Discovery of Intron Function Interrupts Evolutionary Paradigm

Untitled 7

Nobody likes to be interrupted when they are talking. It feels disrespectful and can be frustrating. Interruptions derail the flow of a conversation.

The editors tell me that I need to interrupt this lead to provide a “tease” for what is to come. So, here goes: Interruptions happen in biochemical systems, too. Life scientists long thought that these interruptions disrupted the flow of biochemical information. But, it turns out these interruptions serve an important function, offering a rejoinder a common argument against intelligent design.

Now back to the lead.

Perhaps it is no surprise that some psychologists study interruptions1 with the hope of discovering answers to questions such as:

  • Why do people interrupt?
  • Who is most likely to interrupt?
  • Do we all perceive interruptions in the same way?

While there is still much to learn about the science of interruptions, psychologists have discovered that men interrupt more often than women. Ironically, men often view women who interrupt as ruder and less intelligent than men who interrupt during conversations.

Researchers have also found that a person’s cultural background influences the likelihood that he or she will interrupt during a discourse. Personality also plays a role. Some people are more sensitive to pauses in conversation and, therefore, find themselves interrupting more often than those who are less uncomfortable with periods of silence.

Psychologists have learned that not all interruptions are the same. Some people interrupt because they want the “floor.” These people are called intrusive interrupters. Cooperativeinterrupters help move the conversation along by agreeing with the speaker and finishing the speaker’s thoughts.

Interruptions are not confined to conversations. They are a part of life, including the biochemical operations that take place inside the cell.

In fact, biochemists have discovered that the information harbored in genes, which contains the instructions to build proteins—the workhorse molecules of the cell—experience interruptions in their coding sequences. These intrusive interruptions would disrupt the flow of information in the cell during the process of protein synthesis if the interrupting sequences weren’t removed by the cell’s machinery.

Molecular biologists have long viewed these genetic “interruptions” (called introns) as serving no useful purpose for the cell, with introns comprising a portion of the junk DNA found in the genomes of eukaryotic organisms. But it turns out that introns—like cooperative interruptions during a conversation—serve a useful purpose, according to the recent work of two independent teams of molecular biologists.

Introns Are Abundant

Noncoding regions within genes, introns consist of DNA sequences that interrupt the coding regions (called exons) of a gene. Introns are pervasive in genomes of eukaryotic organisms. For example, 90 percent of genes in mammals consists of introns, with an average of 8 per gene.

After the information stored in a gene is copied into messenger RNA, the intron sequences are excised, and the exons spliced together by a protein-RNA complex known as a spliceosome.


Figure 1: Drawing of pre-mRNA to mRNA. Image credit: Wikipedia

Molecular biologists have long wondered why eukaryotic genes would be riddled with introns. Introns seemingly make the structure and expression of eukaryotic genes unnecessarily complicated. What possible purpose could introns serve? Researchers also thought that once the introns were spliced out of the messenger RNA sequences, they were discarded as genetic debris.

Introns Serve a Functional Purpose

But recent work by two independent research teams from Sherbrooke University in Quebec, Canada, and MIT, respectively, indicates that molecular biologists have been wrong about introns. They have learned that once spliced from messenger RNA, these fragments play a role in helping cells respond to stress.

Both research teams studied baker’s yeast. One advantage of using yeast as a model organism relates to the relatively small number of introns (295) in its genome.


Figure 2: A depiction of baker’s yeast. Image credit: Shutterstock

Taking advantage of the limited number of introns in baker’s yeast, the team from Sherbrooke University created hundreds of yeast strains—each one missing just one of its introns. When grown under normal conditions with a ready supply of available nutrients, the strains missing a single intron grew normally—suggesting that introns aren’t of much importance. But when the researchers grew the yeast cells under conditions of food scarcity, the yeast with the deleted introns frequently died.2

The MIT team observed something similar. They noticed that during the stationary phase of growth (when nutrients become depleted, slowing down growth), introns spliced from RNA accumulated in the growth medium. The researchers deleted the specific introns that they found in the growth medium from the baker’s yeast genome and discovered that the resulting yeast strains struggled to survive under nutrient-poor conditions.3

At this point, it isn’t clear how introns help cells respond to stress caused by a lack of nutrients, but they have some clues. The Sherbrooke University team thinks that the spliced-out introns play a role in repressing the production of proteins that help form ribosomes. These biochemical machines manufacture proteins. Because protein synthesis requires building block materials and energy, during periods when nutrients are scarce, protein production slows down in cells. Ratcheting down protein synthesis impedes cell growth but affords them a better chance to survive a lack of nutrients. One way cells can achieve this objective is to stop making ribosomes.

The MIT team thinks that some spliced-out introns interact with spliceosomes, preventing them from splicing out other introns. When this disruption happens, it slows down protein synthesis.

Both research groups believe that in times when nutrients are abundant, the spliced-out introns are broken down by the cell’s machinery. But when nutrients are scarce, that condition triggers intron accumulation.

At this juncture, it isn’t clear if the two research teams have uncovered distinct mechanisms that work collaboratively to slow down protein production, or if they are observing facets of the same mechanism. Regardless, it is evident that introns display functional utility. It’s a surprising insight that has important ramifications for our understanding of the structure and function of genomes. This insight has potential biomedical utility and theological implications, as well.

Intron Function and the Case for Creation

Scientists who view biology through the lens of the evolutionary paradigm are quick to conclude that the genomes of organisms reflect the outworking of evolutionary history. Their perspective causes them to see the features of genomes, such as introns, as little more than the remnants of an unguided evolutionary process. Within this framework, there is no reason to think that any particular DNA sequence element, including introns, harbors function. In fact, many life scientists regard the “evolutionary vestiges” in the genome as junk DNA. This clearly has been the case for introns.

Yet, a growing body of data indicates that virtually every category of so-called junk DNA displays function. We can now add introns—cooperative interrupters—to the list. And based on the data on hand, we can make a strong case that most of the sequence elements in genomes possess functional utility.

Could it be that scientists really don’t understand the biology of genomes? Or maybe we have the wrong paradigm?

It seems to me that science is in the midst of a revolution in our understanding of genome structure and function. Instead of being a wasteland of evolutionary debris, most of the genome appears to be functional. And the architecture and operations of genomes appear to be far more elegant and sophisticated than anyone ever imagined—at least within the confines of the evolutionary paradigm.

But what if the genome is viewed from a creation model framework?

The elegance and sophistication of genomes are features that are increasingly coming into scientific view. And this is precisely what I would expect if genomes were the product of a Mind—the handiwork of a Creator.

Now that is a discovery worth talking about.


  1. Teal Burrell, “The Science behind Interrupting: Gender, Nationality and Power, and the Roles They Play,” Post Magazine (March 14, 2018),; Alex Shashkevich, “Why Do People Interrupt? It Depends on Whom You’re Talking To,” The Guardian (May 18, 2018),
  2. Julie Parenteau et al., “Introns Are Mediators of Cell Response to Starvation,” Nature 565 (January 16, 2019): 612–17, doi:10.1038/s41586-018-0859-7.
  3. Jeffrey T. Morgan, Gerald R. Fink, and David P. Bartel, “Excised Linear Introns Regulate Growth in Yeast,” Nature 565 (January 16, 2019): 606–11, doi:10.1038/s41586-018-0828-1.

Reprinted with permission by the author
Original article at:

Believing Impossible Things: Convergent Origins of Functional Junk DNA Sequences



In a classic scene from Alice in Wonderland, the story’s heroine informs the White Queen, “One can’t believe impossible things,” to which, the White Queen—scolding Alice—replies, “I daresay you haven’t had much practice. When I was your age, I always did it for half-an-hour a day. Why, sometimes I’ve believed as many as six impossible things before breakfast.”

If recent work by researchers from UC Santa Cruz and the University of Rochester (New York) is to be taken as true, it would require evolutionary biologists to believe two impossible things—before, during, and after breakfast. These scientific investigators have discovered something that is hard to believe about the role SINE DNA plays in gene regulation, raising questions about the validity of the evolutionary explanation for the architecture of the human genome.1 In fact, considering the implications of this work, it would be easier to believe that the human genome was shaped by a Creator’s handiwork than by evolutionary forces.


One of the many classes of noncoding or junk DNA, short interspersed elements, or SINE DNA sequences, range in size from 100 to 300 base pairs (genetic letters). In primates, the most common SINEs are the Alu sequences. There are about 1.1 million Alu copies in the human genome (roughly 12 percent of the genome).

SINE DNA sequences (including Alu sequences) contain a DNA segment used by the cell’s machinery to produce an RNA message. This feature allows SINEs to be transcribed. Because of this feature, molecular biologists also categorize SINE DNA as a retroposon. Molecular biologists believe that SINE sequences can multiply in number within an organism’s genome through the activity of the enzyme, reverse transcriptase. Presumably, once SINE DNA becomes transcribed, reverse transcriptase converts SINE RNA back into DNA. The reconverted DNA sequence then randomly reintegrates back into the genome. It’s through this duplication and reintegration mechanism that SINE sequences proliferate as they move around, or retrotranspose, throughout the genome. To say it differently, molecular biologists believe that over time, transcription of SINE DNA and reverse transcription of SINE RNA increases the copy number of SINE sequences and randomly disperses them throughout an organism’s genome.

Molecular biologists have discovered numerous instances in which nearly identical SINE segments occur at corresponding locations in the genomes of humans, chimpanzees, and other primates. Because the duplication and movement of SINE DNA appear to be random, evolutionary biologists think it unlikely that SINE sequences would independently appear in the same locations in the genomes of humans and chimpanzees (and other primates). And given their supposed nonfunctional nature, shared SINE DNA in humans and chimpanzees seemingly reflects their common evolutionary ancestry. In fact, evolutionary biologists have gone one step further, using SINE Alu sequences to construct primate evolutionary trees.

SINE DNA Is Functional

Even though many people view shared junk DNA sequences as the most compelling evidence for biological evolution, the growing recognition that virtually every class of junk DNA has function undermines this conclusion. For if these shared sequences are functional, then one could argue that they reflect the Creator’s common design, not shared evolutionary ancestry and common descent. As a case in point, in recent years, molecular biologists have learned that SINE DNA plays a vital role in gene regulation through a variety of distinct mechanisms.2

Staufen-Mediated mRNA Decay

One way SINE sequences regulate gene expression is through a pathway called Staufen-mediated messenger RNA (mRNA) decay (SMD). Critical to an organism’s development, SMD plays a key role in cellular differentiation. SMD is characterized by a complex mechanism centered around the destruction of mRNA. When this degradation takes place, it down-regulates gene expression. The SMD pathway involves binding of a protein called Staufen-1 to one of the ends of the mRNA molecule (dubbed the 3´untranslated region). Staufen-1 binds specifically to double-stranded structures in the 3´untranslated region. This double strand structure forms when Alu sequences in the 3´untranslated region bind to long noncoding RNA molecules containing Alu sequences. This binding event triggers a cascade of additional events that leads to the breakdown of messenger RNA.

Common Descent or Common Design?

As an old-earth creationist, I see the functional role played by noncoding DNA sequences as a reflection of God’s handiwork, defending the case for design from a significant evolutionary challenge. To state it differently: these findings mean that it is just as reasonable to conclude that the shared SINE sequences in the genomes of humans and the great apes reflect common design, not a shared evolutionary ancestry.

In fact, I would maintain that it is more reasonable to think that functional SINE DNA sequences reflect common design, rather than common descent, given the pervasive role these sequence elements play in gene regulation. Because Alu sequences are only found in primates, they must have originated fairly recently (when viewed from an evolutionary framework). Yet, they play an integral and far-reaching role in gene regulation.

And herein lies the first impossible thing evolutionary biologists must believe: Somehow Alusequences arose and then quickly assumed a central place in gene regulation. According to Carl Schmid, a researcher who uncovered some of the first evidence for the functional role played by SINE DNA, “Sine Alus have appeared only recently within the primate lineage, this proposal [of SINE DNA function] provokes the challenging question of how Alu RNA could have possibly assumed a significant role in cell physiology.”3

How Does Junk DNA Acquire Function?

Still, those who subscribe to the evolutionary framework do not view functional junk DNA as incompatible with common descent. They argue that junk DNA acquired function through a process called neofunctionalization. In the case of SMD mediated by Alu sequences in the human genome, evolutionary biologists maintain that occasionally these DNA elements will become incorporated into the 3´untranslated regions of genes and regions of the human genome that produce long noncoding RNAs, and, occasionally, by chance, some of the Alusequences in long noncoding RNAs will have the capacity to pair with the 3´untranslated region of specific mRNAs. When this happens, these Alu sequences trigger SMD-mediated gene regulation. And if this gene regulation has any advantage, it will persist so that over time, some Alu sequences will eventually evolve to assume a role in SMD-mediated gene regulation.

Is Neofunctionalization the Best Explanation for SINE Function?

At some level, this evolutionary scenario seems reasonable (the concerns expressed by Carl Schmid notwithstanding). Still, neofunctionalization events should be relatively rare. And because of the chance nature of neofunctionalization, it would be rational to think that the central role SINE sequences play in SMD gene regulation would be unique to humans.

Why would I make this claim? Based on the nature of evolutionary mechanisms, chance should govern biological and biochemical evolution at its most fundamental level (assuming it occurs). Evolutionary pathways consist of a historical sequence of chance genetic changes operated on by natural selection, which also consists of chance components. The consequences are profound. If evolutionary events could be repeated, the outcome would be dramatically different every time. The inability of evolutionary processes to retrace the same path makes it highly unlikely that the same biological and biochemical designs should appear repeatedly throughout nature.

The concept of historical contingency embodies this idea and is the theme of Stephen Jay Gould’s book Wonderful Life. According to Gould,

“No finale can be specified at the start, none would ever occur a second time in the same way, because any pathway proceeds through thousands of improbable stages. Alter any early event, ever so slightly, and without apparent importance at the time, and evolution cascades into a radically different channel.”4

To help clarify the concept of historical contingency, Gould used the metaphor of “replaying life’s tape.” If one were to push the rewind button, erase life’s history, and let the tape run again, the results would be completely different each time. The very essence of the evolutionary process renders evolutionary outcomes nonrepeatable.

Gould’s perspective of the evolutionary process has been affirmed by other researchers who have produced data, indicating that if evolutionary processes explain the origin of biochemical systems, they must be historically contingent.

Did SMD Evolve Twice?

Yet, collaborators from UC Santa Cruz and the University of Rochester discovered that SINE-mediated SMD appears to have evolved independently—two separate times—in humans and mice, the second impossible thing evolutionary biologists have to believe.

Though rodents don’t possess Alu sequences, they do possess several other SINE elements, labeled B1, B2, B4, and ID. Remarkably, these B/ID sequences occur in regions of the mouse genome corresponding to regions of the human-harboring Alu sequences. And, when the B/ID sequences are associated with the 3´untranslated regions of genes, the mRNA produced from these genes is down-regulated, suggesting that these genes are under the influence of the SMD-mediated pathway—an unexpected result.

But, this finding is not nearly as astonishing as something else the research team discovered. By comparing about 1,200 human-mouse gene pairs in myoblasts, the researchers discovered 24 genes in this cell type that were identical in the human and mouse genomes. These identical genes performed the same physiological role and possessed SINE elements (Alu and B/ID, respectively) and were regulated by the SMD mechanism.

Evolutionary biologists believe that Alu and B/ID SINE sequences emerged independently in the rodent and human lineages. If so, this means that the evolutionary processes must have independently produced the identical outcome—SINE-mediated SMD gene regulation—24 separate times for each of the 24 identical genes. As the researchers point out, chance alone cannot explain their findings. Yet, evolutionary mechanisms are historically contingent and should not yield identical outcomes. This impossible scenario causes me to question if neofunctionalization is the explanation for functional SINE DNA.

And yet, this is not the first time that life scientists have discovered the independent emergence of identical function for junk DNA sequences.

So, which is the better explanation for functional junk DNA sequences: neofunctionalization through historically contingent evolutionary processes or the work of a Mind?

As Alice emphatically complained, “One can’t believe impossible things.”



  1. Brownyn A. Lucas et al., “Evidence for Convergent Evolution of SINE-Directed Staufen-Mediated mRNA Decay,” Proceedings of the National Academy of Sciences, USA Early Edition (January 2018): doi:10.1073/pnas.1715531115.
  2. Reyad A. Elbarbary et al., “Retrotransposons as Regulators of Gene Function,” Science 351 (February 12, 2016): doi:10.1126/science.aac7247.
  3. Carl W. Schmid, “Does SINE Evolution Preclude Alu Function?” Nucleic Acid Research 26 (October 1998): 4541–50, doi:10.1093/nar/26.20.4541.
  4. Stephen Jay Gould, Wonderful Life: The Burgess Shale and the Nature of History (New York: W. W. Norton & Company, 1989), 51.
Reprinted with permission by the author
Original article at:

The Human Genome: Copied by Design



The time my wife Amy and I spent in graduate school studying biochemistry were some of the best days of our lives. But it wasn’t all fun and games. For the most part, we spent long days and nights working in the lab.

But we weren’t alone. Most of the graduate students in the chemistry department at Ohio University kept the same hours we did, with all-nighters broken up around midnight by “Dew n’ Donut” runs to the local 7-Eleven. Even though everybody worked hard, some people were just more productive than others. I soon came to realize that activity and productivity were two entirely different things. Some of the busiest people I knew in graduate school rarely accomplished anything.

This same dichotomy lies at the heart of an important scientific debate taking place about the meaning of the ENCODE project results. This controversy centers around the question: Is the biochemical activity measured for the human genome merely biochemical noise or is it productive for the cell? Or to phrase the question the way a biochemist would: Is biochemical activity associated with the human genome the same thing as biochemical function?

The answer to this question doesn’t just have scientific implications. It impacts questions surrounding humanity’s origin. Did we arise through evolutionary processes or are we the product of a Creator’s handiwork?

The ENCODE Project

The ENCODE project—a program carried out by a consortium of scientists with the goal of identifying the functional DNA sequence elements in the human genome—reported phase II results in the fall of 2012. To the surprise of many, the ENCODE project reported that around 80% of the human genome displays biochemical activity, and hence function, with the expectation that this percentage should increase with phase III of the project.

If valid, the ENCODE results force a radical revision of the way scientists view the human genome. Instead of a wasteland littered with junk DNA sequences (as the evolutionary paradigm predicts), the human genome (and the genomes of other organisms) is packed with functional elements (as expected if a Creator brought human beings into existence).

Within hours of the publication of the phase II results, evolutionary biologists condemned the ENCODE results, citing technical issues with the way the study was designed and the way the results were interpreted. (For a response to these complaints go herehere, and here.)

Is Biochemical Activity the Same Thing As Function?

One of the technical complaints relates to how the ENCODE consortium determined biochemical function. Critics argue that ENCODE scientists conflated biochemical activity with function. For example, the ENCODE Project determined that about 60% of the human genome is transcribed to produceRNA. ENCODE skeptics argue that most of these transcripts lack function. Evolutionary biologist Dan Graur has asserted that “some studies even indicate that 90% of transcripts generated by RNA polymerase II may represent transcriptional noise.”In other words, the biochemical activity measured by the ENCODE project can be likened to busy but nonproductive graduate students who hustle and bustle about the lab but fail to get anything done.

When I first learned how many evolutionary biologists interpreted the ENCODE results I was skeptical. As a biochemist, I am well aware that living systems could not tolerate such high levels of transcriptional noise.

Transcription is an energy- and resource-intensive process. Therefore, it would be untenable to believe that most transcripts are mere biochemical noise. Such a view ignores cellular energetics. Transcribing 60% of the genome when most of the transcripts serve no useful function would routinely waste a significant amount of the organism’s energy and material stores. If such an inefficient practice existed, surely natural selection would eliminate it and streamline transcription to produce transcripts that contribute to the organism’s fitness.

Most RNA Transcripts Are Functional

Recent work supports my intuition as a biochemist. Genomics scientists are quickly realizing that most of the RNA molecule transcribed from the human genome serve critical functional roles.

For example, a recently published report from the Second Aegean International Conference on the Long and the Short of Non-Coding RNAs (held in Greece between June 9–14, 2017) highlights this growing consensus. Based on the papers presented at the conference, the authors of the report conclude, “Non-coding RNAs . . . are not simply transcriptional by-products, or splicing artefacts, but comprise a diverse population of actively synthesized and regulated RNA transcripts. These transcripts can—and do—function within the contexts of cellular homeostasis and human pathogenesis.”2

Shortly before this conference was held, a consortium of scientists from the RIKEN Center for Life Science Technologies in Japan published an atlas of long non-coding RNAs transcribed from the human genome. (Long non-coding RNAs are a subset of RNA transcripts produced from the human genome.) They identified nearly 28,000 distinct long non-coding RNA transcripts and determined that nearly 19,200 of these play some functional role, with the possibility that this number may increase as they and other scientific teams continue to study long non-coding RNAs.3 One of the researchers involved in this project acknowledges that “There is strong debate in the scientific community on whether the thousands of long non-coding RNAs generated from our genomes are functional or simply byproducts of a noisy transcriptional machinery . . . we find compelling evidence that the majority of these long non-coding RNAs appear to be functional.”4

Copied by Design

Based on these results, it becomes increasingly difficult for ENCODE skeptics to dismiss the findings of the ENCODE project. Independent studies affirm the findings of the ENCODE consortium—namely, that a vast proportion of the human genome is functional.

We have come a long way from the early days of the human genome project. When completed in 2003, many scientists at that time estimated that around 95% of the human genome consisted of junk DNA. And in doing so, they seemingly provided compelling evidence that humans must be the product of an evolutionary history.

But, here we are, nearly 15 years later. And the more we learn about the structure and function of genomes, the more elegant and sophisticated they appear to be. And the more reasons we have to think that the human genome is the handiwork of our Creator.



  1. Dan Graur et al., “On the Immortality of Television Sets: ‘Function’ in the Human Genome According to the Evolution-Free Gospel of ENCODE,” Genome Biology and Evolution5 (March 1, 2013): 578–90, doi:10.1093/gbe/evt028.
  2. Jun-An Chen and Simon Conn, “Canonical mRNA is the Exception, Rather than the Rule,” Genome Biology 18 (July 7, 2017): 133, doi:10.1186/s13059-017-1268-1.
  3. Chung-Chau Hon et al., “An Atlas of Human Long Non-Coding RNAs with Accurate 5′ Ends,” Nature 543 (March 9, 2017): 199–204, doi:10.1038/nature21374.
  4. RIKEN, “Improved Gene Expression Atlas Shows that Many Human Long Non-Coding RNAs May Actually Be Functional,” ScienceDaily, March 1, 2017,

Is 75% of the Human Genome Junk DNA?


By the rude bridge that arched the flood,
Their flag to April’s breeze unfurled,
Here once the embattled farmers stood,
And fired the shot heard round the world.

–Ralph Waldo Emerson, Concord Hymn

Emerson referred to the Battles of Lexington and Concord, the first skirmishes of the Revolutionary War, as the “shot heard round the world.”

While not as loud as the gunfire that triggered the Revolutionary War, a recent article published in Genome Biology and Evolution by evolutionary biologist Dan Graur has garnered a lot of attention,1 serving as the latest salvo in the junk DNA wars—a conflict between genomics scientists and evolutionary biologists about the amount of functional DNA sequences in the human genome.

Clearly, this conflict has important scientific ramifications, as researchers strive to understand the human genome and seek to identify the genetic basis for diseases. The functional content of the human genome also has significant implications for creation-evolution skirmishes. If most of the human genome turns out to be junk after all, then the case for a Creator potentially suffers collateral damage.

According to Graur, no more than 25% of the human genome is functional—a much lower percentage than reported by the ENCODE Consortium. Released in September 2012, phase II results of the ENCODE project indicated that 80% of the human genome is functional, with the expectation that the percentage of functional DNA in the genome would rise toward 100% when phase III of the project reached completion.

If true, Graur’s claim would represent a serious blow to the validity of the ENCODE project conclusions and devastate the RTB human origins creation model. Intelligent design proponents and creationists (like me) have heralded the results of the ENCODE project as critical in our response to the junk DNA challenge.

Junk DNA and the Creation vs. Evolution Battle

Evolutionary biologists have long considered the presence of junk DNA in genomes as one of the most potent pieces of evidence for biological evolution. Skeptics ask, “Why would a Creator purposely introduce identical nonfunctional DNA sequences at the same locations in the genomes of different, though seemingly related, organisms?”

When the draft sequence was first published in 2000, researchers thought only around 2–5% of the human genome consisted of functional sequences, with the rest being junk. Numerous skeptics and evolutionary biologists claim that such a vast amount of junk DNA in the human genome is compelling evidence for evolution and the most potent challenge against intelligent design/creationism.

But these arguments evaporate in the wake of the ENCODE project. If valid, the ENCODE results would radically alter our view of the human genome. No longer could the human genome be regarded as a wasteland of junk; rather, the human genome would have to be recognized as an elegantly designed system that displays sophistication far beyond what most evolutionary biologists ever imagined.

ENCODE Skeptics

The findings of the ENCODE project have been criticized by some evolutionary biologists who have cited several technical problems with the study design and the interpretation of the results. (See articles listed under “Resources to Go Deeper” for a detailed description of these complaints and my responses.) But ultimately, their criticisms appear to be motivated by an overarching concern: if the ENCODE results stand, then it means key features of the evolutionary paradigm can’t be correct.

Calculating the Percentage of Functional DNA in the Human Genome

Graur (perhaps the foremost critic of the ENCODE project) has tried to discredit the ENCODE findings by demonstrating that they are incompatible with evolutionary theory. Toward this end, he has developed a mathematical model to calculate the percentage of functional DNA in the human genome based on mutational load—the amount of deleterious mutations harbored by the human genome.

Graur argues that junk DNA functions as a “sponge” absorbing deleterious mutations, thereby protecting functional regions of the genome. Considering this buffering effect, Graur wanted to know how much junk DNA must exist in the human genome to buffer against the loss of fitness—which would result from deleterious mutations in functional DNA—so that a constant population size can be maintained.

Historically, the replacement level fertility rates for human beings have been two to three children per couple. Based on Graur’s modeling, this fertility rate requires 85–90% of the human genome to be composed of junk DNA in order to absorb deleterious mutations—ensuring a constant population size, with the upper limit of functional DNA capped at 25%.

Graur also calculated a fertility rate of 15 children per couple, at minimum, to maintain a constant population size, assuming 80% of the human genome is functional. According to Graur’s calculations, if 100% of the human genome displayed function, the minimum replacement level fertility rate would have to be 24 children per couple.

He argues that both conclusions are unreasonable. On this basis, therefore, he concludes that the ENCODE results cannot be correct.

Response to Graur

So, has Graur’s work invalidated the ENCODE project results? Hardly. Here are four reasons why I’m skeptical.

1. Graur’s estimate of the functional content of the human genome is based on mathematical modeling, not experimental results.

An adage I heard repeatedly in graduate school applies: “Theories guide, experiments decide.” Though the ENCODE project results theoretically don’t make sense in light of the evolutionary paradigm, that is not a reason to consider them invalid. A growing number of studies provide independent experimental validation of the ENCODE conclusions. (Go here and here for two recent examples.)

To question experimental results because they don’t align with a theory’s predictions is a “Bizarro World” approach to science. Experimental results and observations determine a theory’s validity, not the other way around. Yet when it comes to the ENCODE project, its conclusions seem to be weighed based on their conformity to evolutionary theory. Simply put, ENCODE skeptics are doing science backwards.

While Graur and other evolutionary biologists argue that the ENCODE results don’t make sense from an evolutionary standpoint, I would argue as a biochemist that the high percentage of functional regions in the human genome makes perfect sense. The ENCODE project determined that a significant fraction of the human genome is transcribed. They also measured high levels of protein binding.

ENCODE skeptics argue that this biochemical activity is merely biochemical noise. But this assertion does not make sense because (1) biochemical noise costs energy and (2) random interactions between proteins and the genome would be harmful to the organism.

Transcription is an energy- and resource-intensive process. To believe that most transcripts are merely biochemical noise would be untenable. Such a view ignores cellular energetics. Transcribing a large percentage of the genome when most of the transcripts serve no useful function would routinely waste a significant amount of the organism’s energy and material stores. If such an inefficient practice existed, surely natural selection would eliminate it and streamline transcription to produce transcripts that contribute to the organism’s fitness.

Apart from energetics considerations, this argument ignores the fact that random protein binding would make a dire mess of genome operations. Without minimizing these disruptive interactions, biochemical processes in the cell would grind to a halt. It is reasonable to think that the same considerations would apply to transcription factor binding with DNA.

2. Graur’s model employs some questionable assumptions.

Graur uses an unrealistically high rate for deleterious mutations in his calculations.

Graur determined the deleterious mutation rate using protein-coding genes. These DNA sequences are highly sensitive to mutations. In contrast, other regions of the genome that display function—such as those that (1) dictate the three-dimensional structure of chromosomes, (2) serve as transcription factors, and (3) aid as histone binding sites—are much more tolerant to mutations. Ignoring these sequences in the modeling work artificially increases the amount of required junk DNA to maintain a constant population size.

3. The way Graur determines if DNA sequence elements are functional is questionable. 

Graur uses the selected-effect definition of function. According to this definition, a DNA sequence is only functional if it is undergoing negative selection. In other words, sequences in genomes can be deemed functional only if they evolved under evolutionary processes to perform a particular function. Once evolved, these sequences, if they are functional, will resist evolutionary change (due to natural selection) because any alteration would compromise the function of the sequence and endanger the organism. If deleterious, the sequence variations would be eliminated from the population due to the reduced survivability and reproductive success of organisms possessing those variants. Hence, functional sequences are those under the effects of selection.

In contrast, the ENCODE project employed a causal definition of function. Accordingly, function is ascribed to sequences that play some observationally or experimentally determined role in genome structure and/or function.

The ENCODE project focused on experimentally determining which sequences in the human genome displayed biochemical activity using assays that measured

  • transcription,
  • binding of transcription factors to DNA,
  • histone binding to DNA,
  • DNA binding by modified histones,
  • DNA methylation, and
  • three-dimensional interactions between enhancer sequences and genes.

In other words, if a sequence is involved in any of these processes—all of which play well-established roles in gene regulation—then the sequences must have functional utility. That is, if sequenceQperforms functionG, then sequenceQis functional.

So why does Graur insist on a selected-effect definition of function? For no other reason than a causal definition ignores the evolutionary framework when determining function. He insists that function be defined exclusively within the context of the evolutionary paradigm. In other words, his preference for defining function has more to do with philosophical concerns than scientific ones—and with a deep-seated commitment to the evolutionary paradigm.

As a biochemist, I am troubled by the selected-effect definition of function because it is theory-dependent. In science, cause-and-effect relationships (which include biological and biochemical function) need to be established experimentally and observationally,independent of any particular theory. Once these relationships are determined, they can then be used to evaluate the theories at hand. Do the theories predict (or at least accommodate) the established cause-and-effect relationships, or not?

Using a theory-dependent approach poses the very real danger that experimentally determined cause-and-effect relationships (or, in this case, biological functions) will be discarded if they don’t fit the theory. And, again, it should be the other way around. A theory should be discarded, or at least reevaluated, if its predictions don’t match these relationships.

What difference does it make which definition of function Graur uses in his model? A big difference. The selected-effect definition is more restrictive than the causal-role definition. This restrictiveness translates into overlooked function and increases the replacement level fertility rate.

4. Buffering against deleterious mutations is a function.

As part of his model, Graur argues that junk DNA is necessary in the human genome to buffer against deleterious mutations. By adopting this view, Graur has inadvertently identified function for junk DNA. In fact, he is not the first to argue along these lines. Biologist Claudiu Bandea has posited that high levels of junk DNA can make genomes resistant to the deleterious effects of transposon insertion events in the genome. If insertion events are random, then the offending DNA is much more likely to insert itself into “junk DNA” regions instead of coding and regulatory sequences, thus protecting information-harboring regions of the genome.

If the last decade of work in genomics has taught us anything, it is this: we are in our infancy when it comes to understanding the human genome. The more we learn about this amazingly complex biochemical system, the more elegant and sophisticated it becomes. Through this process of discovery, we continue to identify functional regions of the genome—DNA sequences long thought to be “junk.”

In short, the criticisms of the ENCODE project reflect a deep-seated commitment to the evolutionary paradigm and, bluntly, are at war with the experimental facts.

Bottom line: if the ENCODE results stand, it means that key aspects of the evolutionary paradigm can’t be correct.

Resources to Go Deeper


  1. Dan Graur, “An Upper Limit on the Functional Fraction of the Human Genome,” Genome Biology and Evolution 9 (July 2017): 1880–85, doi:10.1093/gbe/evx121.