DNA is maybe 60-750MB of data
46 points
21 hours ago
| 6 comments
| dynomight.net
| HN
biomcgary
16 hours ago
[-]
Your genome only needs to store the information necessary for the lineage leading to you to survive and compete in the range of environmental variation that actually happened.

Consequently, your DNA has less information about how to survive on Jupiter or in the absence of oxygen.

However, your genome contains a fair amount of data on how to identify a mate that will maximize your reproductive success in an environment similar to the one your lineage experienced, e.g., a preference for symmetrical faces.

Until we can measure the environment of humans accurately, all the algorithmic complexity measures applied to the genome are going to be missing the relevant context.

reply
boshalfoshal
15 hours ago
[-]
The way I think about it, DNA is just a metaprogram. The "programs" this metaprogram create (brain, gut, other organs, etc). can be far more information dense and complex than the actual metaprogram that initially created them.

The real interesting part is _how_ this small metaprogram can generate something like a brain, which is ostensibly multiples more complex than the DNA that produced it, since obviously DNA cannot possibly encode the data for every possible synaptic connection or protein or whatever losslessly. I think this is more of a testament to how complex the human body is, that it has such complex seemingly emergent behavior from a very sparse set of initial conditions.

reply
ueueejdnsk
9 hours ago
[-]
This! DNA is like the ISA, the environment and biological processes created by it is the software and therefore dominates

DNA maxis exist because the idea of being genetically superior is deeply attractive to the privileged and powerful to justify their position independently of their circumstances of origin

reply
ueueejdnsk
9 hours ago
[-]
To add, obviously the hardware and ISA matter on a per-task basis and might provide some competitive advantages - eg sport, conventional/western beauty standards, some intellectual capabilities, but again, survival of your genes is dominated by birth circumstances
reply
rowanG077
7 hours ago
[-]
I'm not sure what DNA maxis means. But there certainly is precedent that some genetic code is superior for some environments than others. That's basically the entire argument evolution hinges on.
reply
ueueejdnsk
7 hours ago
[-]
that’s what I mean by ‘maxis’ — the people that rarely, if ever, acknowledge epigenetics.

They focus entirely on DNA / Darwinian evolutionary theory from centuries ago, ignoring environmental evolution, or to put it another way — evolution that happens as part of your lived experience opening the door to the heretical idea that they are where they are because of birth luck

reply
rowanG077
7 hours ago
[-]
Isn't "good" DNA the ultimate form of birth luck? An extreme example: down syndrome. No matter what you let the average person with down syndrome experience experience, they will never keep up intellectually. I still don't really understand what you are really saying here, and how epigenetics factor into it. I'm not an expert, but from what I understand epigenetics are essentially the sauce on top.
reply
ueueejdnsk
6 hours ago
[-]
No rich parents in a rich country are
reply
rowanG077
6 hours ago
[-]
So you'd take having down syndrome in the US with rich parents over being born with Paul Allen level intelligence to poor parents in the US? That's certainly an argument.
reply
ueueejdnsk
6 hours ago
[-]
That’s a straw man of my argument. Born with Down Syndrome in the US versus Paul Allen level intelligence in Gaza?
reply
rowanG077
4 hours ago
[-]
You didn't say war torn prison. You said not rich. Paul Allen in Rwanda vs down syndrome in the US. For sure I'd choose Paul Allen in Rwanda.
reply
ueueejdnsk
1 hour ago
[-]
You took an extreme position with Down Syndrome, and so it feels odd to take issue with an extreme position on country/geographic location

Would women excluded from higher education, or people who live in rural communities without access to much education at all likewise be unfair? How about tribes that live in forests, or other extremely isolated places?

I think taking extremes distracts from the majority of people the people were talking about but they do indicate something, surely. That environment can dominate

And so we’re back to arguing over degrees. My point is that many people disregard environment and epigenetics and in my opinion that’s an outdated and politically motivated view. See the other comment trying to bring in left/right categorisations.

How about we look at this from another angle. I think these are both broadly agreed scientifically: 1. DNA can have strong correlations with human attributes 2. Gene expression can differ based on the environment 3. The data encoded in DNA can’t encode the entire variability of a human’s biology by a significant margin

I think, taken together, it feels quite bizarre to conclude that DNA dominates the construction of a person.

It seems more plausible that DNA in combination with other biological processes affects the biological makeup of a person — epigenetic gene expression, diet, education, traumatic experiences, exercise, divergent access to resources and opportunities etc.

It feels ridiculous to not think that the environment can dominate, and so we’re simply talking about to what extent you can discount the environment.

Clearly you agree being in a war torn country is negative, and indeed being imprisoned. So really what we’re arguing over is a narrower definition — rich vs poor in a country like the US.

Even with that, the money you are born with isn’t encoded in your DNA and so how do we explain the correlation between rich family’s kids being more successful than poor families. Let’s restrict that to just in America.

If environment is dominated by DNA you’d have to believe the money you are born into is a function of your ancestral DNA, but that’s kinda of ludicrous because we know some family groups couldn’t amass wealth for the same period as others due to war and other violence excluding them from the same process, at least for the same period of time.

The argument that DNA dominates, to my mind, can only hold if you normalise the environment with a basic standard of living for all, as best you can. If DNA dominates then the same genetic winners will persist even if you cancel any environmental privilege.

It’s rare to see DNA maxis advocate for equality of opportunity, which lends further weight to the idea that even they don’t truly believe in their exceptional genetics

Long answer, but I find the Darwinian argument over used and under powered to explain anything relevant to modern society

reply
motrm
14 hours ago
[-]
DNA makes me think of ASN.1 in that a short sequence of bits can convey rather a lot of information - but it makes no sense without knowing which message those bits represent. Only with that knowledge can you turn those bits into something useful.
reply
hulitu
16 hours ago
[-]
> DNA is maybe 60-750MB of data

maybe

reply
waynesonfire
6 hours ago
[-]
I've scrolled and read through this wall of deep, philosophical conversations. People questioning the very essence of who we are, all hinging on how tiny our DNA is. Then, I encountered your comment and it's potent, capable of shutting down each of those philosophical threads.

And it made me laugh--only because it lives at the very bottom.

reply
jiggawatts
15 hours ago
[-]
For a long time now, a thought has been repeating in my mind: How exactly are high level behaviours like sexual attraction encoded in our genes!?

It such a subtle thing too! We're attracted (or not) to the tiniest differences in physiology. If you doubt this, try this exercise: Pretend you've just met green aliens and have to explain to them how to reliably tell the difference between men and women from appearance alone! Now explain why that particular girl (or boy) is very pretty/handsome, but not that one.

It's one of those topics where the more you know, the more freaky it is.

DNA does not -- to our knowledge -- directly encode the "weights" of our neurons! It can't possibly because there are far more synapses than there bits of information in our genes. Also, most of those genes are dedicated to non-brain parts of the body plan and to the low-level machinery of our cellular biochemistry.

Secondly, DNA has only an indirect effect of our development: it encodes for proteins, which then provide chemical signals such as concentration gradients that guide cell division. It's a bit like playing SimCity, where the players' control is limited to zoning and road topology. The individual Sims are not directly controllable and behave stochastically.

Solving this problem is so freakishly difficult for even the incredible brute force of parallel search of evolution only managed to discover a solution a few times in a billion years.

Our attraction to our partners is a genetic heritage shared with all mammals, going back hundreds of millions of years. That's why Furry is a thing, but not Featherry. Birds are a different class from us mammals and don't share the same "partner attraction wiring" genes. (This is closely related to why all mammal babies are cute to humans, but baby bird chicks are generally repulsive.)

Because this is a hard problem to solve, the few solutions that were discovered had to be reused by entire classes of Animalia. I would hazard a guess that this is precisely what defines a “class” in taxonomy! If there were intelligent birds, their equivalent of Furry would be Featherry, and their crimes of bestiality would be with other non-intelligent birds, not mammals.

With LLMs, we got to see a glimpse into the possible mechanisms of intelligence, and what it might take to design or evolve one.

The LLM equivalent of this kind of encoding would be to design a model architecture that falls in love with a specific, narrowly selected, subset of its users. Keep in mind that I'm not talking about a learned or specifically tuned set of model weights! The architecture is where the attraction is encoded, such as selecting some complex variant or combination of Transformers, Mamba, or CNNs that just "so happen" to result in the model preferentially learning to be attracted to certain styles of conversation, but not others.

Worse still, the direct equivalent to what genes do is that you can't even choose an architecture directly, instead you can only contribute to PyTorch. You have to design its API such that naive developers using it stochastically tend towards the desired architecture of their own accord by simply tab-completing often.

That's essentially what evolution figured out, at least five or six times, but tunable, so that individual species can be attracted to each other but much less to even very closely related species.

And then, evolution found a way to add a "notch filter" such that despite increased attraction to closely related individuals, most animals (including humans) are repulsed sexually by their parents and siblings.

That's mind-blowing to me.

reply
neuroticnews25
1 hour ago
[-]
Another thing is that evolution doesn't need to store all the information defining an organism in the DNA. It can also be passed down from mother to child during pregnancy, never touching the source code, "reflections on trusting trust" style.

Superinteligent alien given only the DNA code and tasked with reconstructing the organism probably would never be able to guess all the details of the bootstrapping environment right. I find it weird how little people talk about this.

reply
nly
14 hours ago
[-]
> And then, evolution found a way to add a "notch filter" such that despite increased attraction to closely related individuals, most animals (including humans) are repulsed sexually by their parents and siblings.

Not being attracted sexually to close relatives is likely the result of early imprinting when growing up under one roof, and not genetics. Indeed there is some evidence relatives separated at birth etc who meet later are more likely to unwittingly be attracted to one another.

reply
jiggawatts
13 hours ago
[-]
It's obviously imprinting, but the mechanism for that imprinting is implemented in your neurons, which are defined by your genes. Ultimately, it's a "feature" that wouldn't be there if it wasn't for something in your genes encoding for it!

The question is how is that encoded, not just in your genes, but in the final neurons?

There are hundreds of trillions of synapses in the brain that a few megabytes of genes somehow shaped into: "be attracted to faces almost, but not exactly like your parents"... but not actually repulsed, because you have to get along nicely with your parents and not just run away in terror.

reply
brap
2 hours ago
[-]
>That's why Furry is a thing, but not Featherry.

I did not want to read this

reply
E_Evan
16 hours ago
[-]
reply
refurb
12 hours ago
[-]
This article only touches on it but the data stored is far more than just the base pairs.

DNA is like a computer program that when it runs, it provides feedback for the code and determines which parts of it should run. It can also modify the code (DNA methylation).

Then add on top the external environment - external molecules can interact with the machinery which then impact which code is executed.

If code is self-regulating, the amount of information it encodes is far higher than that defined by its base pairs.

reply
flowerthoughts
7 hours ago
[-]
Another way of explaining what you said is that the information is encoded (only) in the base pairs, but it makes assumptions about the environment for correct decompression/interpretation. By correct I mean "the reason that piece of DNA was selected for."
reply