Scientists have discovered the earliest signs of life on Earth using a novel technique that identifies chemical markers of living organisms in ancient rocks. This groundbreaking method, which utilizes machine learning, has the potential to revolutionize our understanding of life's origins and its search beyond our planet. The research, published in the Proceedings of the National Academy of Sciences, reveals evidence of microbial life dating back 3.3 billion years, a time when Earth was just a quarter of its current age. Furthermore, it uncovers molecular traces of oxygen-producing photosynthesis in rocks approximately 2.5 billion years old, indicating that this process occurred much earlier than previously thought. The scientists developed a technique that can distinguish between organic molecules with a biological origin and those without, achieving over 90% accuracy. This approach involves collecting and concentrating carbon-rich molecules, analyzing them to identify thousands of tiny molecular fragments, and then employing machine learning to detect subtle patterns that differentiate once-living molecules from non-living ones. This method has significantly expanded the age range at which we can identify signs of life, from 1.6 billion to 3.3 billion years. It can also distinguish between different types of life, such as photosynthetic organisms. The research has implications for the search for life on other planets, with NASA rovers exploring Mars and potential missions to Saturn's moons Enceladus and Titan, and Jupiter's moon Europa, to sample their organic-rich environments.
