The origins of life and the Fermi Paradox - why multicellular life is vanishingly hard

Inspired by Nick Lane's The Vital Question

So this review is inspired by Nick Lane’s The Vital Question, but I’m writing it probably 6 months after reading it, because it inspired so many questions on my end that it took that long for me to assemble some idea of what was probably going on.

And so, I will replicate that journey for you here, in this post - warning: it’s pretty nerdy. It’s got bacteria and volcanoes and Fermi estimates and who knows what else.

The starting point

So if you’re anything like me, your understanding of life’s probable origins starts with Darwin talking about warm ponds, and maybe the Miller-Urey experiment where they simulated lightning in a likely “early earth” atmosphere, and got amino acids out of it.

“Good enough!” you think, “billions of years is a long time, earth is a big place, the right environment probably existed somewhere for long enough to get everything started, so let’s not worry about it too much.”

But one of the things the genomics revolution showed us was that it actually couldn’t have worked this way, because early life was weird.

Profoundly weird - not in a “they eat volcanic emissions and crap methane or ammonia” way, that’s actually common and expected.

Weird like “the earliest LCA (least common ancestor) to eukaryotes and prokaryotes we can realistically imagine already had DNA, RNA, and protein chains, and incredibly complex metabolic machinery (literally machinery, ATP synthase, a molecular engine), but no cell membranes or cell walls, and how the HELL could that even happen?”

A literal rotating engine, but it rotates ~230 times per second - and it probably existed very near the beginnings of life!

So no, warm ponds actually wouldn’t work - the concentrations of carbon and hydrogen needed with the known (very poor to nonexistent) membrane quality of the earliest life makes it laughably impossible.

Similarly, lightning in atmospheres - the amount of energy needed to sustain the energetics for life in an ongoing way would require an absolutely absurd amount of lightning - Nick Lane:

“I once calculated that to sustain a tiny primitive biosphere, equivalent in size to that before the evolution of photosynthesis, by lightning alone, would require four bolts of lightning per second, for every square kilometre of ocean. And that’s assuming a modern efficiency of growth.”

And spoiler, modern aerobic metabolic efficiency is roughly 10x early chemosynthetic metabolic efficiency, so you can 10x that number again.

Perhaps more surprisingly, it couldn’t have been extremophiles in volcanic vents - the temperature is too high, the metabolic chains too specialized. The organisms that can survive near vents are extremely developed and specialized, and volcanic vent colonization only happened after developing membranes and cell walls.

So I’ll start with my high level conclusions after a lot of research:

The reasons multicellular life is rare

1. Even simple (prokaryotic) life requires very specialized environments that are ~100x - 1,000x rarer than typical volcanic vents and fields, so vulcanism alone is not enough (this is The Vital Question’s thesis)

2. Single cell life is bounded in size and complexity because all energy exchange happens at the cell wall, but as you get larger, your interior volume increases faster than your perimeter, and this limits how big and complex you can get - you need some paradigm shift change in energetics to get larger and more complex cells, and ultimately multicellular life (also Vital Question)

3. That step change in energetics (mitochondria in our case) is ridiculously hard to pull off, and probably also requires oxygen (outside research)

So, let’s take it piece by piece - what crazy environment COULD have fostered early life?

Life is weirder than we think - how to have DNA, proteins, ATP synthase, proton pumps and more, but no membranes or cell walls

So to make “replicating life forms” you need some sort of energy, eating and pooping, and templated reproduction.

Today, those things are done via the sun, digestive systems / ATP synthase, and DNA / RNA.

Surprisingly, the only difference when we look back at the LCA (least common ancestor) of prokaryotes and eukaryotes is that chemosynthesis, or metabolizing and growing and replicating with hydrogen and sulfur and iron, are the energy inputs rather than the sun.

Nick Lane makes the detailed case that to get from “inorganics” to “life” you need the following:

“So how could a cell be built from scratch? There must be a continuously high flux of reactive carbon and usable chemical energy, flowing past rudimentary catalysts that convert a modest proportion of that flux into new organics. This continuous flux must be constrained in some way that enables the accumulation of high concentrations of organics, including fatty acids, amino acids and nucleotides, without compromising the outflow of waste. Such a focusing of flow could be achieved by a natural channelling or compartmentalisation, which has the same effect as the channelling of flow in a water mill – it increases the force of a given flux in the absence of enzymes, so lowering the total amount of carbon and energy required. Only if the synthesis of new organics exceeds their rate of loss into the outside world, enabling their concentration, will they self-assemble into structures such as cell-like vesicles, RNA and proteins.”

Ah ha! Vulcanism and volcanic vents!

You’ve got high flux, you’ve got lots of methane and hydrogen and things like iron and sulfur that can serve as catalysts, you’ve got rocks with little channels, it sounds perfect!

But it actually isn’t - for one thing, the metabolic processes¹ going on in present day vents actually rely on oxygen, and the lifeless abiotic world didn’t have oxygen. There are some other theoretical chains that people thought might work (pyrite pulling), but they actually wouldn’t work (and no life grows around pyrite pulling processes).

Additionally, there are problems of temperature - the water at these vents is 200-400C, much higher than boiling temp, and organic synthesis can’t happen at those temps without a lot of adaptation. By the time the water is cool enough to use for organic synthesis, you’ll be farther away from the mineral catalysts, and they’ll be extremely diluted. Finally, volcanic vents are acidic, and relatively unstable - they form, collapse, and reform over decades, which is an eyeblink when it comes to evolution - not nearly enough time.

No, the real hero of the story are alkaline hydrothermal fields, which are ~100x - 1000x more rare than volcanic vents, and better in every respect for early life formation. The primary site we’ve studied is called Lost City.

“Unlike black smokers, alkaline vents have nothing to do with magma, and so are not found directly above the magma chambers at the spreading centres, but typically some miles away. They are not superheated, but warm, with temperatures of 60 to 90°C. They are not open chimneys, venting directly into the sea, but riddled with a labyrinth of interconnected micropores. And they are not acidic, but strongly alkaline. ”

The literal incubator for all life

We know of ~700+ white and black smokers (and remember, they form and reform over decades, so there’s been billions of them over time), and only one Lost City, with another potential 1 or 2 alkaline hydrothermal fields known in the entire world (Bay of Prony, Mariana Forearc).

We also think that alkaline hydrothermal fields might exist on Saturn’s moon Enceladus, so if we ever get a probe up there, we might find some single cell life.

Why are alkaline hydrothermal fields so great?

So Lost City is better for life formation for a lot of reasons - it’s got a lot of porosity and microtubules that have both alkaline and acid consistencies near each other, which can basically provide a hydrogen / proton gradient for free even if you don’t have a cell wall. Those microtubules concentrate inorganics massively, thousands or even millions of times more than the surrounding seawater. The temperature is just right for organic synthesis (60-90C). Lost City itself has been around and stable for at least 100k years, so there’s the time to actually evolve. There’s a decent amount of energy flow, a lot of potential catalytic minerals like iron and sulfur, and a ton of surface area to try out different environments and hyperparameters and iterate towards what actually works.

Basically, it’s the empyrean ideal in all respects for kickstarting life from inorganics - if you really want to get into the details, I heartily recommend the book, it’s a pretty fascinating read and he makes his case pretty well.

Broadly, he takes us through a story that all these good things allow life to kick start, develop RNA, develop proteins, ATP metabolics and ATP synthase, and stumble upon the one carbon fixation mechanism that’s energetically net positive in this environment (acetyl CoA), and that leads us to LUCA, or the least common ancestor of prokaryote and eukaryotes.

So that was our first hurdle - the places life formation can happen are much more rare than we thought (between 1 and 3 on our entire earth that we know about), and you’d need both water and vulcanism to get there.

But you know, there’s zillions of planets, there’s probably plenty with water, surely some of THOSE have vulcanism, and what’s a factor of only 100x - 1000x rarity between friends after that?? Still looking pretty good for life forming out there, right?

Why simple life may be vastly more common in the universe than complex life

Well, simple life is the easier lift. The basic divide is between energetics and complexity.

There’s a huge divide between prokaryotes (simple life) and eukaryotes (complex life) - prokaryotes focus on metabolic diversity and adaptability:

“There have been massive environmental upheavals in that time. The rise of oxygen in the air and oceans transformed environmental opportunities, but the bacteria remained unchanged. Glaciations on a global scale (snowball earths) must have pushed ecosystems to the brink of collapse, yet bacteria remained unchanged. The Cambrian explosion conjured up animals – pastures new for bacteria to exploit. Through our human prism, we tend to see bacteria mainly as pathogens, but the agents of disease are a mere tip of prokaryotic diversity. Yet throughout these shifts, the bacteria remained resolutely bacterial. Never did they give rise to something as large and complex as a flea. Nothing is more conservative than a bacterium.”

Broadly, they can adapt and fill any niche by changing how they metabolize, and they’re really good at it in the aggregate - surviving major environmental changes, upheavals, and more.

Eukaryotes focus on complexity and higher energy density and double down on morphological diversity, and so lead to all plants, animals, and other multicellular life.

“structural constraint enabled the eukaryotes alone to explore the realm of morphological variation. In the broadest of terms, prokaryotes explored the possibilities of metabolism, finding ingenious solutions to the most arcane chemical challenges, while eukaryotes turned their back on this chemical cleverness, and explored instead the untapped potential of larger size and greater structural complexity.”

But the step between them is an incomprehensible gulf - giving up your cell membrane and becoming symbiotic in a long-term sustainable and net-energy-positive way (eventually leading to mitochondria and the much better energetics that allow complexity) was a big step that was seemingly never repeated in the ~4B years since prokaryotes have been around, in the sense that we don’t see any evidence of different “lines” of eukaryotes anywhere in the world - all eukaryotes go back to a singular endosymbiosis event.

What I originally thought:

So I originally thought that the endosymbiosis event was driven by one bacteria engulfing and “eating” another bacteria - phagocytosis. Now, we’re all (literal) animals, so we think eating stuff is just the default. But it’s really not - we think that because we’re from a line of beings that eat stuff. But back when things were simple, practically nobody ate anybody.

Why is that? Why was “eating” so hard to invent?

Broadly, to specialize in “phagocytosis” and to do it regularly you need pretty complex machinery, both externally and internally. It requires a flexible membrane with a complex cytoskeleton, and most prokaryotes have a rigid cell wall. It requires you to have “digestive apparatus,” whether that’s enzymes or something else.

You also need to be BIG - not just for “throw your weight around” reasons, but for raw logistics reasons (you need to engulf your prey). But most prokaryotes are limited in complexity and size because size is energetically expensive, and because all prokaryote “energetic engines” rely on boundary-level proton pumps on their cell walls / outsides. This means you can’t get too big, because there’s a surface area to volume problem - your volume always increases more than your surface area, but you need enough energy to cart your whole self around, so having your energetic engines limited to your surface area limits how big you can get.

And when reaching up for that greater complexity, prokaryotes generally replicate via ring-shaped DNA with a single “replication point” on it (vs the straight DNA with multiple replication points that we use). This means there is continual selection pressure for smaller and simpler genomes, because as you grow and complexify, it will take longer and longer to unravel your ring-DNA from the single replication point and transcribe and replicate it, and mistakes become more likely.

So we would have needed an odd bird of a prokaryote to specialize in phagocytosis - just at the threshold of how big and complex a prokaryote can get, who then ate just the right smaller bacteria.

And it gets worse and more complex from there - the Big Lift

So bacteria-on-bacteria phagocytosis isn’t that rare, we’ve seen it happen multiple other times with prokaryotes. Aside from daptobacter, which is eating but isn’t proper engulfing, we’ve seen actual engulfment events several times.

This was part of my problem when reading The Vital Question. I’m sputtering my toast and coffee, like “bruh! Phagocytosis isn’t even hard, and there’s tons of evolutionary gradients pushing towards it! Of course if you’re surrounded by other life forms and can turn them into “food” somehow, evolution wants you to exist in that all-you-can-eat buffet! And we know of prokaryotic examples today, much less eukaryotic examples, much less over deep time! This just isn’t that hard a lift!”

I mean, just think of the ecological niche! It’s why animals eat each other - sure, you can specialize in finding rich deposits of hydrogen and CO2, and put your head down and really toil at cranking out that meager ~4% metabolic edge,² but ultimately isn’t it tempting to look up and notice you’re in a whole bacterial mat of 4%-toilers, and if you just figured out how to eat them instead, you’d have infinite free food for way less effort?

But (unsurprisingly, given we haven’t empirically seen prokaryote endosymbiosis happen more than once in ~4B years) this probably didn’t work out that way. A few more things need to happen to get to eukaryotes, and from there to multicellular complex life, and those are more likely on a different path:

1. There needs to be an immediate benefit to both organisms

2. Some of the engulfed organism’s functions needs to migrate to the host’s nuclear DNA

3. Energetics need to become net favorable to the host

4. A postal system with labeling and input / output gates needs to be built (TIM / TOM)

It actually turns out that eukaryotes probably originated via syntrophy (different organisms existing in symbiosis) vs trying to eat each other.

Let’s take a look.

The immediate benefit

So it seems a lot more likely that we started with two already-buddies, than with a bigger guy eating a smaller guy and it turning into an odd-couple comedy. This is because when you’re eating the smaller guy, you’re probably not allowing enough immediate benefit to enter into a deeper-in-time relationship to evolve the rest of the steps you need.

So what if instead of eating your smaller organism, both of you got mutual benefit out of the deal, because you coexist in the same environment, and one of you produces hydrogen as waste, and the other one uses that hydrogen as an energetic input, and then does something good in return? Sounds great, circle of life, etc.

For our future eukaryote, the genesis of their partnership was the Great Oxygenation, that poisonous upwelling of oxygen-based respiration and metabolism, deadly to all anaerobes.

We now think that a larger anaerobic Asgardian Archaeon³ became so syntrophic with a smaller alphaproteobacterium that they basically merged. We think this, because the only Asgardian Archaeon we’ve been able to culture in a lab does this naturally (and this actually took 11 years to figure out, anaerobes are hard mode).

These guys literally reach out with pseudopoda and grab other bacteria, and start syntrophic relationships where they share resources and exchange inputs / outputs. They’re the original super-friendly “this guy” who’d give you the shirt (or at least the hydrogen) off their back.

I’m going to call the bigger Asgardian Archaeon Odin (because of course) and the smaller alphaproteobacteriaum Njord (Norse god of the sea, to stay on our alkaline hydrothermal theme) from here on, for simplification.

Our single celled progenitors, Odin and Njord

So Odin was anaerobic, and Njord was an oxygen metabolizer, so there was indeed an immediate benefit - the poisonous stuff that hurts Odin and makes life harder was suddenly less poisonous, because Njord is eating it up and producing tasty byproducts. Less poison, more food. Njord also consumes the hydrogen that would build up as waste around Odin, and may have even returned some higher quality carbon byproducts usable for carbon fixation. Maybe we should keep this guy around!

And Njord got a benefit, too - it gets free hydrogen and small organics from its host, oxygen to metabolize, and a nice safe space away from it all.

Best of all, together they were able to go to areas that were hostile to both aerobic and anaerobic life forms. Because they can navigate and metabolize both sides of the equation, there were likely a lot of resource-rich areas that had mixed conditions, all of which were an entire ecological greenfield that no other organism could exploit. All you can eat buffets, as far as the eye can see!

That’s SYNERGY, son!

And not just the kind that bad bosses write in big letters and underline twice on a whiteboard, I mean the kind that can actually make you rich and feed your family for umpteen generations!

Internal DNA simplification and migration

The next step that needs to happen for true symbiosis is DNA migration - intertwined fates. As long as Njord has its own fully functional DNA, it could turn into a parasite, and consume or multiply at the expense of its host, if the environment shifted to more oxygenated, as one example. And right now, Njord actually takes energy from Odin, it’s not contributing, and could evolve to take even more.

Bacteria in general are always trying to simplify, because Muller’s Ratchet continually pushes them towards the simplest possible DNA configuration that still works and can reproduce.⁴ So slowly, what happens is that a protein coding for something in Njord gets a mutation and doesn’t work any more, but host Odin has a working copy, and the host’s copy takes up the slack.

This intertwines their fates - now the engulfed organism is truly along for the ride, and can’t engage in parasitic dynamics or leave. Over time more DNA can migrate to Odin’s nucleus, allowing specialization, and this is actually a more stable and efficient arrangement over deep time too, rather than having an internal organism duplicating functionality and complexity entirely, and being prone to degradation / mutation.

Now, the big lift - input / output mechanisms

You may have been wondering back there - “Wait, how does that actually work? If the host’s protein is in the host’s cytoplasm, how on earth does it get through Njord’s cell wall so it can function inside Njord?”

That is a great question, and brings us to the biggest lift, the change that was the hardest and most finicky to make, and which allowed the massive, orders-of-magnitude change in energetics that allowed for eukaryotes and multicellular life to happen at all!

So throughout this story, Njord has basically been a parasite, energetically. He is definitively NOT “the powerhouse of the cell” yet.

Sure, he depoisons the oxygen for Odin, and allows them to graze in environments rich in resources and low in competition, but that’s the only way he’s paying for his ride. It was good enough for both of them, obviously - it stuck around long enough for more to happen.

Two more things now had to happen.

First, an ADP / ATP translocase gate had to get installed. This likely happened via normal bacterial horizontal gene transfer (the AAT gene⁵). Broadly, once this was installed in Njord, then Odin could give Njord ADP and hydrogen and phosphates, and Njord could metabolize that aerobically into ATP and give the ATP back to Odin. And ATP has more energy than ADP (three phosphates versus two in ADP). Odin just got free energy out of waste products! It’s like your trash powering your air conditioner!

The Odin + Njord partnership just got strongly net beneficial.

Now remember all that stuff about limitations on complexity and size, due to bacteria processing all energy at cell walls? THIS was the defining revolutionary adaptation that gave us true eukaryotic potential and allowed us tens to hundreds of thousands of times the energetic density, and opened up a world of complexity, size, and multi-cellularity. This is the defining “eukaryotic moment.”

So now there’s a really strong partnership and things are going great. They’re kicking ass and taking names and chewing metaphorical bubblegum as they expand at the expense of weaker and less energetic lifeforms.

But what if Njord loses an internal protein because of Muller’s Ratchet? And this WILL happen over time - it’s something you need to address.

Even if Odin makes lots of that protein and is happy to send them Njord’s way, you need some sort of “postal service” function to shuttle Njord the needed proteins, and the waste / energetic outputs to Odin, and you need some sort of labeling to make sure the right things are going to the right places, and you need both inner and outer cell wall gate complexes in Njord to pull it off!

This one is probably the reason that it only happened once in ~4B years.

Because it’s irreducibly complex, and several adaptations / exaptations have to happen basically all at once (in geological time) for it to work out.

Now they both had to stumble upon proteins that could coincidentally form a primitive channel, others that could act as receptors, and still others that could provide a targeting signal.

This gradually evolves into the TIM / TOM inner/ outer gate complex and signaling system that current mitochondria have.

• Targeting Signals: Each protein destined for the mitochondrion needs a specific "address label" or targeting sequence

• Translocase Complexes (TOM and TIM): A suite of multi-protein complexes had to evolve in the outer (TOM - Translocase of the Outer Membrane) and inner (TIM - Translocase of the Inner Membrane) mitochondrial membranes. These act as gates and channels, recognizing the protein's address label and guiding it to its correct location within the organelle

This is basically verging on the paradox of the watchmaker, as near as I can tell, because so many things had to go JUUUUSSST right for it to even make it. It’s a big jump on the evolutionary landscape, and seemingly all at once.

It’s a very lucky thing the AAT mutation had happened, because it’s in a similar family that is used in the TIM gate complex, and was probably exapted to create the TIM gate. The TOM complex was likely exapted from Njord’s existing beta barrel protein already being used on Njord’s outside membrane in his career as a separate alphaproteobacterium. But it was definitely a skin of the teeth sort of thing.

Usually, if Njord loses a protein due to Muller’s Ratchet, it just dies. And then the host Odin probably dies, too, poisoned by oxygen. And this probably happened 10^^bignum times, until somehow, somewhere, there was an Odin / Njord pair with just the right combination of existing protein signaling and gate complex precursors that could be exapted quickly enough into something like the TOM/TIM signaling system mitochondria have today.

But if you’re able to roll that once-in-4B-years Nat 20 and pull it off? You’re in gravytown, baby!

Why eukaryotes are awesome

The key insight from Lane is that the barrier to greater cellular complexity isn't simply about physical size—it's about the energy-per-gene needed to build and maintain complex cellular machinery. Without the energy boost from specialized organelles like mitochondria, even giant prokaryotes face severe limitations in genomic and functional complexity.

What makes eukaryotes special isn't their size per se, but their energy-per-gene ratio that allows them to maintain thousands of new protein families that enable complex cellular behaviors in the fields of energetics, metabolics, cell specialization, multi-cellularity, and more.

“On average, bacteria have around 5,000 genes, eukaryotes have about 20,000, ranging up to 40,000 in the case of large protozoa, like the familiar pond-dwelling paramecium (which has twice as many genes as we do). The average eukaryote has 1,200 times as much energy per gene as the average prokaryote.”

“If we correct for the number of genes by scaling up the bacterial genome of 5,000 genes to a eukaryote-sized genome of 20,000 genes, the bacterial energy-per-gene falls to nearly 5,000 times less than the average eukaryote. In other words, eukaryotes can support a genome 5,000 times larger than bacteria, or alternatively, they could spend 5,000 times more ATP on expressing each gene, for example by producing many more copies of each protein; or a mixture of the two, which is in fact the case.”

So once you get there, you’ve cracked the code - not only can you scale on both complexity and size, but now you can diversify morphologically.

Prokaryotes are haikus, eukaryotes are everything else - from sonnets to long form literature

The surface-area-to-volume problem has been flipped and now it’s an advantage - you have exponentially more area for energy production now - a typical eukaryotic cell can generate way more energy per second than a bacterium of similar size.

“scaling up a bacterium to the size of an average eukaryote increases ATP synthesis by 625-fold, but increases the energy costs by up to 15,000-fold. The energy available per single copy of each gene must fall 25-fold. Multiply that by the 5,000-fold difference in energy per gene (after correcting for genome size), and we see that equalising for both genome size and cell volume means that giant bacteria have 125,000 times less energy per gene than eukaryotes. That’s an average eukaryote.

Large eukaryotes such as amoebae have more than 200,000 times the energy per gene than a giant scaled-up bacterium. That’s where our number came from.”

In addition to the much more favorable energetics, numbers of genes, and complexity, you also get “economies of scale” from offloading most of the mitochondrial genes to the central nucleus. Modern mitochondria have <40 genes, Njord would have had 1-2k genes. That gives you more efficient gene regulation, lower replication costs, and lower error rates. And you get the division of labor, where mitochondria can specialize in energy production while other cellular components handle different functions. This specialization also supports greater cellular complexity.

Ultimately, of course, eukaryotes are how we get multicellular life, with different types of cell specializing, then aggregates of types of cells specializing, and finally creating plants, insects, animals, mammals, and everything else.

So broadly, here’s the Fermi funnel, as I understand and SWAG it:

1. You need water and vulcanism, but a special 100 - 1,000x rarer type of vulcanism, to get single celled life to begin with. Let’s call this a 100x discount, 10^^-2

2. You (probably) need oxygen for complex life, because chemosynthesis is literally 10x less energetically efficient than respiration, in terms of cellular energy gained (~4% vs ~40%), and we have currently found zero rocky planets with oxygen in the atmosphere (55 Cancri e might be the first one, though). We’ve only been able to observe a few dozen to this accuracy, so let’s be optimistic and just call this another 100x discount, 10^^-2

3. You need some sort of symbiosis, cooperation, or mutually-beneficial phagocytosis to unfold and be beneficial at each deep-time step to reach complex life. This is a mix - phagocytosis is probably around 1 / 100B, but syntrophy is pretty common in the early life environment, say 1-10% or so, so let’s be generous call it a 10% discount, 10^^-1

4. You need AAT horizontal gene transfer with a syntrophic pair such that the engulfed one starts being energetically net positive: let’s say 10^^-2 over deep time⁶

5. You need to cobble together a TIM / TOM gate complex and postal system to shuttle the right things in and out: seems like the hardest one, earlier I estimated 10^^bignum so let’s call bignum 14, for kicks, because 14 is big.⁷ So 10^-14

Fermi equation says:

Let’s say tens of billions of rocky planets with vulcanism in Milky Way: 50B = 5e¹⁰

Now we go through our discounts, 1-5, and see how many we have left:

• Alkaline hydrothermal fields: 5e⁸

• Oxygen: 5e⁶

• Syntrophy: 5e⁵

• Successful AAT transfer, function, and fixation: 5e³

• TIM/TOM and signaling: 5e^-11 - whoops. So even if 14 is a crazy power, and it was actually merely one in hundred million or a billion (10^^-(8-9)) or something because there was a million times more anything, the odds look pretty bad just from a “this planet probably exists” sort of standpoint - basically one in a million odds of getting just multicellular complex life (because there’s discounts from there to “intelligence!”), and that’s starting with 50B planets already starting with water and vulcanism

• You can actually triangulate the odds of conjoined points 3-5 from the other side, too, and reach similar numbers.⁸ We’ve actually seen successful engulfment, net positive energetics, and genetic translocation to the host happen empirically about once every billion years. 2B years ago it was Odin + Njord giving us eukaryotes, then 1B years ago, a eukaryote absorbed a cyanobacteria and turned it into chloroplasts / photosynthesis and all plant life, and about 100M years ago a Paulinella amoeba ate a different cyanobacteria and is photosynthetic and 20-30% of the way through relocating the engulfed’s DNA into the central DNA. I would like to point out - eukaryotes engulfing other bacteria is much easier than prokaryotes doing this, for size and complexity reasons. The original Odin + Njord pairing was still amazingly lucky / rare. And even for the later successful eukaryotic engulfments, think how many organisms this represents, over what amount of time, and how many “shots on goal” that translates to in terms of the odds of success! I do the math in footnote 7 linked above, but it’s basically 1e^-34 odds or thereabouts, roughly in line with what I estimated the other way. Long, loooooonnnng odds to reach “complex life,” requiring lots of shots on goal, oxygen, water, and untold numbers of organisms over deep time.

And of course, lots of these are pretty speculative. What if you need oxygen, but it’s actually a lot more than 1/100 rare? We really don’t have an idea there, and I think there’s a good chance that if you were stuck with chemosynthesis you wouldn’t ever attain the energetics to achieve complex life.

Fermi paradox (at least in terms of complex, multicellular / intelligent life) directionally explained?

You decide.

Mea culpa

All of this AAT horizontal transfer and TIM/TOM stuff and crazy Fermi estimates are more or less the picture I’ve come up with from researching various things over the last few months, not Nick Lane’s, and it could definitely be wrong.

Ultimately, my knowledge of microbiology is fairly limited - I did do microbio research (in cyanobacteria, at that), and published in the field many moons ago, but that just makes it even more likely I’ve misunderstood or misstated something, to my mind. So for any real bio scientist in the audience (Stetson? Metaculus? halvorz?), if I got anything wrong here, please let me know.

What’s in The Vital Question that’s worth reading yourself

1. A lot of fun triangulation around energetics and chemical concentrations and gradients, and why various environments and scenarios for life formation wouldn’t work

2. A deep dive into specifically why and how alkaline hydrothermal fields are an ideal environment for generating single celled life

3. Really interesting deep dives into the surprisingly negligible differences in entropy between different biological states - like a functional spore and one that’s been ground up have basically the same entropy, and a useless form of a protein and useful one, likewise

4. Lots of flavor on eukaryotic complexity and energetics and how great eukaryotes are

1. Hydrogen sulfide reduction, donating one electron to oxygen, to drive ATP synthesis

2. Astoundingly, chemosynthetic and thermophilic bacterial metabolization is something like only 4% efficient, if you’re counting the energetics of the ATP they generate to power cellular activities. Pretty much everything else that respires with oxygen gets 10 times as much efficiency, generally about 40%.

3. Which incidentally, I feel a swell of pride that the literal origin of life is so nominatively metal af - like coming from an “Asgardian Archaeon” just SOUNDS impossibly epic, a fittingly majestic beginning for complex life as we know it

4. And this leads to fun dynamics like a broader “exaptome” accessible via horizontal gene transfer - for instance, a given E. coli organism has ~4k genes. But the metagenome across all E. Coli features ~18k distinct genes, accessible via horizontal gene transfer between different organisms. It’s like having a library of skills and capabilities that you can freely install from any neighbor you meet, it’s a pretty cool concept.

5. Which AAT gene possibly came from the Chlamydiae family, so if so, I guess we can thank chlamydia for once?

6. AAT Fermi estimate:

• 10^^{12 total cells per cubic meter, but there’s only 20-200k usable cubic meters bc we’re still in early life time periods}

• ^{say 1/1M are a hybrid archaeon / alphaproteobacteria, so you’re down to 10^6, and they interact with a potential AAT donor once per year, so you’re at 10^6 per year.}

  ^{Now the discounts:}

• ^{1/1M interactions lead to transfer – 10^-6}

• ^{Has to be integrated 1/10k = 10^-4}

• ^{Has to be expressed 1/1k = 10^-3}

• ^{Newly synthesized protein has to be directed to inner membrane, so an additional step, call it 1/100k = 10^-5}

• ^{Fixation – 50% chance}

• ^{So multiply all that out and you get 50% * 10^-18}

• ^{But we have a million cells cranking on it, for lets say a million years over 20k cubic meters (so 6+6+4), to give us enough time to do the big TIM/TOM lift – that brings us down to 10^-2 odds of it happening.}

^{We’ve got a few billions years to play with overall, so that’s not so bad. If we run just “AAT fixating” for a hundred million years, we’re at 1:1 odds! Not too shabby! It works if you squint!}

^{But then you’ve still got to pull the whole TIM/TOM system out of your pocket. Womp womp.}

7. Which honestly, just thinking of the complexity of all the steps that have to line up and basically all be functional at once, is very much giving me Watchmaker Paradox feels. If you think about this, you can definitely see why this one is the big lift and phenomenally unlikely:

TIM / TOM Fermi estimate

• Say we’ve run for a hundred million years, and have a good chance of having a fixated AAT syntrophic pair. It’d probably be phenomenally successful because of it’s superior energetics and environmental capabilities, so let’s call it 2% of the total bacterial sample. So 5e¹⁰ per cubic meter of seawater, and let’s upgrade our available environment and go with 200k cubic meters of usable seawater. We’re starting from a great place!

• You need to make some sort of cell signaling / tagging to direct needed proteins to the mitochondria, and to signal what things are needed: on the way in, this is a 20-50 amino acid signal appended to any needed mitochondrial proteins today, and it needs to be recognized by both the outer and inner gates, and then cut off by a peptidase within the mitochondria to actually be used. On the way out, it’s usually a bunch of redox products that inspire a whole command center of evaluation and machinery in Odin to evaluate what’s needed, start up synthesis, tag the things, and get them out to the cytoplasm - super complex: 10^^-12

• You need to develop a TIM gate on the inner membrane, consisting of at least 12 distinct proteins and 3 distinct aggregate subunits working together to escort pre-proteins in and shear the tags off - super complex again: 10^^-12

• You need to develop a TOM gate on the outer membrane - “The complete mitochondrial protein translocase complex includes at least 19 proteins: several chaperones, four proteins of the outer membrane translocase (Tom) import receptor, five proteins of the Tom channel complex, five proteins of the inner membrane translocase (Tim) and three "motor" proteins.” These are exceedingly complex molecular machines at this point, and if you tasked a biologist with building them, they’d look at you like you were crazy. But you know, somehow one of these literally brainless bacterial hybrids actually did build an MVP version of all three of these, through pluck and luck and sheer derring-do: 10^^-12

• Now the whole system needs to work together well enough to function and provide benefit reliably and long term and ultimately fixate 10^^-2

• Multiply all that out, and you get 10^^-38 odds

• How many years do we have to run towards it? About 2B if we’re generous, 2e⁹.

• We’ve got 2e⁵ cubic meters and in those cubic meters, 5e¹⁰ per cm^3

• So we’re down to 38-9-10-5 = 10^^-14 which is roughly 1 / 100 trillion odds.

• So even if we’re cumulatively off by a factor of a million on any combination of these - the amount of usable seawater, how hard it is to do any of the TIM/TOM steps, and so on, we’re still looking at 1/100M to 1/1B odds of any one of the 50B planets in our galaxy having multicellular life

• Outer triangulation - this happens successfully about once every billion years - 2B years ago it was Odin and Njord, 1B years ago it was a eukaryote that absorbed a cyanobacteria that turned into chloroplasts and became photosynthetic, giving birth to all plant life, and about 100M years ago, a Paulinella amoeba did the same trick with a different cyanobacteria, and is about 20% of the way through the process of integrating it, implying the process takes 200-300M years.

• Succeeding once every 1B years implies some ungodly huge number of shots on goal - considering the cyanobacteria ones, which are post-eukaryote and post “boring billion,” think 1e⁶ interacting organisms in every cubic meter over 1e¹⁸ cubic meters of seawater over 1e⁹ years. Paulinella is super slow, and divides once every 7 days or so, so let’s go with that as a conservative lower bound. That’s 52 shots per year, over 1e⁹ years, so we’re at ^{6+18+9+1 = 34} ; 1.9e-³⁵ odds. What did I estimate the other way? 10^^-38 to stack all those chains up successfully? Lol, close enough. I guess maybe “signaling” was a thousand times easier or something (or replication was faster than 7 days, which would bump up the 35 by an OOM or two)

8.