Junk DNA, ENCODE, and the Design Debate

In September 2012 a consortium of scientists published the headline results of the ENCODE project, a large, publicly funded effort to map the working parts of the human genome. The lead paper in Nature said its data "enabled us to assign biochemical functions for 80% of the genome." Newspapers and science magazines soon announced the end of "junk DNA," the idea that much of our DNA does nothing useful for the body. Intelligent Design (ID) writers greeted the news as the fulfilment of a prediction they had made years earlier.

A sharp counterattack followed. Evolutionary biologists argued that ENCODE had used a definition of "function" so loose that almost any stretch of DNA would qualify, and that the real functional fraction was nearer 10 percent. By 2014 several of ENCODE's own leaders had published a more cautious account of what their data could and could not show.

This article asks what happened, what the evidence now suggests, how the design argument used the episode, and whether either side overstated its case. A Christian argument built on a scientific headline is only as strong as the headline.

In brief

ENCODE's 2012 figure of 80 percent described DNA that showed at least one reproducible biochemical event, such as being copied into RNA or bound by a protein, in at least one of the cell types studied. It did not show that 80 percent of the genome does something the body needs. Critics such as Dan Graur and W. Ford Doolittle argued that biochemical activity is a poor test of function, and ENCODE scientists themselves wrote in 2014 that it "is not at all simple" to say how much of the active genome is functional. Estimates based on evolutionary constraint and mutation rates mostly fall between about 8 and 15 percent, though a minority of scientists, such as John Mattick, argue for much more. Some "junk" sequences really have turned out to have functions. But the claim that ENCODE "overturned" junk DNA or vindicated design went beyond the evidence, and the older argument that junk DNA disproves a designer rested on theological assumptions science cannot test.

Thesis: ENCODE did not show that 80 percent of the human genome is functional in the sense that matters for the design debate, the true fraction remains disputed, and Christians should neither rest the case for design on "no junk" nor fear the possibility that much of the genome is unused.

What "junk DNA" meant

The phrase was already in use in the 1960s but is usually credited to the geneticist Susumu Ohno, whose 1972 paper was titled "So much 'junk' DNA in our genome." Ohno was thinking chiefly of pseudogenes: copies of genes that have been disabled by mutation. Alexander Palazzo and T. Ryan Gregory note that the term "was quickly extended to include many types of noncoding DNA."¹ Only about 1 to 2 percent of the human genome codes for protein. The largest component of the rest consists of transposable elements, sequences that copy themselves around the genome; Palazzo and Gregory report estimates from one-half to two-thirds of the genome, most of it degraded and inactive.²

Two lines of reasoning led many biologists to expect much non-functional DNA. One was genome size. Genome size varies enormously and does not track complexity. The "onion test," proposed by Gregory in 2007, puts this sharply: the onion has about five times as much DNA as a human, so anyone who proposes a universal function for non-coding DNA should explain why an onion needs so much more of it.³ The other was population genetics: selection must protect functional DNA against harmful mutations, and there are limits to how much it can protect in a species like ours.

"Junk" never had to mean worthless. Sydney Brenner drew a distinction that later critics of ENCODE quoted: "There is the rubbish we keep, which is junk, and the rubbish we throw away, which is garbage." On this view, extra DNA persists because it is harmless and because the processes that add DNA outpace those that remove it.⁴

What ENCODE claimed in 2012

The 2012 lead paper reported 1,640 datasets across 147 cell types. In the main text the consortium said that "the vast majority (80.4%) of the human genome participates in at least one biochemical RNA and/or chromatin associated event in at least one cell type."⁵

The definition was stated openly. The paper said that "Operationally, we define a functional element as a discrete genome segment that encodes a defined product (e.g., protein or non-coding RNA) or displays a reproducible biochemical signature (e.g., protein-binding, or a specific chromatin structure)." In the same paragraph it acknowledged that comparative genomic studies "suggest that 3–8% of bases are under purifying (negative) selection." The paper also gave a spectrum: 2.9 percent of bases in protein-coding exons, 8.5 percent in specific protein-binding sites, and about 80 percent at the broadest level.⁶

Headline readers missed these qualifications. Science ran a news story titled "ENCODE Project Writes Eulogy for Junk DNA," and a Nature feature summarized the result as assigning "some sort of function to roughly 80% of the genome."⁷ {{very-strong: ENCODE's 80 percent figure counted any reproducible biochemical activity in any tested cell type; it was not a measurement of how much DNA the body needs.}}

The backlash

The best-known reply came from Dan Graur and five colleagues in Genome Biology and Evolution in 2013, under the deliberately provocative title "On the Immortality of Television Sets: 'Function' in the Human Genome According to the Evolution-Free Gospel of ENCODE." Their abstract argued that if 80 percent of the genome is functional while fewer than 10 percent is conserved by purifying selection, then about 70 percent would have to be immune to harmful mutation, which they called an "absurd conclusion."⁸ They argued that ENCODE had used a "causal role" idea of function (what a sequence does chemically) rather than a "selected effect" idea (what natural selection maintains it for), and that much transcription and protein binding could be biochemical noise. Hence the title: a television left on and unattended will not still work after a million years, and DNA that selection does not protect will not stay functional.⁹

W. Ford Doolittle made a related argument in the Proceedings of the National Academy of Sciences. He asked what ENCODE's methods would find in much larger genomes. If the number of "functional elements" stayed roughly constant, the extra DNA in large genomes would look like junk. If it rose with genome size, then large-genomed organisms are more complex than we are, or ENCODE's definition was counting sites that biologists would not normally call functional, or the same functions are spread more diffusely in larger genomes; he judged the last two plausible.¹⁰

In 2014 a group of thirty authors, among them leading ENCODE scientists such as Ewan Birney, one of the 2012 paper's lead analysts, published "Defining functional DNA elements in the human genome." They wrote that "it is not at all simple to establish what fraction of the biochemically annotated genome should be regarded as functional." They also defended their approach, arguing that evolutionary methods can miss real function: "absence of conservation cannot be interpreted as evidence for the lack of function." Their conclusion placed the value of ENCODE in its maps, which they believed "far more important than any interim estimate of the fraction of the human genome that is functional."¹¹ The 2014 paper did not defend 80 percent as a measure of biological function.

How much of the genome is functional?

No one has an exact answer, and the estimates depend on what "functional" means.

Approach	Study	Estimate
Biochemical activity in any cell type	ENCODE 2012	about 80% (any activity)
Candidate regulatory elements only (not a total for all function)	ENCODE 2020	7.9% of the human genome
Constraint along the human lineage	Rands et al. 2014	8.2% (range 7.1 to 9.2%)
Human variation in non-conserved active DNA	Ward and Kellis 2012	at least an additional 4%
Constraint across 240 mammals	Zoonomia, Christmas et al. 2023	at least 10.7% (a lower bound)
Mutational load (theoretical maximum)	Graur 2017, corrected 2019	no more than 15%

ENCODE's 2020 registry of 926,535 candidate regulatory elements covers 7.9 percent of the genome.¹² Rands and colleagues estimated in 2014 that "8.2% (7.1–9.2%) of the human genome is presently subject to negative selection and thus is likely to be functional."¹³ Ward and Kellis, using human population data, found that some active DNA not conserved across mammals is still under selection, "with at least an additional 4% of the human genome subject to lineage-specific constraint."¹⁴ The Zoonomia project's 2023 alignment of 240 mammalian genomes found that "At least 332 million bases (~10.7%)" of the human genome are unusually conserved, a minimum figure.¹⁵

Graur used the mutational load argument in 2017 to set an upper limit. As first published the paper gave 25 percent; a 2019 correction added a missing factor to one equation, and the corrected abstract concludes that the functional fraction "cannot exceed 15%."¹⁶ The result depends on assumed mutation rates.

Experiments point in both directions. In 2004 Marcelo Nóbrega and colleagues deleted two large stretches of non-coding DNA from mice, about 1.5 million and 845,000 base pairs long, and the mice were "indistinguishable from wild-type littermates" in the traits measured, which the authors took as support for "potentially 'disposable DNA'."¹⁷ On the other side, some sequences once dismissed have proved active. A 2010 study in Nature showed that PTENP1, a pseudogene of the tumour-suppressor gene PTEN, "is biologically active as it can regulate cellular levels of PTEN."¹⁸ John Mattick and Marcel Dinger, who are not ID advocates, argued in 2013 that conservation-based estimates rest on "circular assumptions" about transposon-derived DNA and that much more of the genome is likely to be functional.¹⁹ Theirs is a minority view, but a serious one.

{{moderate: Most current evidence-based estimates place the functional fraction of the human genome roughly between 8 and 15 percent, with real uncertainty about function that evolutionary methods cannot detect.}}

How the design argument used the claim

ID writers had taken up the issue long before 2012. In First Things in October 1998, William Dembski answered the charge that design is a "science stopper" by pointing to junk DNA. "Thus on an evolutionary view we expect a lot of useless DNA," he wrote. "If, on the other hand, organisms are designed, we expect DNA, as much as possible, to exhibit function." He added that "the most recent findings suggest that designating DNA as 'junk' merely cloaks our current lack of knowledge about function."²⁰ In 2011 the biologist Jonathan Wells published The Myth of Junk DNA with the Discovery Institute Press, surveying research on introns, pseudogenes and other non-coding DNA.²¹

After ENCODE, Casey Luskin of the Discovery Institute's Center for Science and Culture wrote a four-part series for Salvo magazine, whose last three parts appeared in 2015. He described ENCODE as finding that "80 percent of the human genome is biochemically functional—with 100 percent functionality in sight—overturning the concept of junk DNA."²² He answered the critics directly. On the onion test, he pointed to correlations between genome size and cell size and to genome duplication (polyploidy) in plants. On conservation, he argued that the critics' reasoning assumes its conclusion: "an intelligent agent could independently design functional genetic elements with widely divergent DNA sequences in the genomes of different species—no 'conservation' required."²³ He concluded that "ID boldly predicted ENCODE's results, and evolutionary biology did not."²⁴

Some of this deserves a fair hearing. Dembski did point toward function in non-coding DNA, the functions found since then are real, and ENCODE's own scientists conceded that missing conservation does not prove missing function. But ENCODE's authors did not defend 80 percent as biological function by 2014, and the claim that ID "predicted ENCODE's results" depends on reading it that way. {{debated: The ENCODE results confirmed a distinctive Intelligent Design prediction.}}

Overstatement on both sides

On the evolutionary side, junk DNA had become an argument against a designer. Mattick and Dinger summarize this use of the idea, citing Richard Dawkins (1986) and Francis Collins (2006): non-coding DNA was presented as "evidence for the accumulation of evolutionary debris by blind Darwinian evolution," on the assumption that "an intelligent designer would presumably not fill the human genetic instruction set with meaningless information."²⁵ That premise is theological, not scientific: it assumes we know what a Creator would do with a genome. Graur's paper made the stakes explicit in a way that went beyond the data: "ENCODE's take-home message that everything has a function implies purpose, and purpose is the only thing that evolution cannot provide."²⁶ Whether biological processes have purpose is a philosophical question, and percentages of functional DNA cannot settle it either way.

On the design side, the popular story that biologists dismissed non-coding DNA and refused to study it is overstated. Palazzo and Gregory reply that "it is simply not true that potential functions for noncoding DNA were ignored until recently," and that "even the earliest proponents of 'junk DNA' and 'selfish DNA' predicted that further examples would be found."²⁷ Mattick and Dinger, the scientists most sympathetic to a highly functional genome, insist that their view is "entirely consistent with the broad tenets of evolution by natural selection."²⁸ A functional genome, if that is what we have, would not by itself point to design.

This is where the evidence weakens the design argument as it is often presented. "No junk, therefore design" fails for the same reason as "junk, therefore no design." Both turn a disputed percentage into a verdict it cannot deliver (compare Is Intelligent Design a "God of the Gaps" Argument?).

The Orthodox position

The Orthodox Church has no dogma about genomes, and it does not rest faith in the Creator on any particular scientific finding. The Creed confesses God the Father Almighty, "Maker of Heaven and Earth and of all things visible and invisible,"²⁹ and the Church reads the created order as the work of God's wisdom: "O Lord, how manifold are Your works! In wisdom You have made them all" (Psalm 104:24, NKJV; Psalm 103 in the Septuagint numbering). St Paul says that God's "invisible attributes are clearly seen, being understood by the things that are made" (Romans 1:20, NKJV).³⁰ The Fathers commonly saw the order and beauty of creation as a pointer to its Maker. St Basil the Great told his hearers in the Hexaemeron: "I want creation to penetrate you with so much admiration that everywhere, wherever you may be, the least plant may bring to you the clear remembrance of the Creator."³¹

Basil did hold that among plants "not a single thing is useless" (Hexaemeron 5.4), but he meant whole creatures and their uses, not every part of every creature. An Orthodox reader can hold the junk DNA question with the same calm. First, the Church does not teach that creation is now in its finished, pristine state. St Paul writes that "the creation was subjected to futility" and awaits deliverance "from the bondage of corruption" (Romans 8:20-21, NKJV).³² That is a theological statement, not a theory of the genome, but it means a Christian is not committed to the idea that every nucleotide must be useful. Second, this site's own approach to design treats it as an inference to be weighed against alternatives, rather than as a demand that science find no loose ends (see Design as an Inference, Not a Science-Stopper and In DNA, Function, Not Mere Improbability, Matters). The stronger design arguments from biology concern the origin of the coded information in DNA and the cell's machinery for reading it, which is a different question from how much of the genome is in use today (see The Design Inference: DNA, Information, and Function).

Christians are found on both sides of this debate, and both deserve respect. Francis Collins, who led the Human Genome Project, founded BioLogos after The Language of God (2006); it presents what it calls "Evolutionary Creation," believes "that our intelligent God designed the universe," but sees no "scientific or biblical reasons to give up on pursuing natural explanations," and so distinguishes itself from Intelligent Design.³³ Dembski, Wells and Luskin, at the Discovery Institute, argue that design is detectable in biology (see Can a Christian Accept Evolution?). An Orthodox reader can learn from both while keeping faith anchored in the revelation of God in Christ, received in the Church, rather than in a percentage that may change with the next study. For the wider picture, see Has Science Disproved Christianity?

The strongest objections

"A wise designer would not fill our genome with junk, so junk DNA counts against God."

This assumes knowledge of a Creator's purposes that no one has. Scripture describes a good creation that is also, for now, subject to corruption, not a machine optimised for economy. It has also lost ground as particular "junk" sequences, such as PTENP1, have proved useful. A largely unused genome would tell us about the history of genomes, not that there is no God.

"ID predicted ENCODE, and evolutionary biology did not."

Luskin is right that Dembski predicted more function than many biologists expected, and that the junk label was sometimes applied too quickly. But the prediction was loose ("as much as possible"), and evolutionary theory had long allowed for function in non-coding DNA; as Palazzo and Gregory note, functional non-coding sequences have been known for more than half a century. The decisive problem is that the 80 percent figure measured biochemical events, and ENCODE's own scientists stepped back from it as a measure of function. On present evidence the prediction is unconfirmed, though it may yet prove partly right if function turns out to be much more widespread than constraint suggests.

"Conservation-based estimates assume evolution, so they are circular."

This is ID's strongest technical reply, and Mattick and Dinger make it too. Alignment methods can indeed miss function. But not every line of evidence depends on comparing species. Ward and Kellis used variation within living humans. Graur's upper limit uses measured human mutation rates and the arithmetic of harmful mutations, which act on any functional DNA whatever its origin. The mouse deletion experiments are direct tests in living animals. None is final, but together they make the very high estimates hard to sustain.

What we can say confidently

• ENCODE's 2012 figure of about 80 percent counted DNA showing any reproducible biochemical activity in at least one cell type, and the paper said so.
• ENCODE scientists themselves wrote in 2014 that the functional fraction of the biochemically active genome is not simple to establish, and they did not repeat 80 percent as a measure of biological function.
• Estimates of constraint across mammals have risen as more genomes have been compared: the 2023 Zoonomia study found a lower bound of about 10.7 percent, against earlier estimates of at least 5 percent from fewer species.
• Some DNA once called junk, including some pseudogenes and transposable-element sequences, has real functions.
• Large deletions of non-coding DNA in mice have produced no detectable harm in the traits measured, which suggests some DNA is dispensable.
• Both "junk disproves design" and "no junk proves design" rest on assumptions that go beyond the data.

What remains disputed

• The true functional fraction of the human genome. Most estimates fall roughly between 8 and 15 percent, while Mattick and others argue for much more.
• Whether low-level transcription and protein binding are mostly noise or mostly regulated activity.
• How much function is invisible to methods based on conservation.
• Whether the onion test is decisive, or whether cell size and genome duplication explain large genomes, as ID writers argue.
• Whether Dembski's 1998 statement counts as a distinctive, testable prediction of Intelligent Design.

Conclusion

The junk DNA debate shows how quickly a careful paper can become a slogan. ENCODE produced a valuable map of the genome's activity. It did not show that 80 percent of our DNA is needed, and the people who wrote it said as much. Its critics made a strong case that most of the genome is probably not functional in the biological sense, but they too sometimes spoke as if a percentage could answer a question about purpose.

For an Orthodox Christian, none of this unsettles faith in the God who made heaven and earth. The world is the work of a wise Creator whether the genome is one-tenth functional or nine-tenths. The Church asks us to look at creation with the wonder St Basil commended, tell the truth about what we find, and leave room for the next discovery. The design argument is better served by that honesty than by a claim that the evidence has already outgrown.

For how this site rates evidence, see How We Work and the evidence ratings explainer.

Notes

1. Susumu Ohno, "So much 'junk' DNA in our genome," in H. H. Smith, ed., Evolution of Genetic Systems (New York: Gordon and Breach, 1972), 366-370, as cited and described in Alexander F. Palazzo and T. Ryan Gregory, "The Case for Junk DNA," PLoS Genetics 10, no. 5 (2014): e1004351. ↩
2. Palazzo and Gregory, "The Case for Junk DNA," section "Genome Composition." ↩
3. Palazzo and Gregory, "The Case for Junk DNA," section "Genome Size and 'The Onion Test'," citing T. Ryan Gregory, "The onion test" (2007, Genomicron blog). ↩
4. Sydney Brenner, "Refuge of spandrels," Current Biology 8, no. 19 (1998): R669, as quoted in Dan Graur et al., "On the Immortality of Television Sets," Genome Biology and Evolution 5, no. 3 (2013): 578-590. ↩
5. The ENCODE Project Consortium, "An integrated encyclopedia of DNA elements in the human genome," Nature 489 (2012): 57-74, abstract and introduction. ↩
6. ENCODE Project Consortium, "An integrated encyclopedia," introduction and section on the spectrum of coverage. ↩
7. Elizabeth Pennisi, "ENCODE Project Writes Eulogy for Junk DNA," Science 337 (2012): 1159-1161; Brendan Maher, "ENCODE: The human encyclopaedia," Nature 489 (2012): 46-48. ↩
8. Dan Graur, Yichen Zheng, Nicholas Price, Ricardo B. R. Azevedo, Rebecca A. Zufall and Eran Elhaik, "On the Immortality of Television Sets: 'Function' in the Human Genome According to the Evolution-Free Gospel of ENCODE," Genome Biology and Evolution 5, no. 3 (2013): 578-590, abstract. ↩
9. Graur et al., "On the Immortality of Television Sets," sections on "selected effect" and "causal role" functions. ↩
10. W. Ford Doolittle, "Is junk DNA bunk? A critique of ENCODE," Proceedings of the National Academy of Sciences 110, no. 14 (2013): 5294-5300, abstract. ↩
11. Manolis Kellis et al., "Defining functional DNA elements in the human genome," Proceedings of the National Academy of Sciences 111, no. 17 (2014): 6131-6138. ↩
12. The ENCODE Project Consortium et al., "Expanded encyclopaedias of DNA elements in the human and mouse genomes," Nature 583 (2020): 699-710. ↩
13. Chris M. Rands, Stephen Meader, Chris P. Ponting and Gerton Lunter, "8.2% of the Human Genome Is Constrained: Variation in Rates of Turnover across Functional Element Classes in the Human Lineage," PLoS Genetics 10, no. 7 (2014): e1004525, abstract. ↩
14. Lucas D. Ward and Manolis Kellis, "Evidence of Abundant Purifying Selection in Humans for Recently Acquired Regulatory Functions," Science 337 (2012): 1675-1678, abstract. ↩
15. Matthew J. Christmas et al. (Zoonomia Consortium), "Evolutionary constraint and innovation across hundreds of placental mammals," Science 380, no. 6643 (2023): eabn3943, abstract and section "Estimate of genome-wide constraint." ↩
16. Dan Graur, "An Upper Limit on the Functional Fraction of the Human Genome," Genome Biology and Evolution 9, no. 7 (2017): 1880-1885; Dan Graur, "Corrigendum," Genome Biology and Evolution 11, no. 11 (2019): 3158. ↩
17. Marcelo A. Nóbrega et al., "Megabase deletions of gene deserts result in viable mice," Nature 431 (2004): 988-993, abstract. ↩
18. Laura Poliseno et al., "A coding-independent function of gene and pseudogene mRNAs regulates tumour biology," Nature 465 (2010): 1033-1038, abstract. ↩
19. John S. Mattick and Marcel E. Dinger, "The extent of functionality in the human genome," The HUGO Journal 7 (2013): 2, abstract. ↩
20. William A. Dembski, "Science and Design," First Things 86 (October 1998). Later Discovery Institute writers cite it as "Intelligent Science and Design," 21-27. ↩
21. Jonathan Wells, The Myth of Junk DNA (Seattle: Discovery Institute Press, 2011). ↩
22. Casey Luskin, "The ENCODE Embroilment, Part II," Salvo 32 (2015). ↩
23. Casey Luskin, "The ENCODE Embroilment, Part III," Salvo 33 (2015). ↩
24. Casey Luskin, "The ENCODE Embroilment, Part IV," Salvo 34 (Fall 2015). ↩
25. Mattick and Dinger, "The extent of functionality in the human genome," end of the "Review" section, just before "Conclusions," citing Richard Dawkins, The Blind Watchmaker (1986) and Francis S. Collins, The Language of God (2006). ↩
26. Graur et al., "On the Immortality of Television Sets," closing paragraphs of the main argument, citing Dembski (1998; his "Reinstating design within science," Rhetoric & Public Affairs 1, not the First Things essay) and Wells (2004). ↩
27. Palazzo and Gregory, "The Case for Junk DNA," introduction and "Concluding Remarks." ↩
28. Mattick and Dinger, "The extent of functionality in the human genome," "Conclusions." ↩
29. The Nicene-Constantinopolitan Creed, in the translation of the Orthodox Church in America, "The Creed: The Symbol of Faith," oca.org. ↩
30. Scripture quotations are from the New King James Version (Nashville: Thomas Nelson, 1982), checked at Bible Gateway. ↩
31. Basil of Caesarea, Hexaemeron, Homily 5.2 (and 5.4 for "not a single thing is useless"), trans. Blomfield Jackson, Nicene and Post-Nicene Fathers, 2nd series, vol. 8 (Buffalo, NY: Christian Literature Publishing Co., 1895), revised edition at New Advent. ↩
32. Romans 8:20-21, NKJV. ↩
33. BioLogos, "About BioLogos: Our History" (biologos.org/about-us) and "How is Evolutionary Creation different from Evolutionism, Intelligent Design, and Creationism?" (biologos.org/common-questions), both read via the Internet Archive (January 2025 captures) because biologos.org refused automated requests; Francis S. Collins, The Language of God (New York: Free Press, 2006). ↩

Bibliography

Primary sources

• The ENCODE Project Consortium. "An integrated encyclopedia of DNA elements in the human genome." Nature 489, no. 7414 (2012): 57-74. doi:10.1038/nature11247.
• The ENCODE Project Consortium et al. "Expanded encyclopaedias of DNA elements in the human and mouse genomes." Nature 583, no. 7818 (2020): 699-710. doi:10.1038/s41586-020-2493-4.
• Kellis, Manolis, Barbara Wold, Michael P. Snyder, et al. "Defining functional DNA elements in the human genome." Proceedings of the National Academy of Sciences 111, no. 17 (2014): 6131-6138. doi:10.1073/pnas.1318948111.
• The Holy Bible, New King James Version. Nashville: Thomas Nelson, 1982.

Orthodox and patristic sources

• Orthodox Church in America. "The Creed: The Symbol of Faith." oca.org/orthodoxy/prayers/symbol-of-faith.
• Basil of Caesarea. Hexaemeron. Translated by Blomfield Jackson. Nicene and Post-Nicene Fathers, 2nd series, vol. 8. Buffalo, NY: Christian Literature Publishing Co., 1895. Revised edition at newadvent.org.

Other scholarship (including scholars who disagree)

• Christmas, Matthew J., Irene M. Kaplow, et al. "Evolutionary constraint and innovation across hundreds of placental mammals." Science 380, no. 6643 (2023): eabn3943. doi:10.1126/science.abn3943.
• Doolittle, W. Ford. "Is junk DNA bunk? A critique of ENCODE." Proceedings of the National Academy of Sciences 110, no. 14 (2013): 5294-5300. doi:10.1073/pnas.1221376110.
• Graur, Dan. "An Upper Limit on the Functional Fraction of the Human Genome." Genome Biology and Evolution 9, no. 7 (2017): 1880-1885. doi:10.1093/gbe/evx121. Corrigendum, Genome Biology and Evolution 11, no. 11 (2019): 3158. doi:10.1093/gbe/evz231.
• Graur, Dan, Yichen Zheng, Nicholas Price, Ricardo B. R. Azevedo, Rebecca A. Zufall, and Eran Elhaik. "On the Immortality of Television Sets: 'Function' in the Human Genome According to the Evolution-Free Gospel of ENCODE." Genome Biology and Evolution 5, no. 3 (2013): 578-590. doi:10.1093/gbe/evt028.
• Maher, Brendan. "ENCODE: The human encyclopaedia." Nature 489 (2012): 46-48.
• Mattick, John S., and Marcel E. Dinger. "The extent of functionality in the human genome." The HUGO Journal 7 (2013): 2. doi:10.1186/1877-6566-7-2.
• Nóbrega, Marcelo A., Yiwen Zhu, Ingrid Plajzer-Frick, Veena Afzal, and Edward M. Rubin. "Megabase deletions of gene deserts result in viable mice." Nature 431 (2004): 988-993. doi:10.1038/nature03022.
• Palazzo, Alexander F., and T. Ryan Gregory. "The Case for Junk DNA." PLoS Genetics 10, no. 5 (2014): e1004351. doi:10.1371/journal.pgen.1004351.
• Pennisi, Elizabeth. "ENCODE Project Writes Eulogy for Junk DNA." Science 337, no. 6099 (2012): 1159-1161.
• Poliseno, Laura, Leonardo Salmena, Jiangwen Zhang, Brett Carver, et al. "A coding-independent function of gene and pseudogene mRNAs regulates tumour biology." Nature 465 (2010): 1033-1038. doi:10.1038/nature09144.
• Rands, Chris M., Stephen Meader, Chris P. Ponting, and Gerton Lunter. "8.2% of the Human Genome Is Constrained." PLoS Genetics 10, no. 7 (2014): e1004525. doi:10.1371/journal.pgen.1004525.
• Ward, Lucas D., and Manolis Kellis. "Evidence of Abundant Purifying Selection in Humans for Recently Acquired Regulatory Functions." Science 337 (2012): 1675-1678. doi:10.1126/science.1225057.

Sources from the tradition being discussed (Intelligent Design)

• Dembski, William A. "Science and Design." First Things 86 (October 1998).
• Luskin, Casey. "The ENCODE Embroilment," Parts II-IV. Salvo 32-34 (2015).
• Wells, Jonathan. The Myth of Junk DNA. Seattle: Discovery Institute Press, 2011.