More Recent Comments

Monday, June 13, 2022

Manolis Kellis dismisses junk DNA

Manolis Kellis is a professor of computer science at the Massachusetts Institute of Technology (MIT). Sandwalk readers will remember him as one of the ENCODE leaders who participated in the massive publicity campaign of 2012 where they attempted to prove that most of the human genme is functional, not junk. He is the lead author of the semi-retraction that was published eighteen months later. [What did the ENCODE Consortium say in 2012 and 2014?]

Kellis was interviewed in April 2022 and it's interesting to hear his current views on junk DNA especially since MIT has just been rated the top university in the world for the 11th straight year. [QS ranks MIT the world’s No. 1 university for 2022-23].

His response to a question about junk DNA begins at 58 minutes. Kellis makes three points.

  • He doesn't like the word "junk."
  • Lots of noncoding DNA has known functions such as noncoding genes and regulatory sequences.
  • Half of our genome consists of transposon sequences and their regulatory regions fueled the mammalian radiation following the asteroid impact so that modern mammalian genomes now contain a complex and sophisticated network of regulatory sequences.

As I suspected, Kellis still doesn't recognize any of the evidence for junk DNA that was briefly outlined in the Kellis et al. (2014) paper. I find it surprising that after a decade of being exposed to criticism of his stance on junk DNA he is still not capable of presenting a cogent argument against junk.

Kellis, M. et al. (2014) Defining functional DNA elements in the human genome. Proc. Natl. Acad. Sci. (USA) April 24, 2014 published online [doi: 10.1073/pnas.1318948111]


  1. The willingness to concede that one was wrong is the basis of science and being a scientist. According to Lior Pachter Kellis is not open to corrections or any kind of critique:
    Others came to the same conclusion:

  2. It's often said that any discipline with the word "science" in its title isn't a science. "Computer science" certainly fits that description. And it definitely isn't biology.

  3. I wonder how many computer scientists would agree that any random string of characters is computer code due to the fact that we can find some strings of characters that are computer code? Kellis seems to be using the same logic with DNA.

  4. "Unable to sign in to comment. Please check your browser configurations to allow sign-in. Learn more." But "Learn more" leads to nothing relevant. What browser configurations? Help.

    1. I have the same problem when I use Firefox (and also few other problems). I installed Chrome and everything seems to work just fine.

  5. This comment has been removed by the author.