<!DOCTYPE html>
<!––This document was prepared with Bluefish 2.2.7 over Ubuntu. 
  Special symbols from http://www.starr.net/is/type/htmlcodes.html––>
<html>


<head>

<title>
JAVA LAB OF THE FALSIFICATION OF EVOLUTION
</title>

<meta name="keywords" content="falsification of the evolutionary theory,
falsification of evolution; evolution and complexity, 
evolution vs complexity; genetic algorithm, 
genetic programming; evolution with java,
laboratory of evolution; java laboratory of evolution, 
java lab of evolution, 
evolution,
java,
falsification">


<style>
/*Css styling from https://www.w3schools.com/css/default.asp*/
body {
    background-color: #ccffff; /*blue*/
    
}

h1 {
    color: #ff0000; /*red*/ 
    text-align: center;   
    font-size: 40px;
}

h2 {    
    text-align: center;
    font-size: 30px;
}


h3 {
  color: #ff0000; /*red*/ 
    text-align: center;
    font-size: 30px;
}

h4 {
    color: #ff0000; /*red*/   
    font-size: 20px;
}


h5 {
    color: #ff0000; /*red*/   
    font-size: 16px;
}

p {
    font-family: verdana;
    font-size: 20px;
}


</style>
</head>
<body>

<div align="center">
<h1>JAVA LAB OF THE FALSIFICATION OF EVOLUTION </h1>
<h2>Jos&#233; del Carmen Rodr&#237;guez Santamar&#237;a</h2>
<p>The Evoljava Community</p>
<p><a href="http://www.evoljava.com/"target="_blank">http://www.evoljava.com</a></p>
<p>7/X/2017, Bogot&#225;, Colombia.</p>
<h4>Abstract</h4>
</div>



<p>A sample of Java programs has been chosen to convey a 
message: evolution is an amazing tool in the hands of a genius but it is pretty useless otherwise. This is, of course, a 
challenge for everybody in our community. Anyway, evolution is, as any other tool, of limited power. In particular, 
there is no scientific reason to believe that it played an important role in the origin of species. This is my report to you of what I have been making in the last 44 years. By studying it, you can get a lot of ideas to make your own report to the world.   Take in mind that you enjoy an unconditional license to use every element of this site in whatever form you prefer, the present material included. Make your best to get a final product that must be fair, bold, clear for 
beginners but challenging for experts, robust and enticing.</p>

 <h3>1. INTRODUCTION</h3>

<p> Java enables a  thorough testing of the Evolutionary Theory because many questions can be studied in depth. So, we are looking for a natural falsification of that theory. This means that the falsification must arise from the study of evolution instead of being brought forth by ingenious tricks: we must calculate mandatory and ubiquitous predictions and compare them with observations to decide next that they are quite different. So, let us begin by noting that evolution is a methodology that we all use to solve problems of our ordinary life. </p>

<br>



<h3>2. EVOLUTION AS AN ALGORITHM</h3>



<p>The idea of evolution is <i>to change things to improve quality and to enrich options</i> 
by assembling patches, merging a 
proposal from here to a proposal from there. Let us see how this idea is transformed into a very concrete receipt or 
algorithm to solve problems.</p>

<br>

<h4>2.1  Starting with common sense</h4>


<p>We all use evolution as a  tool to solve problems everywhere and every now. Two examples will suffice to show this:</p>


<p>Suppose we need to design a menu that shall be low in sugar. So, one can try <span  style="color: #b3661a">rice + 
potato</span>, change 
<span  style="color: #b3661a">potato</span> by <span  style="color: #b3661a">black bread</span>  and if one has  tried 
 <span  style="color: #b3661a">pasta + rice</span> and also <span  style="color: #b3661a">rice +  beans</span>, 
what about  trying <span  style="color: #b3661a">pasta + beans</span>? In evolution, the 
first operation is called <i>mutation</i> and the second <i>recombination</i>.</p> 

<p>Imagine now that we want to fix a craft. If we buy a 
new part to replace an old and defective one, we are making a mutation. But if we exchange the defective part in the damaged craft by a functioning part from another object, we are executing recombination. </p>


<p>In the first example, we are trying to minimize sugar excess and in the second we try to maximize functionality. We are dealing in both cases 
with  <i>optimization problems</i>. To improve solutions  we use two techniques: mutation and recombination. This 
is <i>evolution</i> and is what we all  do everywhere and every now. With some practice, one can see evolution everywhere. </p>


<p>It happens that this methodology can be automated and generalized as follows: </p>


<ol>
 <li> An <i>optimization target</i> is defined   (minimize sugar excess). 
 <li>  A verbal or <i>genetic description</i> of the objects that are the subject  of  optimization must be given 
(rice, potatoes, bananas are all words that describe real objects). This step is also called <i>encoding</i>. 
<li>  A <i>population</i> of individuals (tentative solutions) is generated by random concatenation of symbols of a 
given alphabet 
(The alphabet for a menu are the ingredients: <span  style="color: #b3661a">rice, onions, potatoes..</span>. A tentative 
solution, a menu, is a 
string of 
those ingredients: <span  style="color: #b3661a">rice +  peas</span>). 
<li> Individuals are <i>mutated</i> 
and <i>recombined at random</i> and the resultant solutions are evaluated. To 
evaluate a solution, that is a verbal description, it must be decoded into an object of experimentation. The process of 
evaluation assigns a score or <i>fitness</i> to each solution. A common definition of the fitness function for 
maximizing problems assigns 0 to 
useless objects and 1 to perfect ones (<span  style="color: #b3661a">rise + potatoes</span> have a low fitness because they 
contain too  much sugar 
in the 
form of starch. By contrast, <span  style="color: #b3661a">beans + rice</span>  are well scored because beans are rich in 
proteins not in sugar).
<li> Fitnesses are used to build <i>a new population</i>, say, the 
probability of cloning is proportional to the fitness. So, fitness incarnates the feedback we are so used to.
<li>  This is <i>done over and over</i>, recurrently, until an acceptable solution is achieved. Each round  is 
called 
a <i>generation</i>. 

</ol>


 
<p>This receipt is one form of defining  <b>GA (Genetic Algorithms)</b> in general according to <i>Darwinian 
Evolution</i>. They are totally mechanistic and so they can be run by computers. Used language results from the fact that historically GAs were inspired by Genetics (things must have a verbal description) and by the Darwinian Evolutionary Theory. Real life contains variations of evolution that do not fit the automatism of 
Darwinism, rather some kind of guide or help is included. Example: a mother is always celebrating the deeds of her baby that he or she could span his or her mind as a source of creativity, of variation. Besides, she furnishes feedback at every moment that the baby could know how good is the performance of his or her decisions. To end this list, her nagging has as purpose to minimize the suffering of inappropriate decisions. This type of conditioning classifies better as 
<i>Lamarckian Evolution</i> and resultant algorithms are created by Artificial Intelligence practitioners (see vol 
VII, Hearts and minds).</p>


<p>Now, what can be achieved by Genetic Algorithms?</p>
 

<h4>2.2 Direct algorithms</h4>
 
 
<p> There are a multitude of problems that are solved right in front of own eyes by genetic algorithms. Let us call them 
 <b>direct problems</b>. Their existence and proliferation allowed many people to imagine that evolution is the final explanation of our existence. 
 A glorious representative of this school of thought is Richard Dawkins 
(<a href="https://cbs.asu.edu/sites/default/files/PDFS/Dawkins%20Puncturing%20Punctuationsim.pdf"target="_blank">1986</a>,
<a href="http://search.barnesandnoble.com/God-Delusion/Richard-Dawkins/e/9780618680009\#EXC"target="_blank">2006</a>). This type of literature is very important for those persons that, like the Author, believe that if evolution explains us then God has no right to judge us.  The persistence influence of Dawkins is captured here by   two 
important and direct  programs that adapt to Java two of his algorithms from the first  cited work:</p>



<ol>
 <li> <span style="background-color: yellow">Shakespeare4</span> from Vol XIV  that matches  a phrase.
 <li> <span style="background-color: yellow">Morphs</span> from Vol IV   that makes drawings that resembles insects and other critters.
 <li> To these we have added <span style="background-color: yellow">EvolArt</span> from Vol IV   that outputs patterns, some of which look artistic. 
</ol>

  
  <p><span style="background-color: yellow">Morphs</span> and <span style="background-color: yellow">EvolArt</span> create variability, while <span style="background-color: yellow">Shakespeare4</span> represents a  GA: it 
uses mutation, recombination, assessment of solutions and  reproduction recurrently to fit a target. The Reader is 
invited to verify on this program that  direct 
algorithms have three properties. First: they steadily gather small change generation after generation until   
perfection 
to the full is achieved 
(<i>to gather small change</i> is a trademark of Dawkins that is a synonym of <i>gradualism</i>, a trademark of 
Darwin). Second: 
when problems are nested,  solutions of tiny problems can be used as partial solutions of  enlarged ones. Third: it is 
productive to select the best exponents of a generation to start a new one. </p>

 
 
 
 
<h4>2.3 Lengthy algorithms</h4>
 
 
 <p>The fitness of a string in the program <span style="background-color: yellow">Shakespeare4</span> measures the number of matching sites. This feedback makes it possible to gather small change, site by site. So, what must a problem have in order to be not direct,  <b>lengthy </b>? All we need is to hinder gathering small change. As an example let us consider the problem in which one must guess a password.  In this case, you have a perfect match else you have nothing. There is no possibility to gather small change. As we know, guessing passwords is difficult and that is why they form part of the modern industry of cyber-security. Now, to understand why, let us compare three 
methods to solve the password  problem: by deterministic search, by randomness and by evolution.</p> 

<p>For sake of easiness, let us suppose that the password  number is  binary of length <i>n</i>, something like 1001110. For a 
deterministic search, we 
can use a for loop. The number of trials ranges from 1 to  2<sup>n</sup>. In average we have  2<sup>n</sup>/2 = 2<sup>n-1</sup> trials.  This problem is solved for binary numbers by randomness in program <span style="background-color: yellow">RandomSearch</span> from Vol 
IV. 
The program is intended to show that randomness is very powerful and that actually it is almighty in a word of unlimited resources. But our world is extremely limited and here randomness gets rapidly frustrated. In fact,  randomness can guess a number in the first trial but it can also fail until the end of time. That is why evolution has 
been sold by science as the obvious alternative to randomness, a propaganda that has been based on direct problems. Now, 
evolution is not better than chance to solve  
lengthy problems  and the    <span style="background-color: yellow">Password</span>  problem is enough to show it. Test this assertion with program  
<span style="background-color: yellow">ChanceVsEvolution</span>. </p>





<p>Intrigue: Is there in biology lengthy problems? If the answer is affirmative, are 
they rare?</p>
 
 
 
 <h3>3. EVOLUTION AS A BIO REALITY</h3>

<p>Astonishing as it might be,   <i>GAs can be implemented with molecules</i>. 
In fact, the first artificial GA that was run on 
DNA  was announced in 1994 by <a href="https://web.archive.org/web/20050206144827/http://www.usc.edu/dept/molecular-science/papers/fp-sci94.pdf"target="_blank">Adleman</a>. 
Nevertheless, GAs always have been running by nature over the DNA 
of every genome. Always.</p>

<p>Example: your black eyes are those of your maternal grandfather while your abundant blond hair is that of your maternal grandmother because the DNA of your grandfather 
recombined with that of your grandmother. At the same time, your bones are stronger than those of both mother and father because of a mutation in the DNA responsible for bone structuring. 
You resulted to be both handsome and bold and that is why you are so loved by girls. </p>

<p>So, what do we have here?  We have here mutation and recombination along a ruthless environment that is ready to award your qualities and 
to chastise your defects by measures that augment else depress your reproductive potential. 
This is the renown biological evolution formally acknowledged by science since 1850 thanks to  a field study made by 
 in Los Galapagos Islands (<a href="http://darwin-online.org.uk/converted/pdf/1861_OriginNY_F382.pdf"target="_blank">Darwin, 1859</a> )  plus a lead triggered by Mendel 
(<a href="http://www.esp.org/foundations/genetics/classical/gm-65.pdf"target="_blank">1866</a>) and that ended in modern Molecular Biology. </p>



<h4>3.1 Evolution in the hands of a genius</h4>



<p>We preach that evolution is as powerful as your imagination. So, it would be good to look at a 
<i>smart example</i>. The preferred one of the Author is that of the immune system to synthesize antibodies that defend us from bacteria, 
viruses and cancer (<a href="http://www.imgt.org/IMGTeducation/Tutorials/ImmuneSystem/UK/the_immune_system.pdf"target="_blank">NIH, 2003</a>). 
Antibodies have two parts, one constant, the other variable. 
The variable part is taken from a 
combinatorial basis and responds for the specificity of each antibody.  The GA consist in the combinatorial basis plus the gluing procedure of the variable and constant parts. 
The used GA is very simple but it is startling because it is the inspiring example of the strategy of <i>using evolution to solve a problem one step off the solution</i>. 
Besides, 
the immune system  uses the very 
modern <span style="background-color: yellow">n-gram</span> technology to distinguish self from 
non-self (<a href="https://en.wikipedia.org/wiki/N-gram"target="_blank">Wikipedia, 2017</a>). This technology is applied in  
<span style="background-color: yellow">Program I22 English</span>  of Vol IX 
to 
discriminate the language used to write a  text, German else English. </p>

 


<h3>4. EVOLUTION AS A THEORY</h3>








 
<p><i>Evolution is strongly tied to Darwin</i>. An important event in his life  was the death of his beloved daughter 
Ann at age 10 (<a href="https://en.wikipedia.org/wiki/Anne_Darwin"target="_blank">Wikipedia, 2017b</a>). 
We adhere to the consideration that the ensued sadness impulsed him to change the faith in a God-with-us 
by a faith in an autonomous process dubbed natural selection that kills less fitted individuals and allows fitted to survive and reproduce. 
After that he published his <i>Origin</i> (Darwin,  l.c. 1859 ):</p>



<p><i>
 It is interesting to contemplate an entangled bank, 
 clothed with many plants of many kinds, 
 with birds singing on the bushes, 
 with various insects flitting about, 
 and with worms crawling through the damp earth, 
 and to reflect that these elaborately constructed forms, 
 so different from each other, 
 and dependent on each other in so complex a manner, 
 have all been produced by laws acting around us ... 
 Thus, from the war of nature, 
 from famine and death, 
 the most exalted object which we are capable of conceiving, 
 namely, the production of the higher animals, 
 directly follows. 
 There is grandeur in this view of life, 
 with its several powers, 
 having been originally breathed into a few forms or into one; 
 and that, 
 whilst this planet has gone cycling on 
 according to the fixed law of gravity, 
 from so simple a beginning endless forms most beautiful 
 and most wonderful have been, and are being, evolved.
</i></p>


<p>While this was written in 1859, the further development of the theory led to the postulate that there is no gap in within ordinary and living matters. 
So, we officially became sons and daughters of the dust of the stars.
Additionally,   GAs were formulated around 1950 following the model of natural evolution. 
Nevertheless, this was not a historical necessity because, as everyone knows by instinct,  GAs derive from common sense, from ordinary life. 
By the same token, Darwin was not a historical necessity. In fact, <i>to invent the Evolutionary Theory is a must for 
everybody</i> if only we merge the following points:</p>

<ol>
 <li> GAs always have being running by nature around us.
 <li> Variants synthesized by natural GAs have different reproductive potential. 
 <li> Our entrails are terribly similar to those of a hen.
</ol>





<p>After summing up we will   exclaim: <i>We are here by evolution!</i> If besides we remember that mutation and 
recombination 
happen at 
random, we will reinforce the  suspicion that God has nothing to do with us.</p> 

<p>More explicitly, our <i>Evolutionary Theory would say</i>: mutation and recombination produce variants with diverse 
reproductive potential so, some individuals leave more spring than others while variability is being created. This process is repeated year over year and in the 
long run we have  that from simple organisms   the Earth has been filled in life with all the  
grandeur 
and richness that we see everywhere.</p>




<h3>5. EVOLUTION AS SCIENCE</h3>





<p>It is good to have thoughts  but  it is advantageous  <b>to test ideas </b> if only this is possible. 
Now, <i>there are too much data supporting the Evolutionary Theory</i>. Let it be enough to recall that the DNA of a 
chimp and 
that of a human differ by no more than 5% (<a href="http://www.pnas.org/content/99/21/13633.full"target="_blank">Britten, 2002</a>). 
Now, if a lot of data supports a theory, is it correct? No, by no means. 
The situation is similar to a cousin that pretends to be your brother in order to share a part of your inheritance. 
He can bring you thousands of true facts to support his claim, family stories and partial-DNA studies included. But all 
we need to prove him wrong is to show that his DNA has long chunks that do not appear in that of your fathers.</p>  


<p>By the same token, Java can help us to bring data that support the Evolutionary Theory but our great commitment is to show that Java is powerful 
to provide insight that allows us to see how false is that theory. Let us see.</p> 

<h4>5.1 The hand as test</h4>

<p>The hand is so important for us that it is a natural election for a person that wants  <b>to test evolution over a 
concrete and important trait </b>. At first sight, a hand is simple. In fact, to make one with pieces of wood  and cords is a task for freshmen. 
Since it seems so easy, this becomes automatically a challenge for us. A portable  
solution is proposed in <span style="background-color: yellow">Program M29 HandOnlyBoxess12</span>  of Vol XIII.  
So, the human hand does not seem to be out of the scope of evolution that has as raw material the hand of 
the chimp (<a href="http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1571064/"target="_blank">Young, 2003</a>).
 <span style="background-color: yellow">Program M124 HandChimpHuman4</span>  of Vol XIII proposes a cartoon of  the 
evolutionary thinking.   By the same token, <i>if we suppose that bones are evolvable</i>   and 
if we already have a theory to explain the architecture and relative measures of the components of the hand, 
then evolution can fit those measures straightforwardly.  This is illustrated by <span style="background-color: yellow">Program M48 FossilRecord5</span>  of 
Vol XIII. 
So, we ask: where is the theory? The intelligent answer is that the best theory is no theory at all because evolution is sufficient. This line of thought  is exemplified as follows: </p>





<p>A finger with 3 phalanges and its corresponding metacarpal is similar to a tail. Thus, the simplest biological boned model of a hand is a tail that we
oversimplified next to a broken line with 4 strokes.  Our immediate 
purpose   is to prove that evolution can reproduce the fact that almost all the vertebrae of the tail of a monkey have 
the 
same length.  Our proof has two parts: first, we reformulate the problem as an optimization one and next we use evolution to solve it 
and to show that the corresponding algorithm is direct because the solution is arrived at just in front of own eyes. So, our optimization problem reads as follows:  the tail must maximize the grabbing of
cylindrical objects such as the arm of a tree. In a 2D cartoon,  the tail must cover a circle allowing maximal grasping. 
Our guess is that this is accomplished when the space in within the circle and the formed quadrilateral achieves a 
minimum. We must observe a constraint: the number of bones is fixed.</p>



<!––
\begin{center}

\psset{unit=2cm}


\begin{pspicture}(0,-1.29)(5.0,1.29)
\pscircle[linewidth=0.04,dimen=outer](1.46,-0.55){0.7}
\pscircle[linewidth=0.04,dimen=outer](4.28,-0.55){0.7}
\psline[linewidth=0.04cm](2.16,-1.25)(0.96,-1.25)
\psline[linewidth=0.04cm](0.96,-1.25)(0.0,1.25)
\psline[linewidth=0.04cm](0.0,1.27)(2.52,-0.43)
\psline[linewidth=0.04cm](2.5,-0.43)(2.14,-1.27)
\psline[linewidth=0.04cm](4.96,-1.25)(3.58,-1.25)
\psline[linewidth=0.04cm](3.58,-1.25)(3.6,0.17)
\psline[linewidth=0.04cm](3.6,0.17)(4.98,0.15)
\psline[linewidth=0.04cm](4.98,0.15)(4.96,-1.27)
\end{pspicture} 

\end{center}

\emph{Figure \thefigure. A tail with four  bones can embrace a circle but the space in within the 
circle and 
the resultant quadrilateral might vary. Our purpose is to select the arrange with the minimal 
  space in excess. We expect a square. Our guess is that this arrangement will favor the 
maximal 
grabbing  of the circle by the tail that we idealize as a polyline with 4 strokes.}
\stepcounter{figure}

––>

<p> The code 
can be found in <span style="background-color: yellow">program P73 EvolutionOfTail3</span> of Vol XV. It allows us to test our expectation. If we succeed, 
we have an 
invitation to formulate more severe tests to the belief that evolution is responsible for the approximate equality of the vertebrae of snakes,  
of the tail of monkeys and of the phalanges of the hand. Beware: we are assuming that bones are evolvable, i.e., that evolution can  deform the shape of bones.</p>





<h4>5.2 Evolution without purpose</h4>

<p>When we assume that the hand has a function, we must face the accusation that we are humanizing evolution because the 
evolution envisaged by Darwinism has no purpose at all. What could be the meaning of such an accusation? 
<span style="background-color: yellow">Program M63 FossilRecordd10</span> of Vol XIII might help us to understand it:  some few random mutations of bones of the hand are 
given to operate in tandem. </p>


<p>We see that our machinery can produce complex structures similar to composed tails or fingers. Nevertheless, the 
Author considers that resulting structures naturally inspire the beholder to see potential weapons:   axes, awls, 
hammers, and maces. Therefore, a prediction immediately springs: a trend towards sophistication of hands that are weapons 
by themselves must have arisen with a clear track in the fossil record.  This prediction makes sense in the presence of a high doses of intelligence, 
both to understand that a hand can be used as a weapon and to exercise sexual selection. If intelligence is lacked, we can think of coevolution. 
If this is also lacked, the only thing that we can predict is too much variability. Nothing like that is found anywhere. 
</p>


<p>If coevolution of the brain and the hand is accepted, it allows us to study the possibility to consider 
that the hand has indeed a function for its owner and so, we can continue with  the study of its evolution.</p>



<h4>5.3 Adding realism</h4>





<p> <b>The human hand is a marvel that is more clever than complex</b> (<a href="https://www.youtube.com/watch?v=ToM51CO0Lyc"target="_blank">BBC, 2014</a>)  hence, we have plenty 
of 
suspicions that evolution is not its explanation. Now, this assertion is 
produced by an 
instinctive reaction. Our directive is that     <b>instinct, common sense, and religious beliefs must provide the fuel 
for hard work that shall be addressed to output objective material </b>. So, let us 
continue 
in the quest for clear-cut falsifications of the Evolutionary Theory. In this regard, let us pay attention to    
a simple trait    that differentiates  human and chimp hands.</p>



<p>Human beings can write but chimps not. That is due to the fact that we have some muscles associated to the thumb that the chimp lacks.
 In principle, four muscles are necessary and sufficient to write:  one that pulls towards the left,  other to the right, 
  a third towards the front and the last towards the rear. Now, chimps can take hold of arms of trees much as we, although their muscles of their thumbs are weaker. So, 
the 
chimp has two of the four muscles that would be needed to write, one that pulls towards the front and the other towards the rear. 
The hand of the chimp lacks two additional muscles to move the thumb to the left and to the right. 
Curiously enough, anatomists speak of three additional muscles instead of two (Young, l.c.  2003). Those muscles 
are: flexor pollicis longus muscle (<a href="https://en.wikipedia.org/wiki/Flexor_pollicis_longus_muscle"target="_blank">Wikipedia, 2016a</a>), the deep head of the flexor 
pollicis brevis (<a href="https://en.wikipedia.org/wiki/Flexor_pollicis_brevis_muscle"target="_blank">Wikipedia, 2016b</a>) and the first volar or palmar  interosseous muscle 
(<a href="https://en.wikipedia.org/wiki/Palmar_interossei_muscles"target="_blank">Wikipedia, 2016c</a>). The Author does not understand the role of that redundancy: 
if more than two extra muscles would be desired, four or eight would be a better figure. 
In general, it seems to me that our Community will need some 30 years of hard and wise work to mature its understanding of the hand. 
The procedure is to formulate hypotheses that could be calculated with Java. Example:</p>






<p>Duplication of fingers is a 
common mutation and so, most people know that there are persons with six fingers in their hands 
whose   neural connections work perfectly but without independence. The author has no idea about 
the 
rate of duplication of separate muscles but no theory  can convince him that such 
separate events cannot exist. Therefore, we will rely on the possibility of duplication of muscles together with a change in their insertion points. So, our stand reads as follows:</p>



<p><i>Dogma: Muscles can be duplicated and their insertion points can migrate. Moreover, 
all these changes happen at random. </i></p>



<p><i>Null Hypothesis: Random mutations that cause duplication of muscles and migration of their insertion points are the sufficient cause 
of the appearing of the human hand beginning from the hand of the chimp or of some related common ancestor. Natural selection of mutants for the sophistication 
of functions - throwing, clubbing,
manufacturing and dexter use of tools-  filtered mutants to end with the type of hand we enjoy now. </i></p>



<p>To fulfill our sacred duty of studying these null hypothesis has been found to be extremely complicated. 
Our glory is just to exhibit a demo that simulates random duplications and relocations of muscles. This is shown by <span style="background-color: yellow">Program 
MuscleMutation3Portable2</span> that is a   simplified version of program 
<span style="background-color: yellow">P83 
MuscleMutation3</span> of Vol XV. By looking at the demo, one tends to believe   that if the human hand were the result 
of 
the evolution of the chimp, then too many malfunctional hands also should have appeared. Nevertheless, to make this 
observation into a mandatory prediction for fossils, we need as yet to determine whether or not the ensuing problem of selecting perfection is direct else lengthy.  
<b>This is an open task</b>.</p>



<p>This case of the hand shows that  <b>to calculate specific predictions of the Evolutionary Theory</b> is an important 
part of 
our duty. But we can also  <b>reason at a general level thanks to abstraction</b> that results from 
capturing specific properties that presumably dominate the scene in question. A proposal for our case follows.</p>



<h4>5.4 The fundamental test</h4>

 <p><b>The exceeding perfection of life that goes on top of  its  overwhelming complexity</b>   constitutes  <b>the 
fundamental test </b> 
that everyone has the duty to present to the Evolutionary Theory. In fact, the path to perfection is paved with  
suffering. Or, objectively, perfection 
is preceded by too many, myriads of non functional products that are followed by mediocre improvements that are 
ensued by imperfect and then by almost perfect 
solutions.   Is that backed by every sort of malformations and malfunctions as in the fossil 
record as in living populations as in our own bodies?</p> 

<p>The wrong answer is affirmative and is defended as follows: unicellular microorganisms  were the first to appear. Then, 
a nucleus with membrane augmented complexity. Next, simple multicellular organism were formed. These were followed by 
fishes, reptiles, mammals, hominids and at last by man. This answer is wrong because a bacterium is a cell as perfect 
as anyone in my body.  More to the point and according to science, bacteria have been tested during more than 
600 million years but my cells during  4 millions only. 

Now, let us notice that most babies along every species are living marvels. In the light of their perfection and 
complexity, this amounts to a severe   falsification of the Evolutionary Theory. Now, can 
we dress our ideas in  objective results? Yes, we can: the first step along the path from subjectivity to objectivity  
is to 
formulate models  because they are precisely defined and so, can be studied and criticized by everybody. Let 
us look at 2 of them. </p>



<h4>5.5 A parable of complexity</h4>

<p>Below we will present an official model of complexity that might look too weird. So, let us present  a very simple 
model that at last conveys the very same message than the official one. It is so simple that its general behavior  
can be calculated right in the head. Let us see.</p>
                                                      

<p>Our first model of complexity was to guess a   number: this model says simply that one 
can achieve any design by 
randomness and/or evolution but that  we would  need eventually too much time to see it, might be more than the age of 
the 
Universe. Many items of our bodies do not fit this model: we can suffer too much damage and nevertheless we can 
survive. We need a model that is compatible with both complexity and partial but not mortal damage. Our proposal 
is to consider the model problem of  guessing an ordered set of k numbers of increasing length, say, 9, 14, 326, 5738, 
...  For this problem one gets 
a point  if one guesses a number. So, the fitness function would take values on the set 1/k, 2/k, ..., k/k. This model 
predicts that functionality must 
grow in leaps from almost nothing to perfection but with long periods of ecstasy in within that are the longer the 
larger are the numbers to be guessed. Functionality is expected here to grow with great laziness from 
incipient functions to mediocre performance to almost perfection to perfection -if only time is unbounded.  This model 
classifies  as a parable because it is too subjective, i.e., it is not an abstraction of  concrete facts of 
biology. What else do we have?</p>



<h4>5.6 An official model of complexity</h4>




<p>Let us depart from  <b>an abstraction of biological function </b>. To fix ideas let us  consider a biochemical function 
of the enzyme Hexokinase that catalyzes the reaction</p> 



<p>&alpha;-D-Glucose + ATP ->  Glucose-6-phosphate (G6P) + ADP.</p> 



<p>So, 
the function of this  enzyme is to activate glucose that it could participate in exo-energetic reactions. Without this 
step there is no 
energy for many forms of life. We can use the language of mathematical functions to encode this biochemical function: 
to the input 
 (&alpha;-D-Glucose , ATP) associate (Glucose-6-phosphate, ADP) but leave intact the rest of the world, Or,    to any 
other input associates the same input. This last statement says that this enzyme is highly 
selective and so enables the on-demand formation of specific biochemical pathways.  

From this we learn that biological function can be encoded in the language of mathematical functions and that every 
functional 
design in as elaboration of a generic proposition: do this  (activate the glucose) but not that (activate or 
change anyhow something else). Let us notice that to the question Did you do 
this? one can answer yes else no, true else false. So, the phrase 'The Hexokinase catalyzes a phosphorylation reaction' 
is true, while 'The Hexokinase can
phosphorylate fatty acids' is false. 
In general, any expression that can be classified as true or false given a   reference set   is called a 
 <b>proposition</b>. True and false also are denoted T and F and as  1 and 0. These are 
appropriate to describe concrete events of 
nature.</p>




<p>One can generate    <b>formulas </b> or compound propositions beginning from elementary propositions or 
variables or from other formulas by the use of logical operators:</p>
<ul>
 <li> The <i>NOT</i> or negation operator: which takes one proposition <i>p</i> and produces the proposition denoted 
<i>NOT(p)</i>  or  &#126;<i>p</i> that is false whenever <i>p</i> is true and is true whenever <i>p</i> is false.
 <li> The <i>AND</i> operator takes two propositions <i>p</i> and <i>q</i> and produces a third one, noted <i>AND(p,q)</i> or 
   <i>p&and;q</i>   that is true if and only if both  of <i>p</i>  <b>and </b> <i>q</i> are true.
 <li> The <i>OR</i> operator takes two formulas <i>p</i> and <i>q</i> and produces a third one, denoted <i>OR(p,q)</i> or <i>p&or;q</i> 
 that is true if one of    <i>p</i>  <b>or</b> <i>q</i> is true and false if both are false.
</ul>



<p>One can see that with the set of operators <i>{NOT, AND, OR}</i> one has more than enough to generate all possible operators 
that take one,  two or any number of inputs to produce one compound proposition. For that reason, we restrict ourselves 
to these operators.</p>



<p> <b>Every proposition encloses a mathematical function</b> that is per se a full description of the possibilities and 
restrictions in regard with design of useful crafts. Example: <i>p &and; ( q &or; (&#126; r)) = AND(p,OR(q,NOT(r)))</i> 
associates true to <i>(true,  true, *)</i> that means  <i>p</i> is true, <i>q</i> is true,  <i>r</i> could be true or false. It also 
associates true to <i>(T, *, F)</i>. It associates False to <i>(F, *,*)</i>.  We have calculated the output for all 8 possible inputs.</p>

<p>In general, we say that a proposition is    <b>satisfiable </b> when we can show a truth assignment 
to its 
variables that makes the proposition true. We also say that the truth assignment satisfies the proposition or the 
formula. So,  <i>p &and; ( q &or; (&#126; r)) </i> is satisfiable - take <i>(T,T,T)</i>.   By contrast,  <i>(r &and; (&#126; r)) = 
AND(r,NOT(r))</i> is not because it is a contradiction, it always outputs <i>F</i>. A contradiction 
might represent  do-undo cycles that so abundantly happen in nature. Example: a world to be ecological in the long 
range needs to undo everything that it does. Anyway, the do and undo functions are separated in most 
cases as in time as in space and executed by possibly unrelated actors. That is why we center our attention over 
satisfiable propositions that represent doing, creating, designing. So, we 
define:</p> 




 <p><b>SAT or the satisfiability problem  of mathematical logic </b> consists in studying the  assignment of  
truth values to the variables of a proposition  with the purpose of making  it true. SAT proposes a question in regard 
with a given proposition: Does this proposition is satisfiable? To answer YES you must show a satisfying assignment 
otherwise you must prove that all assignments evaluate to false.</p>



<p>We can perceive the power of SAT to capture complexity if we embed it into real life. To that aim, let us imagine that  
the chief declares that there is a bonus in big money if all persons fulfill their tasks as scheduled. Tasks are 
realizable but these  demand  effort and concentration. If the team consists of one or two persons, we have no problem. 
But if the team consists of three persons or more,  the chief cannot deny that he is trying to win a lot of money at 
zero cost by manipulating the people with a void promise. If we  reformulate this example for a group of 7 persons as a 
proposition, we get    <i>p &and; q &and; r  &and; s &and; t &and; u &and; v</i>, which  is exactly equal to guessing 
the binary number 1111111. Now, to match by a deterministic algorithm a binary number with <i>n</i> figures one needs 2<sup>n</sup>  
trials in the worst case.
This is an 
exponential function that overgrows very rapidly. Say,  2<sup>300</sup> > 10<sup>80</sup> which is the number of atoms in the 
universe (<a href="https://www.universetoday.com/36302/atoms-in-the-universe/"target="_blank">Villanueva,  2015</a>). That is why both problems, guessing a number and SAT, are lengthy.</p> 






<p>SAT is a mine of experimental possibilities.  One is related to  
 <b>conjunctive normal forms </b>, which are expressions of the form</p>




<p><i>(&#126; p &or;   q &or; &#126; r) &and; (&#126; p &or;   q &or; &#126; r)</i></p>



<p>This formula has 2  <b>clauses </b> on 3 
 <b>variables </b>. The clauses are <i>(&#126; p &or;   q &or; &#126; r)</i> and <i> (&#126; p &or;   q &or; &#126; r)</i> and the 
variables are <i>p,q,r</i>. Conjunctive normal forms have satisfiable 
exponents, such as <i>p &and; q</i> which is true for the <i>TT</i> evaluation. And also have not satisfiable members such as 
<i>p&and; &#126; p</i>, which is a contradiction. Now, we can read the   overall structure of these forms: 
the  
ANDS, that join clauses,  represent 
restrictions: do this and that and that. You are not allowed to do whatever comes to the mind.  But the clauses contain 
diverse options. This means that  conjunctive normal 
forms represent as 
restrictions as complementary optional choices.</p> 

<p>What do we lost if we restrict ourselves to the study of conjunctive normal forms? We consider that we lost not too 
much. Our reasoning goes as follows: the formula  <i>p &and; q &and; r  &and; s &and; t &and; u &and; v</i> belongs 
in the class of conjunctive normal forms. Moreover, it is exactly equal to guessing 
the binary number 1111111, which is a lengthy problem. On the other side, the formula with just one clause and too many 
variables, like <i>p &or; q &or; r  &or; s &or; t &or; u &or; v</i> is almost a tautology because it gets true for all assignments except one: 
<i>FFFFFFF</i>. This means that conjunctive normal forms expand the  world of complexity almost completely.</p> 




<p>Conjunctive normal forms are important for us because they allow a more gradualistic fitness than SAT:</p> 



<p>We redefine the fitness of a valuation as the number of clauses in the proposition that are 
satisfied by the valuation. For instance, if we have the proposition <i>(p &or; q) &and; (&#126; p &or; q)</i>, the fitness 
of 
the valuation   <i>TF</i>, true for <i>p</i> and false for <i>q</i>, is 1 because with that valuation, the clause <i>(p &or; q) </i> gets 
true 
while <i> (&#126; p &or; q)</i> is false. In the SAT semantics, the fitness should had been 0 because the whole proposition 
gets false with the given valuation. This new fitness induces the following  optimization problem:</p>




<p> <b>Max-K-SAT=Maximum K-satisfiability </b> is the problem whose goal is 
to 
find an assignment that maximizes the number of satisfied clauses for a given conjunctive normal  formula over N 
variables, where each clause contains at most K variables and each chosen variable  can be negated. We see that a 
proposition is satisfiable if and only if  Max-K-SAT ends with a valuation that satisfies all clauses, i.e. if the 
maximal fitness in the population is equal to the number of clauses in the proposition. So, Max-K-SAT extends 
SAT. But we have a gradualistic extension because the fitness can effectively take on intermediate values that lie in 
within the extreme ones. So, evolution is expected to outperform randomness when it is given the task to solve MaxKSAT 
problems. This means that the fitness function must increase after the initial generation that is synthesized by 
randomness. But, how much? With  Program  MaxKSATvsEvo one can test the following assertions:</p> 

<ul>
 <li> Randomness is very good to produce mediocre solutions. These appear in the first generation.
 <li> Evolution is good to improve the mark of randomness.
 <li> When the ratio of variables per clause is small the problem is lengthy for evolution because it becomes similar 
to matching the binary number <i>TTT...T</i> and progress comes in 
laps that are interspaced by ever longer periods of ecstasy. Otherwise  the proposition is 
satisfiable and the problem is direct for evolution. 
</ul>








<h3>6. CONUNDRUMS</h3>


<p>Have you ever seen  a kid  trying to use a hammer to intake water? Probably not. It 
seems that we learn very early in life that  <b>no tool is good for everything </b>. This is another fold of 
complexity. 
Let us see now that evolution, being a real tool and  not magical, has conundrums of primordial biological 
importance that it cannot solve. Else, cosmological time is needed to go beyond poor and mediocre solutions.</p>



<h4>6.1 Evolution vs evolution</h4>


<p>Let us study the following problem in which we cause  <b>a head to head collision of evolution as a tool, an 
algorithm, and 
evolution as a theory </b> that says that the tool is almighty in the world of problems.</p> 

<p>We  depart from a tree T, calculate a distance matrix among its leaves M and use evolution to 
restore the tree T from the matrix M. The distance in a tree from 
one leaf to another is   defined as the minimum distance along all paths that connect the two 
leaves.</p> 

<PRE>
 A     D  B      C     E
6|    1|  |      |8    |0 
 |_____|  |7     |_____|
    |c    |        b|
   3|_____|         |8
      a|            |
       |2___________|
 </PRE>
 


<p><i>This tree shows that leave A is more related to D than to E. In fact, the distance from A 
to D is 7 while that of A to E is 19.</i> </p>
 






<p>This problem is given to evolution in   <b>Program K65 EvolFull10 </b> from Vol XI. Our results are impressive: the 
Evolutionary Algorithm can solve the problem for 3 and 4 leaves    but fails 
from 5 on. Now, why is this problem so difficult for
evolution? Let us propose an explanation that goes along the previous
graphic:</p>




<p>Depicted tree conveys information about the distances in within every pair of leaves, data that can be consigned in a 
matrix:</p>



<table>
<tr><td></td><td>A</td><td>B</td><td>C</td><td>D</td><td>E</td></tr>
<tr><td>A</td><td>0</td><td>16</td><td>27</td><td>7</td><td>19</td></tr>
<tr><td>B</td><td>16</td><td>0</td><td>25</td><td>11</td><td>17</td></tr>
<tr><td>C</td><td>27</td><td>25</td><td>0</td><td>22</td><td>8</td></tr>
<tr><td>D</td><td>7</td><td>11</td><td>22</td><td>0</td><td>14</td></tr>
<tr><td>E</td><td>19</td><td>17</td><td>8</td><td>14</td><td>0</td></tr>
</table>



<p>Let us suppose now that the problem is to find the tree whose  distance matrix is</p>


<table>
<tr><td></td><td>A</td><td>B</td><td>C</td><td>D</td><td>E</td></tr>
<tr><td>A</td><td>0</td><td>16</td><td>27</td><td>7</td><td>19</td></tr>
<tr><td>B</td><td>16</td><td>0</td><td>25</td><td>11</td><td>17</td></tr>
<tr><td>C</td><td>27</td><td>25</td><td>0</td><td>22</td><td>10</td></tr>
<tr><td>D</td><td>7</td><td>11</td><td>22</td><td>0</td><td>14</td></tr>
<tr><td>E</td><td>19</td><td>17</td><td>10</td><td>14</td><td>0</td></tr>
</table>




<p>We see that we must match the values of all links of a tree. So, 
this problem is similar to that of guessing a number and so, we expect a lengthy problem. Some lengthy problems allow 
to gather change through the generations.  But proposed program  shows that when the number of leaves is greater 
than 5 
the algorithm does not converge to the targeted tree and matrix. Why?</p>

<p>Let us suppose that we departed from the target matrix cited above and that we have arrived to the depicted tree of the 
figure with its corresponding  matrix. If we 
compare the two matrices, we see that they differ by just one entry, the distance in within C and E. By looking at the 
depicted tree, one might naively think that the obvious solution is to change the length of
the link to C from 8 to 10. This will indeed remedy the problem with the distance
between C and E. But the price we must pay is that all other distances
will change. Say, the distance between A and C will change from 27 to 29. And so
on: every distance will suffer a change. This implies that every single change,
tiny as it could be, will cause the error to increase.</p> 

<p>So, the difficulty of posed problem is due to the fact that every change 
has side effects all around. This allows us to formulate a warning: one usually teaches that an optimization problem 
generates a landscape with peaks and valleys much as a landscape in our planet. But we see that  
 the landscape associated with this problem is not  constant. Rather it changes over every step much as it 
happens with a person that has to walk a slack high-wire.</p> 

<p>The common sense example of this conundrum is the battle against   death. It admits no treatment 
because of the negative side effects created by remedies  and no-remedies alike.   
The same situation is found in genetics  and carries the name of  <b>pleitropy </b> and happens when one gene affects 
many 
traits. In our example of the tree, a gene is 
represented by the value of a link in the tree and the traits are the distances among  leaves.  Pleitropy was known to 
doctors since long ago because when one pleitropic gene fails a lot of problems arise  as in the Marfan Syndrome 
(<a href="https://www.marfan.org/about/marfan"target="_blank">The Marfan Foundation, 2014</a>). 
The  anti-evolutionary role of pleitropy is mathematically well established  
(<a href="http://www.dwaxman.com/Papers/Perfect.PDF"target="_blank">Waxman and Peck, 1998</a>).</p> 
   
   
   
   
<h4>6.2 Evolution vs complexity</h4>  

<p> <b>We have introduced two methods to create complexity: adding requirement over requirement, modeled by SAT,  and 
to use 
pleitropy to 
entangle functions. </b> We have argued that these methods are a terrible burden.  But we also have seen that evolution 
is 
 powerful. What could be the net effect of these two factors operating together over fossils? If we play with both 
factors shoulder to shoulder, complexity and evolution, we shall predict a huge amount of fossils amidst a lazy trend 
to increasing complexity. This is so self evident that it becomes a challenge for us. So, we have prepared a Java   
simulation  in which worms must crawl up the screen. Initially they 
have just some few links but   successful worms get enlarged by one link. This captures 
the power of evolution to increase complexity.  At the same time, a synchronized work of all 
links is necessary to crawl up  so, this problem is like guessing a number and therefore it is lengthy. Moreover, we 
have a process of fossilization that runs in background. Fossils can be studied a posteriori in order to answer a 
question:  What is the 
proportion of perfect fossils that climb the screen to that of  imperfect ones that do not? The code can be seen in 
 <span style="background-color: yellow">Program L112 WormsRandJFXa10</span> of Vol XII.</p>  


   
   
  <p> We must recognize  now that  <b>a model is no more than an idea </b>. But  <b>science compares ideas with facts 
of nature </b>. So, let 
us work out some possible realizations of pleitropy in biology.</p>





<h4>6.3 Living fossils</h4>



<p><span style="background-color: yellow">Program N28 DisfiguredSphere2</span> of Vol 14 shows that mutation can erode shape immediately: not in vain the  
left 
and 
right 
sides of our 
bodies are always different. So, how is it that we human beings are all 
so similar one to another along history and continents? How is 
that possible? Our idea is to involve pleitropy as follows:</p>




<p>Shape in biology must be a natural byproduct of  programmed reproduction,
differentiation and death of cells. 
Now, the reproduction of cells is described by a tree: the mother cell is the root and its 
descendants form the arms of the tree. Let us assume that cell reproduction is equivalent to cell 
splitting so, 
our trees are binary. This representation of temporal events depicts a mother cell that disappears 
to give rise to two daughters. But if we pay attention to space events, a cell can give rise to 
many children that fill surrounding space. In space representation, reproduction would be best 
depicted by  a tree with various children. The number of children may be supposed to be self 
regulated   to fill in vacua.</p> 

<p>On the other hand, our previous example of  pleitropy   resulted from the 
problem of fitting a tree to a distance matrix. Let us apply the very same model to ontogeny: the tree is given by 
cellular reproduction    and the 
overall fitness function that depends on many interconnected traits is the    shape itself that 
is possibly determined 
by the  cellular interactions   with the rest of the body.</p>  


<p>To test the idea that this setting is a realization of pleitropy and that therefore it is a conundrum for evolution,  
we propose the following simulation: evolution is given the task of fitting  a cube of 
size <i>l</i> by a tree. This must fill the interior of the cube as well as its boundary points for integer values. This 
simulates 
the 
ontogeny of a  bone. Evolution does it directly for <i>l = 1</i>. For <i>l = 2</i> evolution also succeeds although sometimes 
fails to do it immediately. Incredibly, evolution succeeds for <i>l = 3</i>.  <span style="background-color: yellow">Program   N114  TreeToShapes17</span> of 
Vol 
XIV 
shows us that a perfect fitting is too ambitions for <i>l > 3 </i>  so, one posits as task to fit a fraction  
of points, say 0.7. This is easy for evolution. But perfection is outside its scope: evolution cannot navigate freely 
in the space of trees that fit into a cube of side 4 or larger. This is our answer to the question: are bones 
evolvable? </p>

<p>Thus,   pleiotropy is all pervading along biological shape if only shapes have  some selective value. Now,  we give a 
very high selective meaning to  wings, the hands of birds, and so, we are happy when we see how they land. 
Nevertheless, 
we deny  an important selective meaning to the shape of the human hand.     Now, by looking at own hands, how perfect 
they are, one feels that our explanation is partial and that something else must exists.</p> 



<h5>The genetic code</h5>











<p>A  <b>living fossil </b> is the name given to a species that lives amidst us and whose first specimen  appeared
millions years ago (<a href="http://www.ucmp.berkeley.edu/bacteria/bacteriafr.html"target="_blank">Waggoner,  2013</a>; 
<a href="http://rstb.royalsocietypublishing.org/content/361/1470/887"target="_blank">Brasier et al,  2006</a>; 
<a href="http://onlinelibrary.wiley.com/doi/10.1002/bies.201200145/full"target="_blank">Casane, Laurenti, \cite{casane13}, 2013</a>). 
The most strange of all living fossils is 
 <b>the 
genetic code </b>. In fact, 
the genetic code is at the bottom of everything that means life so, it is is used and reused everywhere and every now. 
As a consequence, it enjoys the highest possible degree of pleiotropy.  So, our bet shall be that there is no 
variability of the genetic code. Nevertheless, this is blatantly false: the  mitochondria and nuclei of human 
cells have different genetic codes (<a href="https://www.ncbi.nlm.nih.gov/Taxonomy/Utils/wprintgc.cgi"target="_blank">Elzanowski et al,  2016</a>)  
and various simulation studies  show us that it is worth considering the idea 
that observed variants of the genetic code  appeared by evolution 
(<a href="http://resolver.caltech.edu/CaltechETD:etd-09042007-091804"target="_blank">Ofria,  1999</a>). 
This is a fascinating theme for research that cannot be treated in isolation.</p> 


<h5>Enzymes and monomorphism</h5>


<p>We abstracted SAT from the study of functions and we found it to generate as lengthy as direct problems. Nevertheless, 
SAT  is simple per se: in the space of truth assignments one tries to answer the question: does a system that is 
described by a given formula have at least one useful property for design? So,   our next step is 
to consider the evolution of functions inside a binary description (no generality lost). In this case, we are 
investigating the bounds of evolution in the space of formulae, which are 
filled in pleitropy because the same variables operate over possibly many clauses to build together a function. Now, 
formulae might represent mathematical functions that express both possibilities and restrictions for design. So,   
 evolutionary implications are expected. Anyway, to sustain our claims we always need  <b>systems  that use and 
reuse the same set of interacting units to build great functional projects </b>. Biochemistry offers us the right 
example:</p>


<p>The epitome of function in biochemistry is given by   enzymes. Pleitropy arises as follows: all enzymes have the same 
recurrent set of building blocks that are called amino-acids, around 20. Enzymes are encoded by DNA so, they are 
linear structures of, say, 100 amino-acids long. Thanks to molecular interactions with water and among amino-acids 
themselves, the linear polypeptides bend into spatial structures forming, say, globules. The enzymatic activity depends 
on this 3D structure and it is sensible to any tiny change. In fact, that instability allows the immense diversity of 
catalyzed reactions by enzymes, their highly selective power and also the control of their 
activity by ligands that may interact with the globule through, say, van der Waals forces
 (<a href="https://www.youtube.com/watch?v=qBRFIMcxZNM"target="_blank">RCSBProteinDataBank, 2013</a>;  
<a href="https://www.youtube.com/watch?v=lNh0Km4bv18"target="_blank">Freer \cite{Freer11} 2011</a>). </p>








<p>So, do enzymes have the imprinting of pleitropy? 
Yes, it is called  <b>monomorphism </b> and means that a given enzyme has no variability. Additionally, many cases are 
known of 
 <b>conserved sequences </b> that are the very same through various species 
 (<a href="http://ucbase.unimore.it/"target="_blank">UCBase, 2013</a>). 
But evolution is powerful enough to deeply surprise us:  not all enzymes are monomorphic  
(<a href="http://www.blackwellpublishing.com/ridley/tutorials/Molecular_evolution_and_neutral_theory10.asp"target="_blank">Ridley,  2004</a>). 
Moreover, it seems that observed variants in  <b>polymorphic loci </b> have 
almost the same functional performance. This was first observed by Motto Kimura  in 1968, who was able to use this 
finding to formulate  and calculate  his <i>Neutral Theory</i> according to which <i>molecular evolution is 
mostly restricted to equivalent replacements over the background of sampling effects</i> 
(<a href="http://biodiv.sinica.edu.tw/~ceegl/Allen%20Chen%27s%20Coral%20group/Chinese/class%20file/MMPE291205.pdf"target="_blank">Allen Chen's Coral group,  2000?</a>).</p> 



<p>Our words seem to imply that our war against pleitropy is a priori lost. That is not the case. Rather, we propose the 
following slogan:    <b>to use evolution in design one must invent important problems that are one step off the 
solution </b>. This is actually possible and directed evolution is a fundamental method of modern  enzyme design 
(<a href="https://www.nature.com/nrg/journal/v16/n7/pdf/nrg3927.pdf"target="_blank">Packer et al, 2015</a>).  
Now, what is the meaning of our slogan in this specific situation? It means to slightly 
deform an enzyme to catalyze a slightly different reaction that does not happen in nature.</p>




<h4>6.4 GP (Genetic Programming): bugs  over bugs.</h4>




<p>DNA conveys various types of information. One of them consists in verbal receipts to build polypeptides, say, proteins. 
If we say algorithms instead of verbal receipts, we are saying that  <b>the  genome is software </b>. If we add that 
GAs 
can be 
implemented over DNA and that nature always have being playing with them to finely tune a species to its environment 
and to 
synthesize antibodies, we are saying that   <b>evolution is a natural software developer </b>. The Evolutionary Theory 
now 
reads: 
<i>evolution is the software developer that is responsible for the software that enlivens all living beings</i>.   
So, 
this 
theory   <b>predicts </b> that  <b>evolution shall be useful to  develop software. This prediction is 
true </b>. In 
fact, all human developers run in their minds a sort of evolutionary algorithm to develop software: from a possible 
humble beginning more functional complexity is added, tested, corrected  and  so on. 
The automation of this program is 
known as GP (Genetic Programming) which has produced fruits since long ago (<a href="http://www.genetic-programming.org/"target="_blank">Koza John, 2007</a>). </p>







<p>In its turn, Java 
has provided a solution that solves the problem once and for all: it  comes with a method to convert a string   into 
Java code that can be compiled and  run. This is shown by 
Program R9 <span style="background-color: yellow">RuntimeCompilerExample</span> of Vol XVII version 2. If you can do this, all you need   to make GP is to evolve the 
encoding string to fit into your needs. Java provides various possibilities for GP including Clojure that is a language 
of the Java clan and that seems to be very appropriate for GP. So, we give for  verified the mandatory prediction of 
the 
Evolutionary Theory  that the evolutionary algorithm shall be useful to  develop software.</p>


<p>On the other hand,  <b>the huge variability, extreme complexity and exceeding perfection of living forms demand 
from   GP to produce high quality software of whatever complexity and for whatever function </b>. This is a mandatory 
prediction. Thanks to Java it is testable. And what did happen? After too much work to get started,  <b>one cannot 
see anything like that </b>. Instead the situation is similar to that found in the Antarctic: amidst the most severe 
cool, 
there are various research stations where you can find shelter, food, company, instrumentation and hard work. Very hard 
work. Likewise, we can enjoy today a relatively easy world in Java GP due to the efforts of many pioneers. So, GP is 
for people that love harsh adventures and that  have too  much knowledge and wisdom to fabricate tiny but important 
projects worth a PhD dissertation. Anyway, be ready to use supercomputing. In this regard, we can now assert that if 
the Evolutionary Theory is true then natural evolution has supercomputing powers 
(<a href="https://www.nature.com/news/supercomputer-redesign-of-aeroplane-wing-mirrors-bird-anatomy-1.22759"target="_blank">Gibney,  2017</a>). 
To be honest, its powers must be super-super computing.  If someone thinks like this, let him 
or her  play with <span style="background-color: yellow">Program Q100 BinaryAdder3LH3</span> of Vol XVI that faces to  a very simple problem: can evolution design 
a binary adder using evolutionary native operations?</p> 
 





  
<p>Are you thinking that GP is too esoteric? Not really: since you use evolution to develop software,  <b>your programs
must be   evolvable </b>, i.e., used style must facilitate evolution to higher functionality. Now, one can classify own 
style very easily by answering a simple question: can you understand your programs instantaneously when you come on
Mondays to continue work? How difficult is to   get close to this! But we have no other choice. Thus, make up your mind 
and be brave because you are not alone! Many people use to struggle with evolvability and have shared their experiences 
so, one have a lot to learn from them (<a href="https://dzone.com/articles/clean-code-principles"target="_blank">Biel,   2017</a>).</p>
 
 
<p> <b>Now, GP is a terrible, clear cut  and immediate falsification of the Evolutionary Theory </b>. In fact, software 
is built over use and reuse of the very same interacting elements to conform a great functional unit. So, 
slightly complex software must be pervaded in pleitropy. Explicitly:</p>

<p>The closest reality to every developer is that  <b>there is no software without bugs whose correction creates more 
bugs </b>. Since GP is less efficient than human developers, it must produce still more bugs. This is actually so: a 
bug is just a proposal that does not match the target. And GP burns bugs might be for hours, days, and months   to 
solve extremely simple problems.  Now, nature runs GP on the DNA of living beings. This GP to be responsible for the 
existence of species must be accompanied with tons of bugs in the form of malfunctions and malformations as in the 
fossil record as in extant populations as in our own bodies. Nothing like that is found. End.</p>





<h3>7. CONCLUSION</h3>


<p>The excel quality of life is built over items that fulfill many requirements and whose parts interact to conform great functional units. We cannot exclude the possibility that evolution could make some great deeds in its battle against the arising complexity. But we can retro-predict with absolute certainty that if evolution were the cause of our existence, then the tracks of the ensuing evolutionary activity would be abundantly seen as in the fossil record as in extant populations as in our own bodies in the form of thousands of malformations and malfunctions. The reason is that the predicted type of evolution towards perfection, if any, should have gone in leaps with interspaced long periods of ecstasy. Nothing like that is seen. This is a clear-cut falsification of the Evolutionary Theory. But on the other hand, evolution is an ordinary tool to solve problems. The geniality, as shown by the immune system, is to 
invent clever problems one step off the solution and give them to evolution.</p> 


 
 

</body>
</html>