Can Darwinian Evolution Explain Lamarckism? | Hacker News Reader

Can Darwinian Evolution Explain Lamarckism?

quantamagazine.org

65 pointsErrorcod338 comments

This is oddly poignant to me. I just (literally a few hours ago) submitted my PhD thesis on (broadly) the biophysics of emergent phenomena. People don't usually read theses cover to cover, with good reason, but I included one short paragraph in the preface to be the true "take home" message:

We wish to understand mechanism through the elucidation of design principles, yet evolution does not select for principles, it selects for fitness, an epistatic and emergent property. If similar outcomes can be achieved in different but equivalently fit ways, then given the stochastic nature of evolution this is almost guaranteed to happen. We have specific examples where every statement in the preceding paragraph is true [ed: a collection of proposed mechanisms]. We do not need one person to be right or wrong; our nascent understanding of complex biological systems is that the space of information-processing solutions is astronomical. Think of the diversity observed in structural biology - the repertoire of tertiary structures is enormous. There are countless examples of nearly identical functions being performed by proteins with radically different structure.

This divergence, this variety in structure and function, is what makes evolution robust. It is an inherent bet-hedging mechanism woven into the fabric of statistical physics. On the contrary, the desire to categories and abstract complexity into distinct groups is an inherently human endeavour. Much as we may wish and as convenient as it would be, Nature does not have a plan.

Is your work available to read?

+1 - I'd like to see a copy too.

The thesis won't be publicly available for a while because there is fair amount of (currently) unpublished data that is associated with various collaborations.

+1 as well, very very interested. If you could post some resources that an interested reader could educate themselves with in the meantime, it would be deeply appreciated. Maybe a sample of the bibliography from your thesis?

So for some context, my work is not directly about evolution, but about how amino acid sequence determines function in the context of unfolded/disordered proteins.

That said, here are several at least semi-relevant papers that have influenced my thinking on a bunch of things (no particular order).

[1] Wheeler, L.C., Lim, S.A., Marqusee, S., and Harms, M.J. (2016). The thermostability and specificity of ancient proteins. Curr. Opin. Struct. Biol. 38, 37–43. (Probably paywalled but available on the Harms' lab website - https://harmslab.uoregon.edu/publications/. Mike's work on thinking about the biophysics of evolution is in general super cool. Similarly work by Adrian Serohijos is really interesting, although I am in general less familiar with it http://www.serohijoslab.org/publications.html)

[2] Tikhonov, M. (2016). Community-level cohesion without cooperation. Elife 5. (Open Access, really cool, and publishing a single-author original paper in a top journal in this day-and-age is incredibly impressive).

[3] Riback, J.A., Katanski, C.D., Kear-Scott, J.L., Pilipenko, E.V., Rojek, A.E., Sosnick, T.R., and Drummond, D.A. (2017). Stress-Triggered Phase Separation Is an Adaptive, Evolutionarily Tuned Response. Cell 168, 1028–1040.e19. (Paywalled, but IMO a HUGELY important study for thinking about 'aggregation' in the context of cellular fitness)

[4] Chakrabortee, S., Byers, J.S., Jones, S., Garcia, D.M., Bhullar, B., Chang, A., She, R., Lee, L., Fremin, B., Lindquist, S., et al. (2016). Intrinsically Disordered Proteins Drive Emergence and Inheritance of Biological Traits. Cell 167, 369–381.e12. (Paywalled, but potentially one of the most important discoveries in cellular adaptation in decades. More work to be done though!)

[5] Halabi, N., Rivoire, O., Leibler, S., and Ranganathan, R. (2009). Protein sectors: evolutionary units of three-dimensional structure. Cell 138, 774–786. (Paywalled, but super important for thinking about the relationship between local structural coupling and evolutionary behaviour. In general, everything Rama puts out is just gold.)

You've submitted this. I hope the spelling and grammatical errors aren't a problem for the readers. First of which is a missing 'the' in the first sentence. Later 'categories' instead of categorise.

Still, it seems interesting and I wish you all the best. I hope you are awarded your PhD.

Thank you - categories vs. categorise is a great spot!!! Where are the spelling mistakes? I don't see the missing 'the' in the first sentence? Mechanism here is in the abstract, if that's what you're referring to? That said, I've also not slept a lot in the last week...

FWIW, this is the submission to my committee, so I'm now spending the next few weeks meticulously going through to tighten up the grammar, catch typos etc., before it gets officially submitted in June.

Yes. I think he meant

"We wish to understand mechanism through "

should have been

"We wish to understand the mechanism through "

I had to pause on that spot too because it felt stilted when reading over it, though after reading it again I got what you were talking about and it read fine.

My bad on the missing 'the'. Mechanism is in the abstract. Of course it is. I see it now. Sorry about that.

Probably wise to mention your PhD here on HN. Nerds (apart from me) are the finest proof readers and grammar checkers the world has ever seen.

Seriously, all the best on your PhD submission. Sorry for the confusion.

Not at all - there's some additional text that provides set-up for the idea of mechanism. Re-reading it now cold, I read it exactly as you (and others) did as well. And I (embarrassingly) am relatively confident I wouldn't have caught the categories/categorise error, huge thank you!

> but equivalently fit ways, then given the stochastic nature of evolution this is almost guaranteed to happen

I would probably put different commas:

"but equivalently fit ways then, given the stochastic nature of evolution, this is almost guaranteed to happen"

I suggest this based on intuition alone, but I believe the principle is: http://www.grammar-monster.com/lessons/commas_for_parenthesi...

I don't understand how a huge collection of evidences of emergent phenomena proves that Nature does not have a plan.

It is unprovable, because you can always posit a plan that matches the actual outcome from what was actually a random process. What the evidence does do is to weaken the claim that there is evidence for a plan.

The plan is clearly to do anything to live and reproduce...You can see the innate struggle against increasing entropy in most living things, even as death is certain, because certainty is death.

This isn't proof of a plan, life does anything to live and reproduce because life without those qualities would not be sustainable. Those traits could be accidental though. The plan also isn't consistent when you step back, everything from planets to the universe do not seem to be following this plan.

Nature produce plans - nature produced you and you have plans...And if you have plans it's not too much of a stretch to say other living things have plans too, and their degree of planning varies depending on their level consciousness, memory and intelligence. The detail of planning is a continuum between non-life, virus, microbes, plants, animals, and humans. The intention in planning, however, is mostly constant.

Or both nature and me produce no plans, our actions are the result of some set of variables that have led to this point. This isn't proof.

And you're still focusing on too narrow of a spectrum, you can't even be sure that life must follow these rules just that it has in the one example we have.

Too true - in this life it's good to be not too sure of anything because often we are wrong about everything. In the end it doesn't really matter to our everyday lives if you or I believe one way or the other...

Believe is the key word here. This is impossible or nearly impossible to prove, so you can believe whatever idea you want. It just isn't science.

My personal beliefs are similar but I do expand it to the cosmic stage. I can't answer "why is there something rather than nothing?" But as soon as something exists, either through luck or some plan it does everything it can to ensure it doesn't go back to nothing.

Nature has a plan, it's called extinction. What is it over 90% of things which ever evolved have gone extinct. That's nature's plan to make things which are very, very, very likely to go extinct. Some may say that's it's a sieve, but if it's a sieve then why is the universe not filled with intelligent life, it's not a sieve, it's a death sentence. We're all made to die, eventually forever. Into the deep, dark night. All of us.

Is fitness emergent, or fundamental? If you didn't discuss this, you should add another year to your PhD!

"Fitness" is simply the name we give to the likelihood that an organism will have multiple generations of descendants.

This value of likelihood is an emergent phenomenon given an environment and an organism.

That's not deep enough for my tastes.

> the desire to categories and abstract complexity

categorize

(Please don't kill me)

Haha, man first time I opened my thesis, all fresh from the press, on a random page I saw an error. PhD students don't have editors haha, only professors who probably don't read back to back.

This is pretty straightforward. The existance of the epigenetic mechanism is, itself, a result of Darwinian evolution.

Where epigenetics seems weird to people is simply that it goes beyond the modern synthesis -- i.e. it's not just DNA base pairs. But plain-ol' Darwinian evolution doesn't depend on genetics via DNA base-pairs, it depends simply on inherited characteristics, however they are inherited.

I'm not a biologist. Can someone that knows more here help me out?

From the article it seems epigenetic mechanisms aren't inheritable i.e. they do not change the DNA just how the DNA is interpreted and/or which genes are activated. So even though epigenetic mechanisms might get passed on for several generations once the environmental factor affecting the mechanisms is removed the organisms would revert wouldn't they?

It seems all this does is allow some Lamarckian traits to act as a factors in the natural selection process.

Also the two explanations for how Giraffes got their long necks seem pretty ridiculous to me. Does anyone know of legit research into that topic?

What I find interesting is that methylation of cytosine, one common epigenetic modification, increases local mutation rates. Methylation typically reduces expression levels of nearby genes. A concomitant increase in local mutation rate may further disrupt transcription factor binding sites, also reducing expression. Thus, multi-generation methylation at a locus could increase the odds of genetically encoding the epigenetic information (assuming that the new selection pressure remains constant).

I've glossed over some details (e.g., somatic vs germ-line methylation patterns).

Epigenetic mechanisms are heritable. Just via other methods than DNA base-pair order. For example, through chromatin bookmarking, which effects the state of the DNA (which parts are tightly folded and which aren't), not the code itself:

https://en.wikipedia.org/wiki/Transgenerational_epigenetic_i...

https://en.wikipedia.org/wiki/Bookmarking

Gregory Bateson has a good piece in one of his books (it might have been "Steps to an Ecology of Mind") where he derives from cybernetic principals the result that Darwinian evolution works better than Lamarckian.

The key insight is that we can think of meta-evolution driving the form of the "first-order" evolution, meaning we evolved to evolve Darwinian-ly. (In practice meta-evolution cannot be separated from evolution, there is only one.)

This implies that we should expect limited Lamarckian evolution anywhere it leads to greater fitness than Darwinian.

From this POV epigenetic mechanisms are expected.

(I still look askance and the so-called "Central Dogma" of biology. Why on Earth would you assume that genes can't be changed by soma? If there was adaptive advantage to being able to edit DNA then it seems like cells and viruses would have it. As I typed that my brain said: CRISPR stoopid. SO, yeah, duh.)

> This implies that we should expect limited Lamarckian evolution anywhere it leads to greater fitness than Darwinian.

Not necessarily... evolution is good at finding local maxima, but not necessarily global ones if there's a deep valley in between. Transgenerational epigenetic inheritance is hard to evolve (compared to more Darwinian/DNA sequence-based evolution, it requires extra mechanistic steps and typically the effect size is small), so most of the time it isn't employed. If you look at enough cases where Lamarckian evolution might be advantageous, you'll find a few where it's employed to a limited degree, but in any given situation, it probably won't be.

Since biologists and evolutionists are liken to gather here, I would like to ask something that's been occurring to me lately: does evolution explain all the things? For example, does it explain my personal musical taste? Or perhaps the aesthetic tastes of those birds that mate based on aesthetic choices? If evolution doesn't explain these and other factors that are involved in natural selection, so we can say that external factors (explained by what?) are involved in the process of evolution, so evolution may be being partially caused by other, non-emergent, forces?

I don't know if this is clear or if it makes sense. Please feel free to say I'm stupid.

I am not an evolutionist or biologist to answer the first part, but when a theory(model) cannot explain some phenomenon does not imply external factor. It could also possible be caused by randomness.

Sort of. Here's a sloppy analogy off the top of my head:

Evolution explains your musical taste in the same way that gravity explains why much of the rainfall on the Alps ends up in Lake Constance. It's not a detailed, step-by-step explanation, but it's an explanation at the level of "duh, what else could it be?" Evolution, on a pretty fundamental level, is the reason for every last difference in behavior (down to the chemical level) between a bacterium and you, and no serious biologist doubts this.

Now, evolution is not an isolated process; it's exactly the connection between real-world "external factors" as you say and the survival (or not) of our ancestors. So sure, parts of your genome may have been shaped by the climate in Ethopia or the prevalence of mammoths in Siberia. All those stories are _part_ of "evolution."

I'm not a biologist but I can speculate on what one might say about your musical tastes: Humans and other mammals enjoy rhythms of certain frequencies, maybe because those are what you hear as a child nestled against your parent's chest. Hearing that kind of noise usually meant you were safe. Music in general is enjoyed as a form of social communication (which is a vital to human survival), roughly along the same lines that howling together strengthens the unity of a wolf pack. Beyond that, your individual music tastes, i.e. the tempi, rhythms, tones and harmonies you like, are most likely learned behaviors (and that's why they're often not inherited), influenced by what you heard a lot of in a certain age range. On the other hand, the learning capability itself, your acquisition of your surrounding culture and its musical features, is almost certainly another survival-enhancing evolved skill.

I often thinks 'Evolution' is a shifting goal post.

They just keep redefining it to suit new information.

Given by mass the majority of mammals (ok pretty specific) have been designed. I'm not sure it's currently true.

>They just keep redefining it to suit new information.

Also known as "the scientific method."

The nature is not set on supporting our positivist "yes or no" theories.

Which means, for every theory that says "A is causes by X and not by Y", there will be a few cases where the inverse is true.

The solution is probably to stop making a religion out of knowing a few true yes/no theories, and understand the why's.

Reply on news.ycombinator.com