Can a technique determine the presence of life on other planets? It seems that now – yes. The new method, developed on the basis of artificial intelligence, is able to detect minor differences in molecular structures indicating biological signals – even in samples whose age is calculated hundreds of millions of years. Moreover, the accuracy is 90%, according to a new study published in the journal Procedings of the National Academy of Sciences.
In the future, this system is planned to integrate into the sensors of space robots exploring the Moon and Mars, as well as into spacecraft that examine the worlds, such as Enshalad and Europe.
Robert Hazen, a scientist from the Carnegie Institute in Washington and co-author of the study, stated, “We proceeded from the assumption that the chemistry of life is fundamentally different from the inorganic world. If we could derive these rules, they would help us in creating models of the origin of life or in the search for weak signs of life on other planets.”
The new method is based on the fact that the chemical processes that form biomolecules are fundamentally different from abiotic molecules. Biomolecules, such as amino acids, retain information about the chemical processes that created them.
A team of scientists trained an algorithm for machine learning on 134 samples, of which 59 were biotic and 75 were abiotic. The artificial intelligence system successfully identified samples of biotic origin, such as shells, teeth, bones, and ancient remains, as well as abiotic samples.
This new method can already be used to study rocks in the Pilbara region in Western Australia, where the most ancient fossils on Earth are presumed to be located. If the presence of bacteria in these rocks is confirmed, it will indicate that life on Earth could have existed much earlier than previously thought.