Machine learning Archives

An integrated approach to date bacterial evolution and reconstruct the history of oxygen adaptation. We inferred a bacterial timetree by integrating genomic, fossil, and geochemical data and linking oxygen tolerance and aerobic metabolism to the GOE. Colors denote anaerobic (blue) and aerobic (red) states, whereas shades of purple show the fraction of aerobic lineages within extant bacterial phyla. Mitochondria and plastids were included to leverage the more extensive eukaryotic fossils. Land plants and animals are indicated for temporal comparison. -- Science

AI - Data - Apps - Cybernetics, Biophysics, Biosignatures & Paleobiology, Fossils & Paleontology, Genomics, Proteomics, Bioinformatics, Microbiology & Virology, Origin & Evolution of Life, Press Release

Machine Learning Helps Construct An Evolutionary Timeline Of Bacteria

by Keith CowingApril 4, 2025April 4, 2025

University of Queensland scientists have helped to construct a detailed timeline for bacterial evolution, suggesting some bacteria used oxygen long before evolving the ability to produce it through photosynthesis.

AI - Data - Apps - Cybernetics, Exploration Gear & Tech, Icy Worlds, Imaging & Spectroscopy, Missions & Hardware, NASA Astrobiology Program, Press Release, Reports, Roadmaps, & Plans, Robotics/Rovers/UAVs, ROVs/Submersibles/Diving, Sample Return, Tricorders & Sensors

Science Autonomy Using Machine Learning For Astrobiology – A White Paper For 2025 NASA DARES

by Keith CowingApril 3, 2025May 25, 2025

In recent decades, artificial intelligence (AI) including machine learning (ML) have become vital for space missions enabling rapid data processing, advanced pattern recognition, and enhanced insight extraction.

(a) Visualization of statistical effects of the highly predictive biosignature features using a Regression-based Association-Interaction Network (RAIN) with node numbers in order of NPDR-LURF importance. RAIN encodes feature main effects and statistical interactions in a network. Nodes 2, 3, 4, and 5 have the largest main effects. For nodes 2 and 4, the main effect increases the probability of an abiotic class prediction. The TS feature diff2_acff1 is also a hub in the interaction network, participating in three large interactions with nodes 2, 3, and 5 that influence the probability for both biotic and abiotic predictions. (b) The Random Forest classifier trained on NPDR-LURF features for biosignatures yields a training accuracy of 87.9%, trained on a sample size m = 140. Dark gray and green diagonals represent the number of correct predictions: 81/89 abiotic samples are correctly classified, and 42/51 biotic samples are correctly classified. The precision for biotic samples in the training data is 0.824 and the recall is 0.84. (c) The same classifier yields a high test accuracy of 88.2%: 20/22 abiotic samples are correctly classified, and 10/12 biotic samples are correctly classified. The biotic precision for the test data is 0.833 and the recall is also 0.833. -- Earth and Space Science

Biochemistry & Organic Chemistry, Biogeochemical Cycles & Geobiology, Biosignatures & Paleobiology, Habitable Zones, Icy Worlds, Status Report, Water/Hycean Worlds & Oceanography

Interpretable Machine Learning Biosignature Detection From Ocean Worlds Analogue CO2 Isotopologue Data

by Keith CowingMarch 25, 2025March 25, 2025

Future missions to icy ocean worlds (OW) such as Europa and Enceladus will evaluate the habitability and potential for biosignatures on these worlds.

The computational expense of solving non-equilibrium chemistry equations in astrophysical simulations poses a significant challenge, particularly in high-resolution, large-scale cosmological models. -- astro-ph.IM)

AI - Data - Apps - Cybernetics, Biochemistry & Organic Chemistry, Biophysics, Biosignatures & Paleobiology, Genomics, Proteomics, Bioinformatics, Origin & Evolution of Life, Status Report

Bridging Machine Learning And Cosmological Simulations: Using Neural Operators To Emulate Chemical Evolution

by Keith CowingMarch 17, 2025March 17, 2025

In this work, we explore the potential of machine learning, specifically Neural Operators, to emulate the Grackle chemistry solver, which is widely used in cosmological hydrodynamical simulations.

The [Na/Fe] vs. [Fe/H] abundance plot (Ensemble 4) showing the stars that are most likely to host a planet (green), the known planet hosts (blue), and the stars less likely to host a planet (orange). Histograms are also included that show the [Na/Fe] relative distribution to the right of the scatterplot and [Fe/H] relative distribution at the top. We note the clear sub-solar Na abundances for each experiment’s known and predicted planet hosts (see Section 5). - Small planets -- astro-ph.EP

AI - Data - Apps - Cybernetics, Astronomy & Telescopes, Exoplanets, -moons, -comets, Habitable Zones, Imaging & Spectroscopy, Status Report, Stellar Cartography

Utilizing Machine Learning to Predict Host Stars and the Key Elemental Abundances of Small Planets

by Keith CowingMarch 5, 2025

Stars and their associated planets originate from the same cloud of gas and dust, making a star’s elemental composition a valuable indicator for indirectly studying planetary compositions.

Ratio of the mass-loss rates predicted for the test planets by interpol2021 (panels a and c) and interpol2024 (panels b and d) to the true values given by the hydrodynamic model, against Λ and Λ × K, respectively. The points are colour-coded according to the Jtrans value. In panels (c) and (d), the red points have Jbord = 1, while the red points marked by a yellow plus are those with Jtrans = 2. For reference, the two horizontal black lines in each panel lie at 0.5 and 2. -- astro-ph.EP

AI - Data - Apps - Cybernetics, Atmospheres, Climate, Weather, Exoplanets, -moons, -comets, Status Report

Grid-based Exoplanet Atmospheric Mass Loss Predictions Through Neural Network

by Keith CowingFebruary 12, 2025February 12, 2025

The fast and accurate estimation of planetary mass-loss rates is critical for planet population and evolution modelling.

Library cleaning and accuracy determination workflow. In this study, (A) two GCMS spectral libraries (NIST17 and MoNA) were filtered to include only experimental spectra from monosubstituted α-amino acids. A third library (IOCB) was composed of experimental GCMS spectra from seventeen amino acids, also not found in the NIST dataset. (B) The accuracy of predicted GCMS spectra is measured through three metrics: root mean square error, spectral contrast angle, and weighted cosine similarity. -- chemrxiv.org

Astrochemistry, Biochemistry & Organic Chemistry, Biophysics, Biosignatures & Paleobiology, Genomics, Proteomics, Bioinformatics, Imaging & Spectroscopy, Nanotechnology & SynBio, Status Report

Mapping the Edges of Mass Spectral Prediction: Evaluation of Machine Learning EIMS Prediction for Xeno Amino Acids

by Keith CowingJanuary 15, 2025January 15, 2025

Mass spectrometry is one of the most effective analytical methods for unknown compound identification. By comparing observed m/z spectra with a database of experimentally determined spectra, this process identifies compound(s) […]

A Thunderstorm on an arid Earth-like exoplanet. -- Grok via Astrobiology.com

AI - Data - Apps - Cybernetics, Atmospheres, Climate, Weather, Biogeochemical Cycles & Geobiology, Exoplanets, -moons, -comets, Habitable Zones, Icy Worlds, Press Release, Terraforming & Geoengineering, Water/Hycean Worlds & Oceanography

Planetary Weather Modeling Advance: GenCast Predicts Earth’s Weather With State-of-the-art Accuracy

by Keith CowingDecember 6, 2024December 6, 2024

Editor’s note: We should all stop every once in a while and remember that Earth – our home world – is a planet too. We have the benefit of living […]

Illustration of the main modalities included in the Multimodal Universe, along with typical associated machine learning tasks. In addition, the Multimodal Universe also includes a small amount of hyperspectral images and tabular data, not shown here. -- astro-ph.IM

AI - Data - Apps - Cybernetics, Apps - Games, Astronomy & Telescopes, Status Report

The Multimodal Universe: Enabling Large-Scale Machine Learning with 100TB of Astronomical Scientific Data

by Keith CowingDecember 6, 2024December 6, 2024

We present the MULTIMODAL UNIVERSE, a large-scale multimodal dataset of scientific astronomical data, compiled specifically to facilitate machine learning research.

Schematic comparison of DNFs and CNFs. DNFs (trained with NPE) were first applied to exoplanetary atmospheric retrieval by Vasist et al. (2023), while CNFs (trained with FMPE) and the idea of noise level-conditioning are discussed in this work. -- astro-ph.IM

AI - Data - Apps - Cybernetics, Astronomy & Telescopes, Atmospheres, Climate, Weather, Exoplanets, -moons, -comets, Status Report

Flow Matching for Atmospheric Retrieval of Exoplanets: Where Reliability meets Adaptive Noise Levels

by Keith CowingOctober 30, 2024October 30, 2024

Inferring atmospheric properties of exoplanets from observed spectra is key to understanding their formation, evolution, and habitability.

Machine learning

Machine Learning Helps Construct An Evolutionary Timeline Of Bacteria

Science Autonomy Using Machine Learning For Astrobiology – A White Paper For 2025 NASA DARES

Interpretable Machine Learning Biosignature Detection From Ocean Worlds Analogue CO2 Isotopologue Data

Bridging Machine Learning And Cosmological Simulations: Using Neural Operators To Emulate Chemical Evolution

Utilizing Machine Learning to Predict Host Stars and the Key Elemental Abundances of Small Planets

Grid-based Exoplanet Atmospheric Mass Loss Predictions Through Neural Network

Mapping the Edges of Mass Spectral Prediction: Evaluation of Machine Learning EIMS Prediction for Xeno Amino Acids

Planetary Weather Modeling Advance: GenCast Predicts Earth’s Weather With State-of-the-art Accuracy

The Multimodal Universe: Enabling Large-Scale Machine Learning with 100TB of Astronomical Scientific Data

Flow Matching for Atmospheric Retrieval of Exoplanets: Where Reliability meets Adaptive Noise Levels

The Role Of Minerals In Prebiotic Evolution: Implications For Identifying Worlds With The Potential For Life’s Origin

Gen-EBM: An Energy Balance Model For Terrestrial Planets. Climate Multistability From Snowball To Post-runaway Greenhouse Climates

Astrometric Planet Search Around Southern Ultracool Dwarfs: V. Discovery Of The Exoplanet Candidate PALTA-1 b

Hubble Space Telescope Observations of Active Asteroid (101955) Bennu

SOPHIE And TESS Characterization Of Fourteen TESS Objects Of Interest: Nine Planetary Systems And Five Low-mass Stellar Companions