Thursday, March 10, 2016

Response to Evolution "News"

I don't know why I wasted my time. My edits are in bold italics. Response to 

http://www.evolutionnews.org/2015/09/waiting_for_mut099631.html


Many (hardly any) scientists now recognize the insufficiency of the classic Darwinian story to account for the appearance of new features or innovations in the history of life. They focus on other theories to account for remarkable differences between genomes, the appearance of novel body plans, and genuine innovations like the bat's wing, the mammalian placenta, the vertebrate eye, or insect flight, for example (In the same way that General Relativity improved on "Newtonian Theory" and made Mercury's orbit work?). They realize (Inspired by their religion, they believe) that the traditional story of population genetics (changes in allele frequencies in populations due to mutation, selection, and drift) cannot account for "the arrival of the fittest" and not just the "survival of the fittest." (Hugo DeVries, 1904). <--  Wow, they're reaching DEEP into the historical academic work
One of the reasons many (this tiny fraction of )scientists acknowledge the insufficiency of Darwinism is because they know the refuse to admit that the accounting won't works. The mutation rate, the generation times, the strength of selection versus genetic drift, the population sizes, and the time available don't match up just fine.
Here's where they deliberately misinterpret journal results and compare apples to oranges to make the reader think there's something wrong with science. At its core, this is part of a larger argument from ignorance approach, paraphrased as: 'Science isn't right. Therefore God did it.'
For example, supposedly humans last shared common ancestry with chimps about six million years ago. Since that time, we have accumulated significant differences with chimps -- genetic, anatomical, physiological, behavioral, and intellectual differences, among others. The genetic differences between humans and chimps are much more than the (shrinking) 1.2 percent difference in base pairs that is so often quoted in the media. (Yes, there are different ways of measuring similarity.) Add small insertions and deletions and the differences climb to about 3-5 percent,depending on whose estimate is used. Add another 2.7 percent for large scale duplications or deletions, another 6 percent for new Alu elements (a kind of mobile genetic element) and some unknown number for rearrangements of the DNA, other insertions of mobile genetic elements, or new genes, we have more than 11.7 percent of our genome with unique features not present in chimps. (Only when you struggle to exaggerate moves and deletions. For reference, here is the preliminary analysis as published in Nature.) 
There is only so much time for these differences to have accumulated. Mutations arise and are propagated from generation to generation, so the number of generations limits how many mutations can accumulate. The estimated mutation rate is about 10-8 per base pairs per generation, and we have an average generation time of somewhere between 10 and 25 years. Our estimated breeding population size six million years ago is thought to have been about 10,000 (these are all rough estimates based on numbers currently in use -- see the papers cited below). Based on these numbers, one can estimate how many years it would take to acquire all those mutations, assuming every mutation that occurred was saved, and stored up. (Since the mutation rate is extremely small and uniformly distributed, this assumption is valid.)
But there's a difficulty -- it's called genetic drift. In small populations, like the 10,000 estimate above, mutations are likely to be lost and have to reoccur many times before they actually stick (WTF? It's not like a specific change keeps trying. Many are just lost). Just because of random effects (failure to reproduce due to accidental death, infertility, not finding a mate, or the death of all one's progeny), a particular neutral mutation may have to arise many times before it becomes established in the population, and then many more years before it finally becomes fixed (that is, before it takes over the population and replaces all other versions).
What a strange misinterpretation of the process! The author seems to think that the mutations which fail to reproduce keep trying until they stick. Bizarre! Clearly, lost mutations are simply lost. Just like the 99.9% of all species which have gone extinct. Very few succeed.
How long before a single, new mutation appears and becomes fixed? An estimate from a recent paper (by a Young Earth Creationistusing numerical simulations is 1.5 million years. That is within the range of possibility. (By assuming that there is only a single specific desirable mutation being considered by nature at any given time.) But what if two specific mutations are needed to effect a beneficial change? Their estimate is 84 million years. Other scientists have done this calculation using analytical methods, but their numbers are even worse. One report calculates 6 million years for one specific base change in an eight base target typical of the size of a DNA binding site to fix (In the abstract, the author points out that the sequence needn't be perfect to be valuable, reducing the time to 60,000 yrs), and 100 million years to get two specific mutations. (That work was later amended to 216 million years.) Extrapolating from other published data merely confirms the problem.
Astonishing dishonesty! A brief review of the linked article (100 million years) contains the following quote: "Fortunately, in biological reality, the match of a regulatory protein to the target sequence does not have to be exact for binding to occur. Biological reality is complicated, with the acceptable sequences for binding described by position weight matrices that indicate the flexibility at different points in the sequence. To simplify, we assume that binding will occur to any eight-letter word that has seven letters in common with the target word. If we do this, then the mean waiting time reduces to ∼60,000 years."  Why would Evolution "News" need to misrepresent the author's conclusions in their linked article? Did they think nobody would read it?

Durrett makes a good analogy to the mistake Michael Behe made to reach is 216 million year figure in the linked article: "Behe is not alone in making this type of mistake. When Evelyn Adams won the New Jersey lottery on October 23, 1985, and again on February 13, 1986, newspapers quoted odds of 17.1 trillion to 1. That assumes that the winning person and the two lottery dates are specified in advance, but at any point in time there is a population of individuals who have won the lottery and have a chance to win again, and there are many possible pairs of dates on which this event can happen. The probability that it happens in one lottery 1 year is ∼1 in 200 (Durrett 2009)  
Another paper came up with much shorter time frames by assuming that any 5 to 10 base pair binding site could arise anywhere within 1 Kb of any promoter within the genome. 
Yet in all likelihood many more than two binding sites would be required to change anything significant, and those binding sites must be appropriate in location and in sequence to accomplish the necessary changes. They must work together in order for a specific adaptive change to happen.  (Again, the authors seem to think that these changes are prescribed and can't be happening at random, in parallel, and at different sites.) 
Genes operate in networks, and to shift a gene regulatory network would require many mutations, and not just random ones. Remember there are anatomical physiological, behavioral, and intellectual differences to explain, multiple traits each requiring multiple coordinated mutations. Unless one invokes luck on a large scale, those traits would not have come to be. (Again, all this presumes humanity is the intended end-state. It is because it is.) 
I'm not betting on luck. (Of course you're not). 

Image: Homo georgicus, reconstruction, photo by 120 (Own work (photograph), model by √Člisabeth Daynes) [CC BY-SA 3.0], via Wikimedia Commons.

Well that was as much fun as beating my head against a wall. The basics mistakes are as follows:

  1. Cite ancient journals from 1904 as if they're still relevant
  2. Misrepresent the number of changes
  3. Cite letters to journals which explicitly refute the point you're claiming they make (oops?!)
  4. Presume that the genetic changes were prescribed
  5. Assume intelligent humanity was the universe's desired end-goal
  6. Ignore that a neutral genetic change will not survive without a second change that's necessary to the benefit. Indeed, we could have millions of such single-mutations waiting on a second mutation to confer a benefit.

Sunday, March 6, 2016

Standard Phases of a Debate with a Theist

I've been debating with fundamentalist theists for a while now, and I've started to notice some interesting trends. In this blog, I'll describe the standard tactics theists use when attempting to squash dissent. I believe many of these illuminate pervasive beliefs among the more conservative believers. Many of them strike me as a defense mechanism or a childish black-and-white view where truth is established by authority rather than evidence.

Denial

This tact seems to me like the believer needs comfort that they couldn't become atheists. Every time I see it, it seems to me like the theist is desperately trying to hold on to a belief that's impossible to validate.  All these denialist tactics will come off as an accusation that you're lying. Try to remember that the theist just wants to feel better about the security of their own fragile faith.

You're not actually atheist

It's insulting when a someone accuses you of lying -- especially when they don't even know you.  Try not to take this tact personally. There are several factors at play here. First, the Christian Bible clearly states that all men know god is real. The most thoroughly indoctrinated fundamentalist believers are taught to accept the claims in that book regardless of the world they observe. Since the book must be true, you must be lying.
  • "You're just angry with God, you know God is real. You just want to sin without being punished."
  • "You know god is real. It says so in Romans 1:20."
    • Nope. I really don't believe it's real.
  • "You can't unknow something after you've known it"
    • But some of us can realize and admit that we were mistaken
  • "You just want to sin so you pretend God isn't real. But he's gonna getcha' in the end." 
    • OK. I added that last sentence. But the assertion is ridiculous. It's like saying I can rob a bank if I close my eyes and pretend the police aren't there. It's also a claim to know my mind and an accusation that I'm lying. This is treading awfully close to the block button. I don't take kindly to being accused of dishonesty by someone who doesn't know me.

It Can't Happen to Me!

  • "You were never really Christian if you stopped believing. 1 John 2:18-21 and Luke 8:13 predicted this"
    • As if people weren't leaving religions back then too.  It didn't take a genius to make that prophecy. Just look around.

Finding Bearings Without Absolutes

It's pretty disorienting to let go of a core belief. I once thought all my moral values were based in Christian doctrine.  When the Christian doctrine started to crumble, I was left wondering how I would make ethical choices in my life. Turns out this wasn't very hard, but before I thought it through it felt terrifying.

Morality

  • "For morality to be absolute, there must be an ultimate lawgiver."
    • Neither true nor a valid reason for an 'ultimate lawgiver' to exist. 
    • This whole line of questioning seems to imply that we are either guaranteed absolute morality or that "absolute morality" is somehow inherently obvious. 
  • "Where do morals come from in a godless universe?"
    • This question is obvious to someone who understands that genes are common within a tribe, so supporting the tribe reinforces the reproduction of genes. Successful tribes are those where the individuals within the tribe demonstrate  empathy for other tribe members.
    • Unfortunately, many of the Christians I talk with haven't got the first clue how evolution actually works, so short of a repeat of high school biology class, they're just not going to understand.
  • "Mere humans are not capable of understanding, but it's moral for God to XXX (insert horrific Bible story here)." This is the usual response to the assertion that a bible story is immoral. 

Good and Evil

  • "Without a reference of absolute good, there's no way to judge anything as good or evil"
  • "Evil must exist so that we can recognize good" (response to the problem of evil)

Science Is Unreliable

Many (not all) atheists accept that the scientific method is the most reliable method for knowing and understanding our world. Some theists feel a need to try to tear down science in order to feel like faith is somehow reliable or at the very least, just as good as science.
  • "Science isn't capable of detecting God"
    • Science is a process for measuring anything that's objective, verifiable, and logical.  Which of these is your god unable to satisfy?
  • 'Scientific "facts" are always changing '
    • A distortion of the truth. Interpretation of facts can change.  Repeatable objective measurements do not change.
  • "It takes Faith to do science!" Followed by an attempt to label imagination and foresight as the same as religious faith. 
    • This is equivocation, a sign of poor arguments. Imagination and creativity are an ability to envision that something might work or might be true.  Faith is concluding that a belief must be true.
  • "Great scientists believed in God". this is typically followed with specific examples of famous scientists who believed in God in ancient times. 
    • It's worth remembering that people didn't have a choice in what they professed to believe back then. The punishment for apostasy is death in the Bible. Thank God we've moved past that!

Proselytizing

Sadly, many theists aren't really in it for the discussion. They are used to a preaching style of communication and they think it'll help you find Jesus. 

My Version of Christianity Will Fix You

  • "How many times have you read the Bible cover-to-cover"
    • This one is both a dick-measuring contest and a challenge to believe their special religion.
  • "You need to really READ the Bible."
    • Many of the atheists I know are former Bible scholars.

Quoting Scripture When Things Get Rough

This is like some sort of incantation.  The specific verses vary, but it's like they're trying to reinforce their own beliefs and ward of Satan when you say something that makes them doubt. It's rather amusing when it happens.  I tend to infer that I've likely struck a nerve.

Thinly veiled hell threats

I think these are a good opportunity to ask the believer if they think it's ethical for their God to torture people based on sincere beliefs. It's often a last-resort tactic when the believer needs to make themselves feel better about the fact that they're completely unable to justify their stated beliefs.

  • Every knee will bow
  • Some day, you'll be sorry for what you're saying

Begging the Question

When it's clear they can't possibly demonstrate any of their claims to be true, they'll start working their baseless assumptions as presuppositions to comments and questions. 
It's like they're incapable of even recognizing the places where they've made assumptions. Repeatedly asserting as if it's obvious doesn't make it real.

Genetic Engineering Algorithms

Why do I care?

In discussion with creationists, I'm often told that "randomness cannot design." 

That's utter bullshit. One of the ways I reply to these tweets is with examples of Genetic Design Algorithms, where random mutations coupled with selection and reproduction based on objective performance criteria can will always result in the appearance of design that satisfy the performance criteria.
In the course of these conversations, I've dug up quite a few resources and developed a few memes.

Evolution Works

It works so damn well that it can be used to design things when you're not sure what approach is best. I'm not just saying that, there's really good examples:
  • Field -Programmable Gate Arrays (FPGA's) have been designed to distinguish sounds
  • There are commercial software products like Genetic Designer and Natural Selection which provide tools for genetic design
  • Open-source libraries are free and available, so you can study the techniques and evaluate the performance for yourself. Admittedly, most of the people don't have the technical expertise, time, or patience to learn how to use these tools.

Watch it Happen Yourself

There are some really great examples of genetic design algorithms which can be watched evolving in near-real-time. One of my favorites is called BoxCar2D. In it, you can watch randomly designed cars compete for speed and distance.  Successful cars reproduce with random mutations. You can even manipulate the mutation rates and number of top cars to breed. BoxCar2D is a pretty robust simulation,
There are other, simpler online examples of the same basic concept. Since they're simpler, they converge faster:

  • Genetic Algorithm 2D is a version that runs on pure HTML5 / javascript, so you don't need Flash.
  • Genetic  Cars lets you cross-breed your cars with cars from other online players.
  • Darwin Bots is a more comprehensive simulation that runs on your desktop
  • BioMorph is a simulator where the user interactively selects the successful child in each generation.
I'm sure there are more. If you have a personal favorite, please post a comment or tweet me.