Machine learning framework can search for signs of extraterrestrial life

A machine learning framework can distinguish molecules made by biological processes from those formed by non-biological processes and could be used to analyze samples returned from current and future planetary missions. The results are published in the journal Nexus PNAS.

José C. Aponte, Amirali Aghazadeh and colleagues analyzed eight carbonaceous meteorites and ten terrestrial geological samples using two-dimensional gas chromatography coupled with high-resolution time-of-flight mass spectrometry.

Using these data, the authors developed LifeTracer, a computational framework that processes mass spectrometry data and applies machine learning to identify patterns distinguishing abiotic from biotic origins. A logistic regression model trained on compound-level features achieved over 87% accuracy in classifying samples as meteoritic or terrestrial.

The analysis identified 9,475 peaks in meteorite samples and 9,070 in terrestrial samples, with statistically significant differences between the two sample types in terms of molecular weight distributions and retention times, which describe the time it takes for the compound to move through the two columns of the chromatograph. Organic compounds present in meteorite samples showed significantly lower retention times, consistent with higher volatility in abiotically formed materials.

The framework identified polycyclic aromatic hydrocarbons and alkyl variants as key predictive features, with naphthalene emerging as the most predictive compound for abiotic samples. According to the authors, the approach enables scalable and unbiased biosignature detection and could be a powerful tool for interpreting the complex organic mixtures that will be returned by current and future planetary sample return missions.