Adherents say junk DNA hides an alien message; biologists still fight over how much of the genome has any function.
Two arguments meet under this title. One is a long scientific dispute over the roughly 98 percent of the human genome that does not code for protein: how much of it does anything. The other is the biomessage hypothesis of Vladimir shCherbak and Maxim Makukov, who argue that an intelligent signature was written into terrestrial life at its seeding. The idea of a message written into genomic DNA itself, those authors note, was raised soon after Francis Crick and Leslie Orgel proposed in 1973 that Earth was seeded, and was taken up in later SETI writing. shCherbak and Makukov place their own signal somewhere else, in the genetic code, and say a message in neutral DNA would not survive.
ENCODE's 80 percent, 2012
The ENCODE Project Consortium's integrated paper, published in Nature on 6 September 2012, describes 1,640 datasets across 147 cell types that map transcription, transcription factor binding, chromatin structure, and histone modification. It reports 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," and that these data let the authors assign biochemical functions for 80 percent of the genome. In the same paper, exons of protein-coding genes cover 2.94 percent of the genome, and protein-coding exons 1.22 percent, across 20,687 protein-coding genes.
ENCODE Project Consortium, "An integrated encyclopedia of DNA elements in the human genome": https://pmc.ncbi.nlm.nih.gov/articles/PMC3439153/
The objection from evolution, 2013
Dan Graur and colleagues at the University of Houston answered in Genome Biology and Evolution in February 2013. They wrote that ENCODE's claim that more than 80 percent of the genome is functional runs against estimates that less than 10 percent is conserved through purifying selection. They argued that ENCODE used a "causal role" definition of function rather than the "selected effect" definition most biologists use, under which only sequence shown to be under selection can be called functional. They also faulted ENCODE for favoring statistical sensitivity over specificity and for not separating "junk DNA" from "garbage DNA."
Graur et al., Genome Biology and Evolution, 2013: https://pmc.ncbi.nlm.nih.gov/articles/PMC3622293/
Three ways to measure function, 2014
Manolis Kellis and coauthors reviewed the dispute in the Proceedings of the National Academy of Sciences in April 2014. They wrote that nearly 99 percent of the roughly 3.3 billion nucleotides of the human genome do not code for proteins, and that biochemically active regions cover a much larger share of the genome than evolutionarily conserved ones, which raises the question of whether active but unconserved regions are truly functional. They noted that ENCODE maps of RNA cover more than 75 percent of the genome and transcription factor occupancy about 15 percent, while comparative studies find only 5 percent of mammalian genomes under strong constraint, and concluded that biochemical, evolutionary, and genetic approaches each add complementary information and must be used together.
Kellis et al., "Defining functional DNA elements in the human genome": https://pmc.ncbi.nlm.nih.gov/articles/PMC4035993/
What the non-coding majority is made of
Alexander Palazzo and T. Ryan Gregory argued in PLoS Genetics in May 2014 that the term "junk DNA" is usually credited to Susumu Ohno though it was in use as early as the 1960s, and that it is not true that functions for non-coding DNA were ignored until recently. They wrote that the 98 percent or so of the human genome that does not code for protein is generally well understood, with transposable elements making up between one-half and two-thirds of it by varying estimates, most of them inactive and degraded by mutation. They pointed to the onion test: the onion's genome is about five times the size of the human genome, a puzzle for any claim that most eukaryotic DNA is functional.
Palazzo and Gregory, "The Case for Junk DNA": https://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1004351
The biomessage claim, 2013
shCherbak and Makukov, in Icarus in 2013, proposed that if life on Earth was seeded on purpose, a signal in the genetic code would be a testable result. The genetic code maps codons to amino acids. Arranging that mapping by the nucleon counts of the amino acids, they report arithmetical and ideographical patterns, with sums divisible by 037 shown in decimal notation and a symbol of zero, and they say the null hypothesis that these patterns arose by chance with evolution is rejected with a P-value under 10 to the minus 13. They call the signal irreducible to any natural origin, and say the code stores non-biological information efficiently while staying nearly optimized for biology.
shCherbak and Makukov, "The 'Wow! signal' of the terrestrial genetic code": https://arxiv.org/abs/1303.6739
Why the authors look in the code, not in junk
Makukov and shCherbak returned in the International Journal of Astrobiology in 2017, testing what they call the we-are-them hypothesis. They write that genomic sequence is the most direct place to embed a message, but that a biologically neutral segment carrying one is unlikely to survive billions of years of evolution, and that cross-compared genomes show no segment without function shared by all or most terrestrial life. The only way to keep such an artifact, they argue, is to tie it to the most conserved function, the genetic code, frozen since the last universal common ancestor. Applied to the canonical code, they report a precision structure of interlocked attributes that known variant codes lack, which they read as post-seeding deviations. They acknowledge the mainstream view that the scenario is too speculative to test, and offer their method as an answer to it.
Makukov and shCherbak, "SETI in vivo: testing the we-are-them hypothesis": https://arxiv.org/abs/1707.03382
Where the record stands
How much non-coding DNA is functional remains argued in journals, with estimates from under 10 percent to 80 percent depending on the definition. The biomessage authors place their signature in the genetic code, not in non-coding DNA.