Recently I came across a 2008 article, the authors of which argue that in fact the space of protein sequences is not as large as it might seem, and that life on Earth has most likely already explored this entire space (maybe before the appearance of eukaryotes).

Dear experts, can you explain what this means?

That life on Earth has already explored all the possible space of genomes, and all possible living organisms that are generally possible have already appeared? But after all, only the human genome consists of about 3 billion base pairs, and all possible combinations are a gigantic number, how could life explore them all, not to mention other genomes? And if life before eukaryotes tried every possible genome, then how did other life forms with a different genome come about?

I'm confused, please help.

  • 2
    $\begingroup$ The idea that all possible living organisms have already appeared is readily disproven, by for instance plant breeding. Just with flowers, we see things like artificially-induced polyploidy, and interspecies hybrids that only survive because of things like embryo rescue. WRT that 3 billion base pairs, note that in addition to the combinations, you can make it 3 billion +1 and get a new set, then repeat with 3 billion + 2 &c. $\endgroup$
    – jamesqf
    Commented Jun 14, 2021 at 16:19
  • 3
    $\begingroup$ The article argues about protein space, which is orders of magnitude smaller than genomic space. 1) There are 20 AAs for 64 codon sequences. 2) Proteins average 300-600 aa depending on the organism (source); note that this is much larger than your cited article's claim of <100aa, upon which their math is dependent 3) Only a fraction of most organisms' genome codes for proteins $\endgroup$ Commented Jun 14, 2021 at 17:13
  • $\begingroup$ @Punintended How many orders of magnitude is the space of proteins less than the space of genomes? $\endgroup$ Commented Jun 14, 2021 at 17:28
  • 1
    $\begingroup$ @ArmanArmenpress To compare the two, you'd need to make approximations for both. The amount of work in that task depends on your assumptions/approximations, and would likely be a paper (or two) in its own right. A constraint like "stability" will depend heavily on your precise definition, and since proteins and DNA have different chemical properties (and therefore stability), stability constraints on one will manifest quite differently in approximations of the other. Eg, "stability = 6 months at 4dC" will give very different results for different proteins, never mind comparing it to DNA $\endgroup$ Commented Jun 14, 2021 at 17:41
  • $\begingroup$ @Punintended Thank you. That is, it is not known what proportion of all possible genomes are stable. $\endgroup$ Commented Jun 14, 2021 at 17:45

1 Answer 1


The answer to this question is very much no. Genomes are of arbitrary length and structural organisation (i.e. ploidy), so there isn't a fixed number of 'slots' that different base-pairs could inhabit (like e.g. rolling a dice). As a consequence (and as jamesfq mentioned in the comments), no matter what genome you have, you can always add additional levels of complexity. Take for example the human genome which is approximately 3.1bn base pairs long; you could reconfigure it by adding or removing a base-pair (also known as indels) indefinitely.

That's not even considering the total number of combinations of base pairs at each position in the human genome, which is 4^(3.2e9), which is so astronomically large that it's not even worth entertaining that all those combinations could have occurred in nature.

  • $\begingroup$ There's a significant difference between "possible genomes" and "stable genomes", as the latter constrains protein sequences. As an easy example of the latter, for organisms following the central dogma of DNA -> RNA -> Protein, at minimum the genome needs to encode sequences for replication, transcription and translation $\endgroup$ Commented Jun 14, 2021 at 16:57
  • $\begingroup$ a) yes that makes it more clear, b) agreed. but the question didn't really specify (to my reading) whether or not the genomes were viable or not, but I'll edit that in to be clear. $\endgroup$
    – user438383
    Commented Jun 14, 2021 at 17:00
  • $\begingroup$ @Punintended That is, life on Earth has explored all possible stable genomes? $\endgroup$ Commented Jun 14, 2021 at 17:05
  • $\begingroup$ @user438383 Is it known how much of the genome can be functional and stable? $\endgroup$ Commented Jun 14, 2021 at 17:37
  • 1
    $\begingroup$ @ArmanArmenpress - you mean how many combinations of nucleotides end up being stable? Not sure myself, I think that would be a hard (but interesting) question to answer. $\endgroup$
    – user438383
    Commented Jun 14, 2021 at 17:50

You must log in to answer this question.

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