It's well known that genetic information is stored in DNA. As far as I know, DNA only has information at the protein level. What about higher levels, such as organelles, cells, tissue, organs? Is there any known carrier of information at that level? If not, what guides those levels of structure?

Hypothesis: Above the protein level, there is no real genotype/phenotype. Instead, it's all done through cell division. So, there is no split between information and embodiment. Instead, it's through a prototype-duplication model: cells grow and split, creating more cells. As for differentation into different types of cells, and organization into tissue and organs: the information and control systems for that are currently unknown.

  • 1
    $\begingroup$ Excellent question. Welcome to Biology.SE! $\endgroup$ Sep 16, 2012 at 11:42

1 Answer 1


What a timely question.

Does DNA contain information beyond protein synthesis?

Yes. In fact, protein-coding genes only constitute a tiny part – less than 2% – of the whole DNA. There are of course many other genes which aren’t protein coding: there are genes for ribosomal RNA and we find more and more genes which code for small RNAs, such as tRNA. But even if we count all those genes we won’t come above maybe 10% of the total DNA.

Most of the DNA is instead devoted to the regulation of gene expression, most importantly via the binding of transcription factors (but the picture is much more complex than that). With the conclusion of the ENCODE project, a whole slew of papers were published which show that in fact most of the DNA is actively implicated in the binding of various factors (although it’s not known how much of that actually contributed to the cell’s fitness).

But I’ve hijacked your question a bit here. So let’s come back to what you’re actually interested in:

What about higher levels, such as organelles, cells, tissue, organs?

There is no known mechanism (beyond the already mentioned regulation) which would encode such information in the DNA. Excluding its existence categorically may be hard but given that we haven’t found any machinery which would be necessary to read such information, we can be pretty confident that it doesn’t exist.

If not, what guides those levels of structure?

The higher levels of organisation are to all appearances emergent. That is, they are a consequence of the lower level organisation. For instance, take the cytoskeleton which carries much of the cell’s physical structure. It is composed of different protein complexes which form spontaneously through assembly of globular proteins (such as actin). The are several ways in which the process can be guided but strikingly it’s largely stochastic – i.e. mostly unguided, and it still succeeds in building a stable skeleton, simply by virtue of molecular properties encoded in the proteins.

I think this is a common theme of cell organisation: the elementary building blocks are encoded by the DNA, and their abundance is tightly regulated. Everything else, i.e. higher-level organisation – follows from that: abundance and localisation of the right proteins.

As for differentation into different types of cells, and organization into tissue and organs: the information and control systems for that are currently unknown.

In fact, much is known here, and it goes back to regulation on the level of DNA: we know that genes are differentially expressed depending on the cell type and stage of development (and the stage in the cell cycle). This regulation is highly complex and decoding it is a slow process. Nevertheless, the factors involved here are decoded one by one. This is the domain of developmental biology.

  • $\begingroup$ Thanks! Very helpful. What about my hypothesis that not all biological information is split between genotype/information carrier and phenotype/embodiment, but some actually is closer to prototype + copier? Specifically, organelles and cells are never created ex nihlo from any type of information but always split from a preexisting one? $\endgroup$ Sep 16, 2012 at 16:16
  • $\begingroup$ @S.RobertJames I don’t know. I would guess that most (though obviously not all, e.g. mitochondria / chloroplasts) organelles could also form de novo since their constituents are once again coded for in the nuclear genome and their assembly might again be either spontaneous or aided by other proteins. In fact, a friend just reminded me of things like centrioles which require a helper protein for assembly. $\endgroup$ Sep 16, 2012 at 16:55
  • 1
    $\begingroup$ It really isn't correct to interpret the ENCODE data as showing that most of the DNA in the human genome is "actively implicated in the binding of various factors". According to ENCODE scientist Ewan Birney only 8% of the DNA functions in binding factors. Almost 60% of the DNA is intronic, classed by ENCODE as functional simply because it is transcribed. genomeinformatician.blogspot.co.uk/2012/09/… $\endgroup$
    – Alan Boyd
    Sep 17, 2012 at 17:58
  • $\begingroup$ @Alan It absolutely is. The interaction is shown in vivo. The debate over whether this binding performs a function (and the definition thereof). Whether functional or not, it is, at some point or other, bound to – either by transcription factors or transcription machinery (i.e. transcribed). Ewan says exactly that in the blog post you linked, and he even qualifies his use of the word “functional”. $\endgroup$ Sep 17, 2012 at 18:38
  • 1
    $\begingroup$ Ewan Birney: "...60% of the genome with the new detailed manually reviewed (GenCode) annotation is either exonic or intronic..." As Konrad says, the generally quoted figure for coding (=exonic) DNA is "less than 2%" Therefore 58% of the genome is intronic: this is "almost 60%". $\endgroup$
    – Alan Boyd
    Sep 21, 2012 at 13:37

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .