What distinguishes "coding" from "noncoding" DNA?

Question

I've been reading a bit about "junk DNA" and how much of our genome consists of this "non coding DNA" in comparison to "coding DNA".

I'm just an interested layperson but I thought all combinations of three base pairs encoded one amino acid, with some amino acids being encoded by more than one combination of base pairs.

But if that were true then all of our DNA would encode something.

Or if only a tiny percentage of our DNA is "coding" that would mean that the vast majority of possible combinations of three base pairs don't represent any amino acid.

Or it could mean that there are a small number of "meaningless" combinations of three base pairs, but that those combinations are vastly overrepresented in our genome.

Which is correct? What am I missing?

Answer 1

You are correct in thinking that any sequence of bases corresponds, via the genetic code, to a sequence of amino acids. However not all stretches of DNA are actually transcribed into mRNA for translation into proteins. For this to happen the stretch of DNA requires (DNA-encoded) elements to promote and regulate the transcription and translation processes, and this, very broadly, is what defines a gene: a segment of DNA which has the required components to direct the synthesis of a protein (or in some cases an RNA that will not be translated into a protein). The segments of DNA that correspond to mRNA for proteins and to other RNA molecules are referred to as coding sequences.

Now, I've omitted lots of details here: some genes encode RNA molecules, such as ribosomal RNAs, which are not translated into proteins; eukaryotic genes include stretches of sequence (called introns) that are spliced out of the transcript before the mRNA>protein step. The original definition (if there ever was a definition) of "junk" DNA included these introns, as well as regions of DNA lying outside the coding sequences. We now know that there is useful information stored in much of this DNA, even though it doesn't code for anything directly via the genetic code.

Answer 2

You bring up a good point. "Coding" is a term that obviously carries some historical baggage that is gradually becoming less and less relevant. "Coding DNA" has typically been used to refer to DNA that encodes one or more functional protein products, which are constructed from an mRNA intermediate. As we've been learning over the last several years (and as was confirmed by the recent release of 30+ coordinated, high-profile publications from the ENCODE project), DNA that is not "coding DNA" is not "junk" or "meaningless"--it simply does not, to our knowledge, encode a protein. There is still a lot to learn about what precisely the function and purpose of this DNA is, but we do know that a lot of it is transcribed into RNA and that a lot of it has been associated with human disease.

Answer 3

DNA regions coding for proteins and RNA's comprise a rather small portion of total eucaryotic DNA.They are associated with special upstream signal sequences (promoters,enchancers,"boxes"etc) translation initiation triplets and translation termination signals ,which all together establish a reading frame provide information for the formation of the translation complex and the regulation of the whole translation process. Non coding DNA is by no means junk just because we do not understand it well enough yet.However it is known to play important roles in the formation of chromatin structure and its transformations which are crucial for exposing /hiding genes and their regulatory sequences to the proteins and cofactors working on them.They are characterized by large amounts of repetitive sequences and may also contain silenced ancient genes .