Further on ENCODE (and SETI, and Biologos etc.)

I've been meaning to gin up some further contemplations on the ENCODE results, as a followup to my previous article, and in reply to my friend the thoughtful and sober atheistic commenter "Blue Devil Knight" who commented on my article starting here in the comments. But I've been distracted by other projects.

BDK asks,

"What is it specifically that the naturalists are having so much trouble handling?"

If you're referring to my opening paragraph, I meant that "proponents of completely non-intentional biological development" were "scrambling defensively in various ways" in relation to "ID and creationist proponents hopping up and down"--which combined with how I closed that sentence ("...and frankly, at first I didn't see what the big deal was."), plus my next large paragraph, means I didn't understand why the ID and creationist proponents were hopping up and down. But such activity would naturally lead to scrambles from the opposition, even if the scramblers were the ones in the right. After all, people on my side of the aisle (myself included {g}) tend to scramble to meet even goofy non-theistic arguments when they're put forward in a culturally attention-getting way.

(I was specific about "proponents of completely non-intentional biological development" because not all such people are atheists and/or philosophical naturalists. Biologos.org, for example, is a site run by biologists and other scientists who hold to God having set up a system where random copy-errors or other undirected mutations build up effective biological complexity, up to and including inter-and-intracellular biological machinery, which natural selection processes don't kill off or otherwise destroy as quickly as other similarly generated variants. Consequently, they're currently defending the notion of non-functional DNA still predominating in cells with the reasonable observation that chemical interactions along the chain do not mean functional proteins and other enzymes are being generated, and that prior lines of evidence relating to non-functional DNA have not suddenly disappeared.)

So I didn't mean "scramble" there against naturalists. And despite the arguments in the rest of my article, I did qualify the results rather strongly, too, such as by my reminder at the end that the situation (for all we know) may flip around sooner or later. My title itself poked some self-critical fun at my propensity, and theist observers of biology more generally, to run with the new data!--or do you think if I was 100% gung-ho behind the results I would have questioned in my title whether we ought to start a "marvelous orgy" now? {wry g}

Still, there's scrambling and there's scrambling. The ENCODE researchers themselves, despite knowing that recent (but only relatively recent) research has been steadily undermining the concept of "junk DNA", routinely use words like "surprising" or "unprecedented" and "a lot more than expected" when reporting their findings. In an article for Science a few weeks ago, Elizabeth Pennisi quoted John Stamatoyannopoulos (one of the ENCODE researchers at the University of Washington, and definitely a proponent of neo-Darwinian gradualism based on other public comments) as saying "I don't think anyone would have anticipated even close to the amount of sequence that ENCODE has uncovered that looks like it has functional importance."

If intelligent design is true, the results shouldn't be that surprising at all. (ID proponents have been predicting such results since the 90s.) Random copy error noise from mutation as a basis for building up genetic structure would be more likely to result in significantly large amounts (even supermajority large amounts) of random copy error noise as the genetic structure. ID and non-ID theories can both work with meaningful and unmeaningful genetic sections, as I noted in the article; but ID can work a lot more with a little meaningful information in a primary background of meaningless noise, than non-ID theories can work with a primary background of meaningful information even when some gibberish is included here and there.

(The problem with non-intentionally directed natural selection processes weeding out the gibberish is not just that most "selection" processes would naturally tend to accidentally weed out the useful accidental copy-errors, too--just not as quickly on larger timescales--but that helpful functionality of a copy-error presupposes a non-gibberish situation already in place for the mutations to be helpful in. Natural selection builds up nothing; it's a purely negative process.)

If someone found what looked like operational code in stellar background noise--code for generating prime numbers or which was otherwise applicable to specified operation--that would count strongly toward intentional design of that portion of the radio sequence. SETI researchers don't have to prove that the vast majority of stellar radio signals are operational code in order to detect extra-terrestrial intelligence. But if they proved that more than half (and much much more than half!) of cosmic radio background signals were operational code, that result would certainly be "surprising" and "unprecedented" and "a lot more than expected" for everyone--except for people who think we're surrounded by alien societies in every direction routinely filling the spaces between the world with beamed information (just like Earth does going out).

For this reason I think we're going to see a push for checking whether the functionality is virus-like self-replication, i.e. no real function for the cell or the multi-cellular body where appropriate (since all cells of a body have the same DNA but not all cells have the same functions). My guess that the ENCODE results eliminate that option, as such mere self-replication functionality would only apply during cell division, and the tests for functionality involved protein generations other than cell division. But aside from there being a legitimately scientific interest in ascertaining what the functionality means (since the ENCODE tests didn't and couldn't yet test for verification on what the transcriptions and histone modifications were actually accomplishing), there's going to be an ideological push, too--and to be fair, ideological pushes on all sides of the aisle.

That being said, if the ENCODE test designers had enough sense to eliminate cell division transcriptions, then since the chemical reactions involved in unwinding, transcribing, and rewinding portions of DNA are far from incidentally accidental (it isn't like random chemicals meandering along bump into the outside of a DNA strand until enough happen to cohere together to form a protein after which they randomly float off again), the freakishly detailed "upstream" process functions would be strong presumptive evidence in favor of downstream functionality. Even on non-ID cell development theories, the cell would have to be already set up to survive and provide sufficient functionality despite wasting so much energy and material, which in itself would be more than a little weird.

Biologos, for example, published an article here about how the ENCODE results would identify the Mendel's classic pea alleles of "dominant" purple gene and "recessive" gene as equally functional. But the "recessive" white gene is sufficiently functional for purposes of an ID interpretation of the ENCODE results: it has a real functional difference, due to a difference in the DNA sequence which leads to a premature breakdown of the resultant protein coding for color, leading to a whiter instead of purple flower. But the upstream biochemical processes leading to a generation of mitachondrial RNA for a white pea flower, and that RNA has the tags necessary to leave the area and go into the protein sequencing machine, and the sequencing machine can synthesize a preliminary protein chain out of it that then goes into the folding machine, which folds a protein out of it, and that protein has enough functionality to get through the other machines in the cell to serve a purpose somewhere else, most of them being the same purposes served by a purple color protein. The current (and presumably correct) interpretation of the difference between the white and the purple coding protein, as Dr. Venema (author of the article) mentions, is that there has been a marginally harmful mutation, presumably from copy-error, that resulted in a bit of the information in the coding being gibberished.

Now that I read over the article again, Dr. Venema actually indicates that the mutation renders the protein "too short to work as an enzyme" and too instable so that the mRNA "degrades more readily, resulting in a lower steady-state amount of the mRNA in the cell". But whether the mRNA codes for a nearly-as-functional protein or for one that is so broken it does nothing, the salient point is that the protein is broken.

That still fits the paradigm-changing shift being lauded by ID proponents. The background expectation now changes to a formulation, the "transcript", that presupposes some kind of useful information. As I put it in my previous article: if the information turns out not to be useful, then either we aren't looking at it correctly yet or we should regard the sequence as being broken. Not originally random noise accidentally generated to begin with. Broken information (whether broken accidentally or not).

Dr. Venema warns that "[t]hese limitations should stand as a caution to any group that wishes to adopt the ENCODE definition as the only viable definition of biological function", but ID proponents obviously go very far identifying multi-valent levels of functionality. He suspects that "many of those opposed to evolution" (by which he means common "Young Earth Creationist, Old Earth Creationist, and Intelligent Design") would bristle at the suggestion that the white color allele was equally functional to the purple allele since the white allele (being the recessive gene which doesn't express, or doesn't express as much, when a dominant gene for purple is also present) "represents a clear loss of function in keeping with [those groups'] definitions of loss-of-function alleles, and the propensity of these groups to insist that such mutations destroy functional information." But Dr. Venema's own chosen example involves a mutation destroying loss of original function by mutation, which fits perfectly with such groups' expectations that the utterly complex and multi-valent functioning of microbiological processes could not have been feasibly built up by random mutations. In this case, it is Dr. Venema who "has not yet had time to carefully think through the implications" of embracing the ENCODE results.

Anyway, it isn't that we simply don't know what most of the chemical behavior does after generation; we know a lot (and an increasingly larger amount) of what the chemical behavior has to do to generate those transcripts; and that behavior is certainly not functionally insignificant. We've also previously learned that some of the functionality of what appeared to be non-coding portions of the genome isn't biochemical, but involves structural or other functionality, some of which is absolutely essential to cell survival and operation. In that sense ENCODE uses too narrow a definition of function! But junk DNA proponents are claiming the definition is too broad.

At the very least, if the ENCODE results hold, then we're going to see a brief but intense fight between two different classes of non-ID proponents: those who have been pointing to junk DNA (or the equivalent thereof) as strong evidence for a history of merely random mutations (which of course it would be, although that wouldn't eliminate design in other ways), thus as strong evidence in favor of a totally non-ID development process; and those who have been stressing the natural selection side of neo-Darwinian gradualism and so saying on that ground that we ought to expect majority or even supermajority levels of functionality because natural selection processes would have weeded out the non-efficient portions of DNA. (But as noted above there are significant problems with this, which the scientists who have been leaning hard on neutralistic junk DNA are aware of.)

Richard Dawkins has already flipflopped completely on this; since 1976 in The Selfish Gene, up to three years ago in 2009's The Greatest Show On Earth, he couldn't hammer junk DNA hard enough as evidence against any kind of intelligent design and in favor of being exactly what neo-Dargrad theory would expect. Last month he said in a public debate with rabbi Lord Jonathan Sacks that superfunctional DNA is exactly what a Darwinist would hope for and expect.

Granted, Mr. D is more than a bit of a straw man (which is why he provides a richness of embarrassments for us to mine and feel good about ;) ). But he's also a good barometer of how this will be popularly played out. A majority and even a supermajority of junk DNA is what neo-Darwinistic gradualism would expect and what a neo-Dargrad proponent would hope to find; but a majority and even a supermajority of functional DNA is also what neo-Darwinistic gradualism would expect and what a neo-Dargrad proponent would hope to find. The only problem is that each set of expectations and hopes are based on very conflicting reasons rooted strongly in neo-Darwinian gradualism as a theory.

It's one thing to say that the theory is genuinely flexible enough that wildly different evidential states can fit either way; it's another thing to say that the theory predicts two wildly different evidential states over against itself for conflicting reasons. Either way, though, that means the different evidential states cannot be used as evidence for the superiority of the theory against other theories (although an evidential state in itself could still be used as evidence against another theory).

(Dr. Venema has a second part to his recent essay on interpreting the ENCODE results, over at Biologos.org, which also inadvertently illustrates what I was talking about the in the original article, about what the implications for ENCODE are in regard to ID theories; but I'll leave that as an exercise for the readers until/unless I get around to commenting on it further.)



Jason Pratt said…
Registering for comment tracking.
Thanks for clarifying. The final blog post I linked to has a very good discussion by some very competent scientists on this (it does seem some of the authors in the ENCODE studies have overstated things, which is what scientists tend to do in press releases and statements to the press, which is why I get my science from the original articles whenever I can).

But as I stated, I have always found the way Dawkins and others pounced on junk DNA as revealing their confirmation bias as much as facts about biology.

I guess I'm just pretty underwhelmed here, and see this as a fairly typical ID overreaction to your garden-variety naturalistic biology marching along just fine.
In case it got lost this is a very good discussion by people who are very good molecular biologists.

I highly recommend that post to anyone who is interested in this debate.

On your previous commments at last post...

You are focusing a lot on 'random copy errors' but point mutations are just one of many mechanisms of producing genetic variation. Recombination, transposable elements (jumping genes), gene duplication, are important too. Careful you don't want to end up sounding like Gish talking about tornadoes making 747s.

If you just focus on the production of variation, and ignore the question of how these variants change in prevalence over time (i.e., evolution!), then I can see how you might be tempted to take talk of 'biology producing complexity' as implying teleological pixie dust, rather than shorthand for a naturalistic story.
Useful clarification here by an ENCODE author on the misrepresentation of the work in the popular press.

Note also that ENCODE looked only at humans, so extrapolating to all of biology is weird, and especially extrapolating backwards in time to the first living things (and prebiotic ancestors) would be even less justified.
tahir sumar said…
This comment has been removed by a blog administrator.

Popular posts from this blog

Martin Luther King, Jr., Jesus, Jonah and U2’s Pride in the Name of Love

How Many Children in Bethlehem Did Herod Kill?

How Should I Be A Sceptic -- belief and reason

Kierkegaard's Knights of Faith and the Account of Abraham

Bayes Theorem And Probability of God: No Dice!

The Meaning of the Manger

If Christianity were true, would you become a Christian?

The Origin of Life and the Fallacy of Composition

Where did Jesus say "It is better to give than receive?"