AI can help find life on Mars with 90% accuracy, researchers say

Mars rovers could soon be fitted with sensors to help them look for signs of life after scientists successfully used artificial intelligence to check for signs of past or present life.

Published today in Proceedings of the National Academy of SciencesA new paper describes attempts to use a new machine learning technique to solve problems by using algorithms to reveal the biological or non-biological origin of a sample with 90% accuracy.

This routine analytical method has the potential to revolutionize the search for extraterrestrial life and deepen our understanding of both the origin and chemistry of Earth’s earliest life, said Dr. Robert Hazen of the Carnegie Institution for Science, who co-led the study. This opens up the possibility of using smart sensors on robotic spacecraft, landers and rovers to look for signs of life before the samples return to Earth.

It could even soon be used on the surface of Mars as part of instruments on NASA’s rovers. The Perseverance rover is now sampling in situ. We may already have the data to determine whether Mars has molecules from an organic Martian biosphere, Hazen said, though he added that work is needed to adapt the method to align with NASA protocols.

Impact on life pursuits

The researchers showed that artificial intelligence can distinguish between samples of biotic nature (plants, animals and bacteria) and abiotic origin (water, soil and atmosphere).

This new analytical method can detect subtle differences in the molecular patterns of samples by identifying the component parts of the sample and determining their molecular weights. The AI ​​was trained on datasets of 134 known abiotic or biotic samples.

This means scientists could soon be looking at samples from Mars and ancient Earth to determine if they were once alive, but it also shows that biochemistry differs from abiotic organic chemistry at a deep level.

According to the researchers, it was 90% accurate in identifying samples that came from:

• Living things (modern shells, teeth, bones, insects, leaves, rice, hairs and cells preserved in fine-grained rock).
• Remains of ancient life altered by geological processing (coal, oil, amber and carbon-rich fossils).
• Samples of abiotic origin (eg pure laboratory chemicals such as amino acids).

These results mean that we may be able to find a form of life on another planet, in another biosphere, even if it is very different from the life we ​​know on Earth, Hazen said. And if we find signs of life elsewhere, we can tell if life on Earth and other planets has a common or different origin.

A gift for astrobiologists

It was hoped that a new AI-driven technique could soon shed new light on the origins of 3.5-billion-year-old black sediments found in Western Australia, which may or may not contain Earth’s oldest fossil microbes. It can also be useful in biology, paleontology and archaeology.

There’s still much to learn, but one day a next-generation version of their system could fly to Mars, assessing the possibility of life on the red planet, while its Earth-bound sisters shed light on ancient life on our planet, said Andrew H. Knoll, Fisher Professor of Natural History and researcher at the Earth and Emeritus Professor of Planetary Sciences, Department of Organismal and Evolutionary Biology, Harvard University. He called it a gift to astrobiologists.

Wishing you clear skies and big eyes.