Do machines have the ability to detect life on distant planets? They kind of already are, after all.
Spacecraft equipped with sensors can find compounds that could be signs of extraterrestrial life. However, organic compounds that offer hints of fascinating biological processes are known to deteriorate with time, making it challenging for present equipment to detect their presence.
However, a recently created technology based on artificial intelligence (AI) can now detect minute variations in molecular patterns that signify biological signals, even in samples that are hundreds of millions of years old. And even better, new data shows that the process provides answers with a 90% accuracy rate.
This AI system may eventually be incorporated into more intelligent sensors aboard robotic space explorers, such as rovers and landers on the moon and Mars, as well as in spacecraft circling planets that may one day support life, such as Enceladus and Europa.
According to Robert Hazen, a researcher at the Carnegie Institution for Science in Washington, D.C., and co-author of the new study, they started with the idea that the chemistry of life differs fundamentally from that of the inanimate world and that there are ‘chemical rules of life’ that influence the diversity and distribution of biomolecules. If those laws could be determined, we could use them to direct our research into simulating the origins of life or searching for minute indications of life on distant planets.
The new approach is based on the idea that biomolecules (such amino acids) retain knowledge about the chemical processes that created them, which makes them fundamentally different from abiotic molecules in terms of how they are formed and operate. The latest study suggests that this is also likely to be true for extraterrestrial life.
On any world, life may create and use a small number of substances in higher concentrations to carry out daily activities. This would set them apart from abiotic systems, and AI can detect and measure these changes, according to the researchers’ statement.
First, the scientists used 134 samples to train the machine learning system, of which 59 were biotic and 75 were abiotic. The data was then randomly divided into a training set and a test set in order to validate the method. The AI technique effectively detected biotic samples from ancient life preserved in certain fossilized fragments made of materials like coal, oil, and amber, as well as from living objects like shells, teeth, bones, rice, and human hair.
According to the latest study, the programme also recognized abiotic materials such as laboratory-produced substances like amino acids and carbon-rich meteorites.
The 3.5 billion-year-old rocks in Western Australia’s Pilbara area, believed to contain the world’s oldest fossils, can already be studied using the new AI technique. When they were first discovered in 1993, these rocks were believed to be the fossilized remains of microbes related to cyanobacteria, the first living things on Earth to create oxygen.
If proven, the bacteria’s existence so early in Earth’s history would indicate that the planet was hospitable to robust life considerably earlier than previously assumed. The data could alternatively be the result of purely geological processes unrelated to ancient life, as study has repeatedly noted. As a result, those conclusions have remained disputed. Maybe the solution is in AI.
On Monday, September 25, a report in the journal Proceedings of the National Academy of Sciences described this research.
