They have developed a simple and highly reliable test that can identify life on other planets, even if they haven’t been around for long.
This can be read in the magazine Proceedings of the National Academy of Sciences. We can safely call it a hack. “Our method has the potential to revolutionize the search for extraterrestrial life and advance our understanding of the origins and chemistry of the first life forms on Earth,” said researcher Robert Hazen.
complicated
To be sure, the work done by Hazen and his colleagues is not an unnecessary luxury; The search for life on other planets has proven to be very complex. The matter may be more complex than scientists could have imagined in the middle of the last century. At that time, researchers discovered that – under the right conditions – mixing simple chemicals could easily create more complex molecules that form the basis of life (think amino acids, for example). He gently hinted that such molecules could also easily see daylight beyond Earth. And the truth; In the years since, many of the essential components of life – such as nucleotides, a component of DNA – have already been found in space. But that was not evidence of the existence of extraterrestrial life. Because it remained unclear whether these molecules had a biological origin – that is, they were the remains of life – or were created by abiotic processes. While that is not clear, we do not know whether we have discovered extraterrestrial life by detecting such molecules.
Amnesty International
In the end, it all boils down to one simple question, Hazen says. Is there a fundamental difference between the chemical processes that underlie life and the chemical processes that occur in a world without life? “We found out it was.” The researchers came to this conclusion after collecting 134 different carbon-rich samples. This included samples of living cells, but also samples of chemicals manufactured in a laboratory, fossil fuels produced by geological processes, and even a piece of a carbon-rich meteorite. These samples were then heated in an oxygen-free environment, causing them to disintegrate. The components that make up it can then be identified. This molecular analysis provided a wealth of information that was then used to train the artificial intelligence system. By showing the system the molecular characteristics associated with biological and abiotic samples, the researchers hope the system will detect subtle but compelling differences between both types of samples. Because once it becomes clear how different they are from each other, the system can also tell them apart. Thus, for example, to determine whether organic molecules collected on Mars are the result of life or abiotic processes.
90 percent
The results are hopeful. Because after training, the AI system was shown to be able to predict with more than 90 percent accuracy whether a sample is biological or non-vital. “In our analysis we do not aim to identify a specific molecule, but to determine the biological or non-biological origin by looking at the context in which the substance is found.”
This is important, because this approach makes it possible to look beyond the implications of life as we know it. “In this way, we should also be able to detect exotic biochemistry,” says Hazen. “This is important, because it is relatively easy to discover molecular biomarkers of terrestrial life, but we cannot assume that extraterrestrial life would also use DNA, amino acids, etc. Our method looks at patterns of molecular distributions that arise from the fact that Life simply requires “functional” molecules.
“Chemistically, the differences between biological and abiotic samples depend on things like water solubility, molecular mass, volatility, etc.,” adds researcher Jim Cleaves. Illustrate this using an example. “The cell has a membrane and an interior, also called the cytosol. The membrane is actually insoluble in water, while its contents are somewhat soluble in water (…) So if you divide such a living cell or tissue into components, “You’ll get a mixture of molecules that dissolve easily in water and molecules that are more difficult to dissolve in water.” It is clear that the distribution of these water-soluble and poorly water-soluble molecules and the frequency of their observation varies from one substance to another. “For example, things like oil and coal have often lost a lot of their water-soluble substances over their long history.” But what is striking is that the distribution of this property of the components of abiotic materials does not differ only from one abiotic material to another. “But it’s also very clearly distinct from the distribution you see with biological materials,” Cleaves emphasizes. This is one of the contexts that enables artificial intelligence to distinguish between living and non-living organisms.
Three groups
The first experiences are promising in this regard. They demonstrated that the system can differentiate between biotic or abiotic origin based on subtle differences in molecular context. Thus, the artificial intelligence system served the researchers according to their desire. In fact, she gave them a little more than they asked for, Hazen says. “What really surprised us was that… Machine learningA way to distinguish between two groups – biotic and abiotic – but the system eventually turned out to be able to distinguish between three groups: abiotic, biotic, and fossiliferous. In other words, it can distinguish between fossil specimens and more recent biological specimens: in other words, for example, between a just-picked lettuce leaf and something that died a long time ago. This is good news. Because this means that the system will soon be able to tell us, for example, whether any biotraces on Mars attest to living or extinct aliens. “In addition, based on this result, we are optimistic that we can also distinguish other properties (of biospecimens, ed.), such as whether they bear witness to photosynthetic life or the presence of eukaryotes (cells with a nucleus).” says guard.
Start fast
The work of Hazen and his colleagues is expected to give the search for life a serious boost. We don’t even have to wait long for that. For example, this method can already be used on samples collected by the Mars rover Curiosity. This rover has been roaming the surface of Mars for some time and has something of a laboratory on board where samples are analyzed molecularly. “We may already have data with which we can determine whether there are molecules on Mars that attest to the Martian biosphere,” Hazen says.
In addition, the scientists’ approach also gives us the opportunity to gain more knowledge about the origins of life on Earth in the short term. For example, this method can also be used on traces of ancient life on Earth, and is controversial. For example, you can think of 3.5 billion-year-old rocks found in Western Australia in which some scientists believe they have found the oldest fossilized microbes, while other researchers don’t want to know anything about them. With the help of artificial intelligence, it should be possible to settle this controversy and thus gain more knowledge, for example, of the moment when life appeared on Earth.
