Where on Earth did life originate, and where else could it occur? A comprehensive answer is most likely a long way off. But it might depend on how many suitable sites for abiogenesis there are on different worlds.
We only have one data point for life: dear old Earth. Examining abiogenesis, the natural process where life originates from non-living matter, can’t be done by observing other places where it occurred. Instead, scientists use models to dig into the big question.
Manasvi Lingam is an astrobiologist at the Florida Institute of Technology. In new research, Lingam and his co-researchers examine the probability of life originating in different sites on Earth. The research is titled “A Bayesian Analysis of the Probability of the Origin of Life Per Site Conducive to Abiogenesis.” It’s published in the journal Astrobiology, and the other authors are Ruth Nichols and Amedeo Balbi.
“We can’t peer back in time. Sometimes you can arrive at answers just through very clever use of limited data… but there is a part that you’ll never know.”
Manasvi Lingam, Astrobiologist, Florida Institute of Technology
A Bayesian Analysis uses existing knowledge—in this case, the appearance of life on Earth—to estimate how probable it is that the same thing will occur elsewhere. Disregarding panspermia, we know that life originated on Earth at least once. Scientists can use it to try to determine how probable it is that life arose elsewhere.
There are many roadblocks on our path to understanding the spontaneous appearance of life. “One of the foremost among these current limitations is our lack of conclusive knowledge regarding the minimal set of conditions necessary for engendering abiogenesis, as well as the absence of definitive data pinpointing the likely location(s) where this process took place,” the authors write.
But the fact that it did arise on Earth, at least once but possibly in multiple locations, is an information-rich fact. But the information doesn’t announce its presence. Scientists have to tease it out. “Nevertheless, the occurrence of abiogenesis on Earth still holds significant informative value,” the authors explain.
In new research, Lingam and his co-researchers developed a model based on urable sites. Urable sites are those that are viable places where life could start. The results were surprising and counter-intuitive.
Urable sites are environments where we think life can arise. They include hydrothermal vents, impact sites, lakes and ponds, and natural atomic reactors like the one that existed in Gabon two billion years ago.
In this work, the researchers compiled a list of urable sites, and each type has a corresponding level of conduciveness for life to get going. They shaped their models according to two questions: on how many sites could life have originated on Earth, and what is the probability of life emerging for each one.
It’s critical to understand that this work can’t tell us how and where life originated. Instead, the goal was to understand how to interpret the models’ results.
In their simulations, the researchers considered three different scenarios, each with a different number of urable sites. One had only 10 urable sites, one had 1016 urable sites, and one had 1031 urable sites. They also worked with optimistic, pessimistic, and uninformative scenarios. The optimistic had a higher probability of life appearing per urable sites, the pessimistic had a lower probability, and uninformative means the results were just that.
The researchers anticipated that a larger number of urable sites would mean a higher probability of life emerging. But to their surprise, the opposite was true. More sites meant a lower probability of life emerging, and fewer sites meant a higher probability.
“That’s the two situations that are here. One where there are lots of sites, but there’s very low probability [of life] per site. And the second where there are very few sites, but there’s a very high probability per site,” Lingam said in a press release.
“Normally ‘the more, the better’ is the attitude for many things in life,” Lingam says. “But more is not always better. If it’s fewer, but it’s the right kind of fewer, then that can actually be better.”
This means that in their model, where Earth had the fewest urable sites, the probability of life emerging on any single site is higher. When there are plentiful sites, the probability of life emerging on any one of them is lower.
Though counterintuitive, Lingam says these results are valuable. There’s no consensus on what urable site life arose on, so different researchers can use them in their experiments to understand their own preferred environments in experiments. “Then they can do laboratory experiments, try to get a feel for how many trials might be needed to actually move to something like life,” Lingam says.
Even with all we don’t know about the origin of life, and even though these models can’t tell us how life arose, Lingam’s work can still help other researchers make progress.
“We can’t peer back in time,” Lingam says. “Sometimes you can arrive at answers just through very clever use of limited data… but there is a part that you’ll never know.”
The phylogenetic evidence has been clear for some years, the predictions are that biology had split from geology in deep ocean hydrothermal vents 4.3 – 4.2 billion years ago. [For the latest passed test, see “The nature of the last universal common ancestor and its impact on the early Earth system”, Moody et al, Nature ecology and evolution, 2024.] Based on such data it is fairly clear that more sites means lowered time to the split, and indeed the short time to life on a geologically active early Earth may attest to that.
Without having access to the paper it is hard to tell what an uninformed model explores, but I assume it mostly explores itself. A different model may then come to different conclusions, perhaps more akin to what we already know.
I’d agree (with you Torbjorn). Another article, Moddy is co-author on, which looks at ATP Synthase gives a similarly early origin of life. Again, without seeing Lingam’s paper I’m not sure of the model. However, I’d disagree if it’s saying that given one confirmed origin of life, then more sites means 1/number of sites, therefore lower probability per site. The issue with that is that you are assuming that only the currently observed life is the only life that ever emerged. That’s a false argument, given the propensity for competition and extinction – particularly when the early Earth was not a pleasant environment by current standards.
Moody’s work, which implies LUCA lived in a community, would also suggest that there was rather a lot of life going on, most of which is no longer evident.