5 ms·

Huh, can we remove those? isn't there a cost to have all those extra sequences lying around?

by sudoaza 7y ago

bananamerica 7y ago
Of course, we’re on HN, someone will suggest a refactoring lol
- quickthrower2 7y ago
  As long as you have unit tests!
  - limomium 7y ago
    Any volunteers?
    - techbio 7y ago
      Use of this service constitutes acceptance of these terms.
Enginerrrd 7y ago
You ever see a cryptic line of code with a comment next to it saying something to the effect of: "No idea what this does, but everything breaks if we delete it"? It could be that old viral code has been repurposed long ago by evolution and serves a purpose.
- kibibyte 7y ago
  I generally get annoyed at most analogies that compare DNA to code because they oversimplify a lot, but I really do like this particular DNA-code analogy. My university molecular genetics course taught me that DNA is very complicated. Many of us know of the simplified model where an RNA polymerase, encouraged by some transcription factors, locates and binds to a promoter sequence, unwinds the strands, and constructs an RNA strand. However, DNA has a 3D structure. Histones can wind up DNA tightly enough to hide genes from transcription proteins, or they can unwind and expose DNA to encourage expression. Some transcription factors can bind to sequences several kilobases away from the target gene, cause that part of the DNA to fold onto itself, and ultimately bind to transcription proteins to encourage expression of the target gene.[0] Extending your comment, it is entirely possible that bits and pieces of old viral code have accidentally led to weird things like this, and that many organisms now depend on this behavior. [0]: https://en.wikipedia.org/wiki/Transcription_factor#/media/File:Transcription_Factors.svg https://en.wikipedia.org/wiki/Transcription_factor#/media/Fi... (In the course of reading about transcription factors, I came across something pretty neat that's related to all this about 3D structures: https://en.wikipedia.org/wiki/Transcription_factories https://en.wikipedia.org/wiki/Transcription_factories) Edit: I should also mention that, if it wasn't obvious, that viral DNA getting inserted into certain places can disrupt these complex interactions by increasing or decreasing the expression of a gene. For the cell/organism, this can be detrimental, benign, or perhaps beneficial.
  - sudoaza 7y ago
    I was worrying about this, specially your latest edit, when i mentioned cleaning up. We could limit to very new additions though just in case. I know that there's maybe no reason to think that this random additions are any worse than the random+selected pieces that were coding before, but just seems ugly somehow.
narag 7y ago
Is there an equivalent to zero-filled memory?
- tdy721 7y ago
  Sorta https://en.m.wikipedia.org/wiki/Non-coding_DNA https://en.m.wikipedia.org/wiki/Non-coding_DNA See we know about computers pretty good. But about DNA we get ignorant.
knome 7y ago
As I remember it, it was stated in "The Selfish Gene" that the vast bulk of genetic code served only a single purpose: ensuring it itself was replicated. It had no known function for the organism, it simply was copied from generation to generation. The primary "goal" of any self-replicator. Since it didn't actively interfere with the genes that do have an affect on the construction of the structures the genes build around themselves to protect and propagate those genes, namely us and any other organisms, the genes would simply continue being copied along indefinitely.
- brokenmachine 7y ago
  Forgive my naive question, but if the host is not affected if that code is lost, then wouldn't you imagine that eventually all that code would be destroyed by random mutation? There's no selection.
  - knome 7y ago
    The book really is a fantastic read. If you've any curiosity on the matter, I would suggest picking up a copy. The premise is that it is not organisms, but genes themselves which are the primary target of evolution. All but the gene is mere trappings that surround their competitive replication. Those genes that manage to slide from generation to generation aren't doing nothing, they are those that are most successful at duplicating themselves into the next generation. If they do not hinder the organism in their duplication, there is no pressure for their removal.
leaf_house 7y ago
There is a cost in terms of energy required to replicate the genome. The cost is relatively small though and grows slowly within the population so it is subject to fairly weak selection. On the other hand deleting DNA is always quite a risky proposition.
zby 7y ago
https://www.discovermagazine.com/planet-earth/mammals-made-by-viruses https://www.discovermagazine.com/planet-earth/mammals-made-b...

Huh, can we remove those? isn't there a cost to have all those extra sequences lying around?

bananamerica 7y ago

Of course, we’re on HN, someone will suggest a refactoring lol

quickthrower2 7y ago
As long as you have unit tests!
- limomium 7y ago
  Any volunteers?
  - techbio 7y ago
    Use of this service constitutes acceptance of these terms.

Enginerrrd 7y ago

You ever see a cryptic line of code with a comment next to it saying something to the effect of: "No idea what this does, but everything breaks if we delete it"? It could be that old viral code has been repurposed long ago by evolution and serves a purpose.

kibibyte 7y ago
I generally get annoyed at most analogies that compare DNA to code because they oversimplify a lot, but I really do like this particular DNA-code analogy. My university molecular genetics course taught me that DNA is very complicated. Many of us know of the simplified model where an RNA polymerase, encouraged by some transcription factors, locates and binds to a promoter sequence, unwinds the strands, and constructs an RNA strand. However, DNA has a 3D structure. Histones can wind up DNA tightly enough to hide genes from transcription proteins, or they can unwind and expose DNA to encourage expression. Some transcription factors can bind to sequences several kilobases away from the target gene, cause that part of the DNA to fold onto itself, and ultimately bind to transcription proteins to encourage expression of the target gene.[0] Extending your comment, it is entirely possible that bits and pieces of old viral code have accidentally led to weird things like this, and that many organisms now depend on this behavior. [0]: https://en.wikipedia.org/wiki/Transcription_factor#/media/File:Transcription_Factors.svg https://en.wikipedia.org/wiki/Transcription_factor#/media/Fi... (In the course of reading about transcription factors, I came across something pretty neat that's related to all this about 3D structures: https://en.wikipedia.org/wiki/Transcription_factories https://en.wikipedia.org/wiki/Transcription_factories) Edit: I should also mention that, if it wasn't obvious, that viral DNA getting inserted into certain places can disrupt these complex interactions by increasing or decreasing the expression of a gene. For the cell/organism, this can be detrimental, benign, or perhaps beneficial.
- sudoaza 7y ago
  I was worrying about this, specially your latest edit, when i mentioned cleaning up. We could limit to very new additions though just in case. I know that there's maybe no reason to think that this random additions are any worse than the random+selected pieces that were coding before, but just seems ugly somehow.

narag 7y ago

Is there an equivalent to zero-filled memory?

tdy721 7y ago
Sorta https://en.m.wikipedia.org/wiki/Non-coding_DNA https://en.m.wikipedia.org/wiki/Non-coding_DNA See we know about computers pretty good. But about DNA we get ignorant.

knome 7y ago

As I remember it, it was stated in "The Selfish Gene" that the vast bulk of genetic code served only a single purpose: ensuring it itself was replicated. It had no known function for the organism, it simply was copied from generation to generation. The primary "goal" of any self-replicator. Since it didn't actively interfere with the genes that do have an affect on the construction of the structures the genes build around themselves to protect and propagate those genes, namely us and any other organisms, the genes would simply continue being copied along indefinitely.

brokenmachine 7y ago
Forgive my naive question, but if the host is not affected if that code is lost, then wouldn't you imagine that eventually all that code would be destroyed by random mutation? There's no selection.
- knome 7y ago
  The book really is a fantastic read. If you've any curiosity on the matter, I would suggest picking up a copy. The premise is that it is not organisms, but genes themselves which are the primary target of evolution. All but the gene is mere trappings that surround their competitive replication. Those genes that manage to slide from generation to generation aren't doing nothing, they are those that are most successful at duplicating themselves into the next generation. If they do not hinder the organism in their duplication, there is no pressure for their removal.

leaf_house 7y ago

There is a cost in terms of energy required to replicate the genome. The cost is relatively small though and grows slowly within the population so it is subject to fairly weak selection. On the other hand deleting DNA is always quite a risky proposition.

zby 7y ago

https://www.discovermagazine.com/planet-earth/mammals-made-by-viruses https://www.discovermagazine.com/planet-earth/mammals-made-b...