically indicated) represents a gross misuse since basic evolutionary theory causation is now at stake via the invalid employment of Fisher's model.

I have attempted to open free and fair discussion on model use and misuse for over a decade. Neither yourself or other professionals who at some point have posted here will enter into any such discussion. In the past, Prof Joe Felsenstein, Dr Larry Moran and Dr Guy Hoelzler refused discussion. Even Dr John Wilkins who maintains an interest in epistemology would not engage this topic. I can only conclude that the pressing question of model misuse within evolutionary theory and other currently critical sciences such as climate change science, was in the past and continues today, evaded. Indeed, most who post here will not even say if they separate a theory from a modeI! I still have no idea of your position with regards to any of these basic questions.

WLH:-

I am more concerned with getting a clear and understandable exposition of the concept rather than whether it is labled a model or theory.

JE:-
That is exactly what labelling them achieves! Unless they always remain separate within the sciences theory causality may become reversed within just-a-model-of-a-theory making an unclear and non understandable theory "exposition". The best modern example I know of is the gene centric reversal of causality within Dawkins "selfish gene" model which as you know was and remains 100% predicated on Inclusive Fitness. In this misused-as-a-theory model of bona fide Darwinian theory, at least one Darwinian variable and even more critically, a central Darwinian constant have been artificially removed. The absolutely deleted from the theory variable is e and the more critical but similarly mistreated constant is D, representing Total Darwinian Fitness (TDF): the total number of strictly adult fertile forms reproduced per parent per population. Neither have any representation within Inclusive Fitness. In Hamilton's Rule e must be included, even within just a simplified model that relatively removes e via artificially fixing e=1. This model must minimally allow e =2 within any sexual species with just a single pair of chromosomes. Much more importantly, D sets a missing-in-the-rule limit to c since in Hamilton's model, c remains UNLIMITED. Without D included, Darwinian cause and effect is easily reversed because the rule has no falsifiable constant to act as a frame of reference. Variables must be measured relative this frame NOT just relative to each other as they have been incorrectly allowed to do within the rule.

WLH:-

Usually it will be clear enough from the context whether is
should be called one or the other but somethimes it is not so clear. I think of when Copernicus
first presented his theory. Out of fear of the church, in his forward he offered it only as a model
to make predictions rather than as a true representation of reality. For some time seafaring
mariners, who may have felt Copernicus's model was a true represention of reality, continued to use Ptolemy for their calculations because Ptolemy's theory/model gave more accurate predictions.

JE:-
Models are clearly and easily separated from theories since only theories can empirically refute and always do so when scientifically misused. In stark contrast, models can only verify/non verify. The proof that Fisher presented a model of a theory, i.e. not a valid theory in its own right, was the simple fact that the use of Fisher to absolutely delete e via a zero heritability Neo Darwinistic ASSUMPTION cannot be empirically falsified only verified/non verified. Without exception, mathematical tautologies cannot be falsified. All of them are employed in the sciences in just the ad hoc way that you and Neo Darwinian establishment employed Fisher. This may constitute either a valid use or unhappily, an invalid misuse via the model replacing the theory it was simplified/over simplified from allowing a reversal of theory cause and effect. In Fisher's case this constituted a gross misuse since some e is empirically heritable. If this wasn't the case Morgan could not have mapped fruit fly chromosomes.


WLH:-

Outside of breeders, most people are left unsatisfied with just a prediction of how the
trait will vary in offspring, they want a causal explanation of why
the trait is different. For that Fisher's statistics are useless.

JE:-
Then you agree that employing Fisher to absolutely delete e from Hamilton's rule represents Neo Darwinistic model misuse? Please answer definitively.

WLH:-

This must be a rhetorical question as I have answered NO I do not agree to you at least a dozen
times in previous threads.

JE:-
The question is not at all rhetorical just a futile attempt to keep your exposition self consistent. You have stated that Fisher is "useless" with regards to understanding "cause and effect". OK. You do not seem to see that this contradicts your use of Fisher to absolutely (not just relatively) delete e from Hamilton 's Rule only allowing an additive and therefore fictitious, competing in FITNESS independent gene level of selection that can now in principle, challenge and win against Darwin's single D fitness maximand, i.e. REVERSE Darwinian THEORY cause and effect via allowing the evolution of parental fitness altruism, but only when employing a "selfish gene" MODEL.

I ask again: given your clear statement concerning the limitations of Fisher's model do you now agree that employing Fisher to absolutely delete e from Hamilton's rule represents Neo Darwinistic model misuse?

I repeat: under no circumstances can e be removed from Hamilton's Rule via employing Fisher, rendering the rule empirically inoperable.

JE:-
Note: if ALL variance is heritable evolution by natural becomes impossible.


WLH:- You have this backward. The response to selection increases as
heritability increases. If there is NO heritability, then selection
becomes impossible.

JE:-
This is the second time you have incorrectly stated I had put something backward. You were incorrect then for the same reason your are incorrect now. I this particular case BOTH situations are prohibited via any bona fide falsifiable theory. IOW, when nothing is proposed heritable and when everything is proposed heritable, evolution by natural selection ceases. If every tiny bit of variation happened to be 100% passed on, the inevitable pilling up of heritable clutter will bring evolution to a halt.

WLH:-

None of this makes sense to me, so I don't know how to respond.

JE:-
I wrote the above in simple to understand English. Do you agree or disagree that in principle, if NOTHING or if EVERYTHING is heritable, evolution by natural selection is halted?

This represents the most potent argument against the Lamarkian inheritance of acquired characters.

WLH:-

Well there have been actual experiments that demonstrate no Lamarkian inheritance for those who prefer experimental evidence over argument..

JE:-
This depends entirely on how you define Lamarckian Inheritance! Please supply a definition.

I repeat:- this is why deeming e to be partly but not fully heritable makes scientific sense. Defining e as absolutely non heritable as Neo Darwinism continues to do in 2013 is just as nonsensical as defining e absolutely heritable.

Also I repeat:-
Please provide at least two scientifically valid explanations for the observed phenomena and explain what test of nature allows anybody to falsify one in favour of the other.

WLH:-

It is not clear but I don't no what the "observed phenomena" can be other than the two-way SNP
interaction and the observed statistical epistasis (with no discussion of heritability) in the
initially mentioned paper. That seems unremarkable so I don't why you would need a "scientifically
valid explanation". Much less two explanations.

JE:-
I will ask this another way: what observation of nature would prove to you that e is partly heritable?

[William L Hunt]

On Wed, 12 Jun 2013 17:36:17 -0400 (EDT), john...@gmail.com wrote:

>
....
[snip]
....

>
>Also I repeat:-
>Please provide at least two scientifically valid explanations for the observed phenomena and explain what test of nature allows anybody to falsify one in favour of the other.
>
>WLH:-
>It is not clear but I don't no what the "observed phenomena" can be other than the two-way SNP
>interaction and the observed statistical epistasis (with no discussion of heritability) in the
>initially mentioned paper. That seems unremarkable so I don't why you would need a "scientifically
>valid explanation". Much less two explanations.
>
>JE:-
>I will ask this another way: what observation of nature would prove to you that e is partly heritable?

When the loci involved are linked, epistasis is partly heritable.
William L Hunt

[john...@gmail.com]

Sent from my iPad

On 18/06/2013, at 6:57 AM, "sci.bio.evolution moderation account" <s...@darwin.ediacara.org> wrote:

Newsgroups: sci.bio.evolution
Approved: jo...@darwin.ediacara.org
From: wlh...@earthlink.net (William L Hunt)
Subject: Re: Missing Heritability
References: <knimif$1p3d$1...@darwin.ediacara.org> <ko39o0$5e0$1...@darwin.ediacara.org> <kpapkh$1rln$1...@darwin.ediacara.org>
Organization:

On Wed, 12 Jun 2013 17:36:17 -0400 (EDT), john...@gmail.com wrote:

....
[snip]
....

Also I repeat:-
Please provide at least two scientifically valid explanations for the observed phenomena and explain what test of nature allows anybody to falsify one in favour of the other.

WLH:-
It is not clear but I don't no what the "observed phenomena" can be other than the two-way SNP
interaction and the observed statistical epistasis (with no discussion of heritability) in the
initially mentioned paper. That seems unremarkable so I don't why you would need a "scientifically
valid explanation". Much less two explanations.

JE:-
I will ask this another way: what observation of nature would prove to you that e is partly heritable?

WLH:-

When the loci involved are linked, epistasis is partly heritable.

JE:-
Linkage depends entirely on how far apart they happen to be on the same chromosome (statistical epistasis). This variable allowed Morgan to accurately map chromosomes. IOW, e CAN NEVER be ABSOLUTELY deleted from Hamilton's Rule as has been allowed since inception!

I noticed you snipped all discussion concerning Neo Darwinian model misuse. These discussions are centred on this subject since evasion is never the answer. You agreed that Fishers's mode can have nothing valid to say about cause and effect. OK. However, you seem to fail to understand/more likely choose to ignore, that the employment of Fisher's heritability model reverses cause and effect within Darwinism for the reasons previously given. Instead of altruistic gene replication servicing a single adult Darwinian organism maximal fitness (TDF), with the help of Fisher only allowing additive and therefore fitness independent gene associations as heritable, the previously prohibited reverse now also applies: "selfish genes" cause altruistic in fitness adult forms. Since BOTH apply within Neo Darwinism it becomes impossible to falsify evolutionary theory since tautologies cannot be falsified.



Consistent to what you have argued, Fisher's model cannot validly exclude ALL e from Hamilton's Rule since whenever doing so Darwinian cause and effect becomes reversible. To remain consistent to what you previously stated concerning cause and effect limitations of models, Hamilton's Rule must be : (r^e)b >c where e has been artificially fixed to e =1 as a modelling simplification. What does this mean? It proves that Inclusive Fitness was never a theory in its own right it, just a simplified model of a theory within which e=2 within a minimally valid model so Inclusive Fitness cannot challenge and win against Darwinian theory as it has been allowed to do for over a century. Hamilton was minimally 100% in error. In Haldane's example four, not just the two brothers related 0.5 are required to produce a minimally valid model.

[William L Hunt]

On Sat, 22 Jun 2013 00:31:11 -0400 (EDT), john...@gmail.com wrote:

>

.....
[snip]
.....

>
>I noticed you snipped all discussion concerning Neo Darwinian model misuse. These discussions are centred on this subject since evasion is never the answer.

We can agree that models can be oversimplified though we would disagree on particular examples.
We can agree that models can be misused though we would disagree on particular examples.
I would even agree that some models involving Hamiliton's Rule and misused but you seem to think
any use of Hamiliton's Rule in a model is a misuse. You seem to think that all such models are being
misused even before examining them.
You accept that Hamiliton' Rule is mathematically true (though not in nature) but offer a
modification that would make it no longer mathematically true but now somehow true in nature.
This makes no logical sense to me and appears to me to be just nonsense.
William L Hunt

[john...@gmail.com]

snip]
.....


I noticed you snipped all discussion concerning Neo Darwinian model misuse. These discussions are centred on this subject since evasion is never the answer.

WLF:-

We can agree that models can be oversimplified though we would disagree on particular examples.

JE:-
Please provide examples....

WLF:-

We can agree that models can be misused though we would disagree on particular examples.

JE:-
This may or may not be the case! Without examples provided by you, the above is only pointless rhetoric.

WLF:-
I would even agree that some models involving Hamiliton's Rule and misused..

JE:-
Please provide your example of a misuse of Hamilton's Rule.

WLH :-

but you seem to think any use of Hamiliton's Rule in a model is a misuse.

JE:-
Absolutely not. In over a decade posting here I have consistently claimed that ANY model, simplified or oversimplified, can be used or misused.


WLH:-

You seem to think that all such models are being
misused even before examining them.

JE:-
Ironically, this in reverse applies to Dawkins' popular gene centricity as it remains based on Hamilton: they seem to think that all such models are used correctly even before examining proposed cause and effect within the model, as it compares to cause and effect in the parent theory! In both Hamilton and Fisher, theory cause and effect has been allowed to 100% reverse in just simplified/oversimplified models of Darwinism.

WLH:-

You accept that Hamiliton' Rule is mathematically true (though not in nature) but offer a

modification that would make it no longer mathematically true ..

JE:-
Without providing an example you are indulging yourself in the worst form of rhetoric.

Below are detailed examples to which you failed to provide any specific criticism:

You agreed that Fishers's mode can have nothing valid to say about cause and effect. OK. However, you seem to fail to understand/more likely choose to ignore, that the employment of Fisher's heritability model reverses cause and effect within Darwinism for the reasons previously given. Instead of altruistic gene replication servicing a single adult Darwinian organism maximal fitness (TDF), with the help of Fisher only allowing additive and therefore fitness independent gene associations as heritable, the previously prohibited reverse now also applies: "selfish genes" cause altruistic in fitness adult forms. Since BOTH apply within Neo Darwinism it becomes impossible to falsify evolutionary theory since tautologies cannot be falsified.



Consistent to what you have argued, Fisher's model cannot validly exclude ALL e from Hamilton's Rule since whenever doing so Darwinian cause and effect becomes reversible. To remain consistent to what you previously stated concerning cause and effect limitations of models, Hamilton's Rule must be : (r^e)b >c where e has been artificially fixed to e =1 as a modelling simplification. What does this mean? It proves that Inclusive Fitness was never a theory in its own right it, just a simplified model of a theory within which e=2 within a minimally valid model so Inclusive Fitness cannot challenge and win against Darwinian theory as it has been allowed to do for over a century. Hamilton was minimally 100% in error. In Haldane's example four, not just the two brothers related 0.5 are required to produce a minimally valid model.

[William L Hunt]

On Thu, 27 Jun 2013 02:07:13 -0400 (EDT), john...@gmail.com wrote:

>snip]
>.....
>
>
>I noticed you snipped all discussion concerning Neo Darwinian model misuse. These discussions are centred on this subject since evasion is never the answer.
>
>WLF:-
>We can agree that models can be oversimplified though we would disagree on particular examples.
>
>JE:-
>Please provide examples....

We disagree on so much surely there can be no question that we would not always agree on whether
a particular model is oversimplified or not. But here is an example of one I consider oversimplified
(the compacted URL): http://tinyurl.com/ntcnyqu

They use a static model that does not include males. I feel that to be sufficiently realistic a
dynamic model that includes males should be used. The resulting equation is then more complex so
some might still the more simplified static model.

>
>WLF:-
>We can agree that models can be misused though we would disagree on particular examples.
>
>JE:-
>This may or may not be the case! Without examples provided by you, the above is only pointless rhetoric.

Yes it may be pointless rhetoric but as we disagree on so much I am sure we would not agree on all
examples of model misuse.

>
>WLF:-
>I would even agree that some models involving Hamiliton's Rule and misused..
>
>JE:-
>Please provide your example of a misuse of Hamilton's Rule.

Explaining praire dog warning behavior by invoking Hamilton's Rule is a misuse since there is no
demonstrated risk.
Also explaining hymenoptera workers by arguing since sisters are more related to sisters than their
mother or daughters they will prefer to make more sisters. Fortunately you are unlikely to see this
argument anymore. Even E. O. Wilson has disavowed it.

>
>WLH :-
>but you seem to think any use of Hamiliton's Rule in a model is a misuse.
>
>JE:-
>Absolutely not. In over a decade posting here I have consistently claimed that ANY model, simplified or oversimplified, can be used or misused.
>

OK, but then please provide an example of the use of a model using Hamilton' Rule that you
consider proper and not a misuse.

>
>WLH:-
>You seem to think that all such models are being
>misused even before examining them.
>
>JE:-
>Ironically, this in reverse applies to Dawkins' popular gene centricity as it remains based on Hamilton: they seem to think that all such models are used correctly even before examining proposed cause and effect within the model, as it compares to cause and effect in the parent theory! In both Hamilton and Fisher, theory cause and effect has been allowed to 100% reverse in just simplified/oversimplified models of Darwinism.
>
>WLH:-
>You accept that Hamiliton' Rule is mathematically true (though not in nature) but offer a
>modification that would make it no longer mathematically true ..
>
>JE:-
>Without providing an example you are indulging yourself in the worst form of rhetoric.
>
>Below are detailed examples to which you failed to provide any specific criticism:

I really can't comment because I can't make sense of these below examples.
William L Hunt

[john...@gmail.com]

I noticed you snipped all discussion concerning Neo Darwinian model misuse. These discussions are centred on this subject since evasion is never the answer.

WLF:-
We can agree that models can be oversimplified though we would disagree on particular examples.

JE:-
Please provide examples....

WLF:-

We disagree on so much surely there can be no question that we would not always agree on whether
a particular model is oversimplified or not.

JE:-
I still have no idea what the Neo Darwinist establishment or yourself consider to be an "oversimplified" model or how this differs from a "simplified" model or, amazingly, how either differ from a bona fide theory. In over a decade of posting questions not a single professional will define these basic terms.

My definions (repeated for over a decade):-

1. Oversimplified model: any model of theory in which at least one algebraic constant has been artificially altered or removed. If this leaves a model without a single algebraic constant then theory cause and effect becomes 100% reversible reducing such a model to just a cause and effect tautology (circular logic with circular causation).

Example: Hamilton's Rule rb>c.

Since absolutely no algebraic constant has been provided, The Rule as it stands only represents a tautology. Without exception, tautologies employed as bona fide theories of science are misused since all theories concern themselves with falsifiable propositions of causation. Reversible propositions as they are demonstrated by the rule cannot be falsified so they remain true by definition only (all definitions are tautologies). The most famous tautological model in the history of evolution is Herbert Spencer's "survival of the fittest". Even W. D. Hamilton employed Spencer's circular jingle to describe what he thought was Darwinism. Ask just about anybody on the street in 2013 what they think evolution is and they will chant Spencer's nonsense providing a damning testament for the education system.

The proper use of an oversimplified model provides incorrect causative views as useful illustrative contrasts, making it easier to view theory causation. Also, an oversimplified model may be gainfully employed to more easily calculate a result that does not significantly differ to the theory result. An example: it is valid to employ Newtonian Mechanics as jan oversimplified model of Special Relativity for most human based velocities. However, it is the height of stupidity to subsequently claim that Newton's causation is correct when doing so. This why Popper stressed that verification is not definitive.

2. Simplified Model: any model of a theory within which a defined theory variable has been artificially altered or removed.

Example: the 100% deletion of genetic epistasis from Hamilton's Rule. Since statistical epistasis can only lower the probability of two alleles on two different loci being inherited together, where this is a simple function of their distance apart on the same chromosome, e has to be represented in any valid Inclusive Fitness model, even when the model reduces e to zero. Hamilton's model is (r^e)b>c where e is fixed to 1 as an explicit modelling simplification. It is critical to include e even when e=1 because only this explicitly acknowledges that The Rule represents a simplified model of Darwinian theory, NOT A COMPETITIVE THEORY. The subsequent ongoing misuse of the rule via Neo Darwinistic gene centricity wrongly employing Inclusive Fitness as a theory not just a model, is demonstratedvby the fact that e was always only artificially fixed to 1 as Hamilton's modelling simplification of Darwinism, where Darwin's single adult organism level of selection included e as partly heritable.

WLF:-

But here is an example of one I consider oversimplified
(the compacted URL): http://tinyurl.com/ntcnyqu

They use a static model that does not include males. I feel that to be sufficiently realistic a dynamic model that includes males should be used. The resulting equation is then more complex so some might still the more simplified static model.

JE:-
I only have one question:-

What was their proposed fitness constant?

WLF:-
We can agree that models can be misused though we would disagree on particular examples.

JE:-
This may or may not be the case! Without examples provided by you, the above is only pointless rhetoric.

WLF :-

Yes it may be pointless rhetoric but as we disagree on so much I am sure we would not agree on all
examples of model misuse.

JE:-
.....more rhetoric...

Model misuse:- any model employed to replace a parental theory e.g. gene centric Inclusive Fitness replacing organism centric Darwinism.

WLF:-
I would even agree that some models involving Hamiliton's Rule and misused..

JE:-
Please provide your example of a misuse of Hamilton's Rule.

WLF:-

Explaining praire dog warning behavior by invoking Hamilton's Rule is a misuse since there is no
demonstrated risk.

JE:-
The risk may be small but it is NOT zero. The continually missed point here is that the cost < gain for each parent ruling out organism fitness altruism which is confused with cooperative, unequal fitness mutualism because no constant appears in The Rule.

WLF:-

Also explaining hymenoptera workers by arguing since sisters are more related to sisters than their
mother or daughters they will prefer to make more sisters. Fortunately you are unlikely to see this
argument anymore. Even E. O. Wilson has disavowed it.

JE:-
Inclusive Fitness relatedness is misused in an ongoing way. Relatedness IBD (Identical By Decent, i.e. the probability that a gene has been replicated from another gene, wrongly employs the gene as a frame of reference, only allowing brothers to share 50% of their IBD genes. Most species share over 99.9% "the same genes" but strictly, NON IBD, i.e. these genes have the same DNA sequence but are not necessarily derived from the same gene as the parent. Gene-to-gene decendancy is irrelevant to organism centricity since the Darwinian fertile organism frame of reference only concerns itself with the impact a gene has on TDF (Total Darwinian Fitness), i.e. has no interest in what gene a gene in question was replicated from.

WLH :-
but you seem to think any use of Hamiliton's Rule in a model is a misuse.

JE:-
Absolutely not. In over a decade posting here I have consistently claimed that ANY model, simplified or oversimplified, can be used or misused.

WLF :-

OK, but then please provide an example of the use of a model using Hamilton' Rule that you
consider proper and not a misuse.

JE:-
A modelling genes-eye-view may be useful in helping researchers to understand organism centricity since it provides a list of false causations, i.e. what a more complex organism centricity, isn't. As an illustration, a logically identical rule to Hamilton's can be formulated for observed rising sea levels. The sea will indeed flood over the land whenever w>l where w and l are respectively, the elevation of the land l, relative to the sea level w. However, it is always incorrect to claim this is caused by an increasing sea level since the reverse may always be true; a sinking land level. It was and remains impossible to separate these contradictory causations unless both w and l are measured relative to an included constant, NOT justvrelative to each other as are the variables in Hamilton's Rule. A necessary constant is obtained by measuring w and l's distance to the moon M. The cause of flooding can now be proven to be caused by rising seas when Mw < Ml (Mw= distance of the sea level to the moon, Ml= the distance of the land level to the moon). For this reason alone, Hamilton's Rule cannot validly claim "selfish geneism" causation because the reverse equally but oppositely applies: selfish organisms cause Hamilton's gene to spread via a donation acting as a mutual but not necessarily equal investment gain (the donation was never a gift). Like the land levels falling and the sea levels NOT rising causing the land to flood, unequal fitness mutualism remains an ignored possibility since a fitness constant has never been included in the rule. By simple deduction, this missing constant can only be TDF since only this Darwinian total can provide a ceiling to c in rb>c. Like theproblem of observed rising sea levels relative to the land, Hamilton's gene will indeed spread whenever rb>c but this is not the point which is: WHY? Any tautology is true but by definition only. Hamilton's tautology likeany other, can say nothing about causation. Therefore, Hamilton's proposed gene centric causation, i.e. an independent in fitness gene level that can outdo the Darwinian organism level causing parental fitness altruism, can only constitute modelling misuse.

1. Please define a theory. Do you agree that theories cannot be scientifically misused?
2. Please also define a model and state the difference between a simplified and oversimplified model.
3. Do you agree that just a small cost to a larger gain is not altruistic ?
4. Do you concur that relatedness IBD is only a gene centric simplified/oversimplified model of organism centric NON IBD relatedness?
5. Do you agree that e is partly heritable?
6. Do you agree that Hamilton's Rule represents a tautological model within which cause and effect reverses?
7. Do you agree that only via the inclusion of a fitness constant within The Rule can the problem of contradictory propositions of causation be settled?