Scientists from Lund University in Sweden have developed an innovative AI tool capable of identifying a person’s recent location based on microorganisms they have encountered during their travels. This breakthrough, published in Genome Biology and Evolution, offers a novel approach to pinpointing whether someone has visited a specific location such as a beach, train station, or park, all by analyzing the microorganisms left behind.
The researchers found that microorganisms, such as bacteria, fungi, and algae, act as microscopic “fingerprints” that can be traced back to specific geographic locations. These microbial communities, much like human populations, exhibit regional characteristics, enabling scientists to develop a tool that can identify where someone has been based on the microbiome they carry.
Unlike traditional GPS, which relies on satellite signals, this AI-driven system uses the concept of the Microbiome Geographic Population Structure (mGPS). This model links the microbial communities found in various environments to specific geographical regions, making it possible to trace a person’s movements through the microorganisms they have encountered.
To develop this cutting-edge tool, researchers trained the AI model using a massive dataset that includes microbiome samples collected from diverse environments. This data set included:
By analyzing this extensive collection of microbiome samples, the AI model learned to recognize the unique “fingerprints” of microbial communities and link them to specific geographical coordinates.
Eran Elhaik, a researcher at Lund University and co-author of the study, explained that, unlike human DNA, the human microbiome changes constantly depending on the environments a person encounters. “By tracing where your microorganisms have been recently, we can understand the spread of disease, identify potential sources of infection, and localize microbial resistance. This also opens up new opportunities in forensics,” Elhaik said.
The mGPS system has shown impressive accuracy in identifying the geographical origin of microbiome samples. According to the study, the AI tool successfully pinpointed the city of origin for 92% of the urban samples it analyzed. In a more challenging test, the system was able to distinguish between subway stations in Hong Kong that were just 564 feet apart. In New York City, it could even differentiate between a kiosk and a handrail located less than one meter from each other.
However, the AI system’s accuracy decreased in London, where it only correctly assigned half of the samples to their geographical clusters. Researchers attributed this reduced accuracy to the poorly maintained conditions of the city’s underground stations.
The development of this AI tool has enormous potential for various fields, including medicine, epidemiology, and forensic science. By analyzing microbiome data, scientists can track the spread of infectious diseases, identify environmental sources of microbial resistance, and even solve criminal cases by linking microbiome evidence to specific locations.
As the system continues to be refined and more microbiome data is collected, the AI model’s accuracy is expected to improve, opening up even more possibilities for its application in real-world scenarios. This innovation could revolutionize how we understand microbial geography and environmental interactions, offering an exciting glimpse into the future of AI-driven research.