Google Research

Applied science

Applying powerful AI to solve grand challenges in natural science for the benefit of the world.

aerial view of the earth

Applying powerful AI to solve grand challenges in natural science for the benefit of the world.

About the team

Computer science and natural science are complementary: breakthroughs in one drive remarkable advances in the other. Google’s Applied Science organization aims to cross-fertilize these two fields across a wide range of scientific disciplines.

Within Applied Science, the Climate & Energy team focuses on technology to mitigate climate change, while the Science AI team leverages AI to accelerate progress in natural sciences. By tackling grand challenges from reconstructing the brain to inventing zero-carbon energy sources, Applied Science advances both computer science and scientific research simultaneously, including using AI to automate code generation for computational experiments.

Team focus summaries

Climate and energy

The Climate & Energy team deploys AI and Google's extensive computing power to combat climate change. We aim to provide the scientific insights and technical tools necessary for global climate mitigation and adaptation. We also investigate both technological and nature-based methods for atmospheric CO2 removal, including sequestration in the oceans.

Primary research goals include:

  • Accelerating clean energy innovations, such as nuclear fusion.
  • Developing strategies to mitigate various warming agents, including aviation-related contrails.
  • Employing AI alongside remote sensing to monitor and manage climate-related features like wildfires, methane leaks, and water resources.

Models of the world

Google researchers and their collaborators use computational power to drive breakthroughs in climate science, environmental modeling and biodiversity mapping. Key initiatives include:

  • NeuralGCM: Combining fundamental physics laws with an ML model trained on historical weather and satellite data to create a hybrid atmospheric model. This open-source model, NeuralGCM, is fast, efficient and can reproduce real-world weather events with high accuracy.
  • Targeting climate forecasting: The development of smaller-scale, specialized models for particular weather and climate prediction needs.
  • Global biodiversity mapping: Using LLMs and geospatial models to map the habitats of hundreds of thousands of species to facilitate local, national, and international conservation efforts.

Multimodal biology and genomics

For more than a decade, the genomics group at Google has leveraged AI to enable more accurate analyses of genetic data. As genetic sequencing technology has evolved, so have our analysis tools. Our suite of open-source AI models such as DeepVariant, DeepConsensus, DeepPolisher and, most recently, DeepSomatic, support genetic analysis in research, medicine and conservation.

Today we also have single-cell sequencing to detect when genes are active, as well as data from medical imaging, fitness trackers, patient history and more. The multimodal biology group at Google seeks ways to combine these various data sources to unlock medical insights.

Neuroscience and connectomics

The human brain contains billions of neurons and fully mapping it remains out of reach. However, Google Research is precisely mapping the connections between every cell in simpler brains to someday understand and even treat mental illnesses. In collaboration with external research partners, we develop AI-based methods to efficiently reconstruct neurons from 2D images and create tools, such as Neuroglancer, to view and analyze that data.

We are working with our research partners to create increasingly ambitious brain maps for model organisms such as the fruit fly, zebrafish, the mouse and even sections of the human brain. Research groups are now using those maps to predict the neural activity and behavior of model organisms.

Next-generation scientific tools

We develop foundational tools for scientific discovery that leverage AI-enabled coding and LLMs to democratize access to advanced analysis and automate complex research tasks.

Featured publications

ZAPBench: a benchmark for whole-brain activity prediction in zebrafish
Jan-Matthis Lückmann
Alex Immer
Alex Bo-Yuan Chen
Peter Li
Mariela Petkova
Nirmala Iyer
Luuk Hesselink
Aparna Dev
Gudrun Ihrke
Woohyun Park
Alyson Petruncio
Aubrey Weigel
Wyatt Korff
Florian Engert
Jeff W. Lichtman
Misha Ahrens
Michał Januszewski
Viren Jain
International Conference on Learning Representations (ICLR) (2025)
Preview abstract Data-driven benchmarks have led to significant progress in key scientific modeling domains including weather and structural biology. Here, we present the Zebrafish Activity Prediction Benchmark (ZAPBench), which quantitatively measures progress on the problem of predicting cellular-resolution neural activity throughout an entire vertebrate brain. The benchmark is based on a novel dataset containing 4d light-sheet microscopy recordings of more than 70,000 neurons in a larval zebrafish brain, along with motion stabilized and voxel-level cell segmentations of these data that facilitate development of a variety of forecasting methods. Initial results from a selection of time series and volumetric video modeling approaches achieve better performance than naive baseline methods, but also show room for further improvement. The specific brain used in the activity recording is also undergoing synaptic-level anatomical mapping, which will enable future integration of detailed structural information into ZAP forecasting methods. View details
Light-microscopy-based dense connectomic reconstruction of mammalian brain tissue
Mojtaba R. Tavakoli
Julia Lyudchik
Michał Januszewski
Vitali Vistunou
Nathalie Agudelo Duenas
Jakob Vorlaufer
Christoph Sommer
Caroline Kreuzinger
Barbara de Souza Oliveira
Alban Cenameri
Gaia Novarino
Viren Jain
Johann Danzl
Nature (2025)
Preview abstract The information-processing capability of the brain’s cellular network depends on the physical wiring pattern between neurons and their molecular and functional characteristics. Charting neurons and resolving the individual synaptic connections requires volumetric imaging at nanoscale resolution and comprehensive cellular contrast. Light microscopy is uniquely positioned to visualize specific molecules but dense, synapse-level circuit reconstruction by light microscopy has been out of reach due to limitations in resolution, contrast, and volumetric imaging capability. Here we developed light-microscopy based connectomics (LICONN). We integrated hydrogel embedding and expansion with comprehensive deep-learning based segmentation and analysis of connectivity, thus directly incorporating molecular information in synapse-level brain tissue reconstructions. LICONN will allow synapse-level brain tissue phenotyping in biological experiments in a readily adoptable manner. View details
Unsupervised representation learning on high-dimensional clinical data improves genomic discovery and prediction
Taedong Yun
Justin Cosentino
Babak Behsaz
Zachary Ryan Mccaw
Davin Hill
Robert Luben
Dongbing Lai
John Bates
Howard Yang
Tae-Hwi Schwantes-An
Yuchen Zhou
Anthony Khawaja
Andrew Carroll
Brian Hobbs
Michael Cho
Cory Y. McLean
Farhad Hormozdiari
Nature Genetics (2024)
Preview abstract Although high-dimensional clinical data (HDCD) are increasingly available in biobank-scale datasets, their use for genetic discovery remains challenging. Here we introduce an unsupervised deep learning model, Representation Learning for Genetic Discovery on Low-Dimensional Embeddings (REGLE), for discovering associations between genetic variants and HDCD. REGLE leverages variational autoencoders to compute nonlinear disentangled embeddings of HDCD, which become the inputs to genome-wide association studies (GWAS). REGLE can uncover features not captured by existing expert-defined features and enables the creation of accurate disease-specific polygenic risk scores (PRSs) in datasets with very few labeled data. We apply REGLE to perform GWAS on respiratory and circulatory HDCD—spirograms measuring lung function and photoplethysmograms measuring blood volume changes. REGLE replicates known loci while identifying others not previously detected. REGLE are predictive of overall survival, and PRSs constructed from REGLE loci improve disease prediction across multiple biobanks. Overall, REGLE contain clinically relevant information beyond that captured by existing expert-defined features, leading to improved genetic discovery and disease prediction. View details
A scalable system to measure contrail formation on a per-flight basis
Scott Geraedts
Erica Brand
Sebastian Eastham
Carl Elkin
Thomas Dean
Zebediah Engberg
Ulrike Hager
Ian Langmore
Kevin McCloskey
Joe Ng
John Platt
Dinesh Sanekommu
Tharun Sankar
Aaron Sarna
Marc Shapiro
Nita Goyal
Environmental Research Communications (2024)
Preview abstract In this work we describe a scalable, automated system to determine from satellite data whether a given flight has made a persistent contrail. The system works by comparing flight segments to contrails detected by a computer vision algorithm running on images from the GOES-16 Advanced Baseline Imager. We develop a `flight matching' algorithm and use it to label each flight segment as a `match' or `non-match'. We perform this analysis on 1.6 million flight segments and compare these labels to existing contrail prediction methods based on weather forecast data. The result is an analysis of which flights make persistent contrails several orders of magnitude larger than any previous work. We find that current contrail prediction models fail to correctly predict whether we will match a contrail in many cases. View details
CURIE: Evaluating LLMs on multitask long context scientific understanding and reasoning
Michael Brenner
Matthew Abraham
Haining Pan
Peter Norgaard
Zahra Shamsi
Muqthar Mohammad
Chenfei Jiang
Ruth Alcantara
Gowoon Cheon
Sameera Ponda
Xuejian Ma
Michael Statt
Jackson Cui
Dan Morris
Martyna Plomecka
Nayantara Mudur
Eun-Ah Kim
Paul Raccuglia
Victor V. Albert
Lizzie Dorfman
Brian Rohr
Shutong Li
Maria Tikhanovskaya
Viren Jain
Drew Purves
Elise Kleeman
Yasaman Bahri
Philippe Faist
Subhashini Venugopalan
Ean Phing VanLee
International Conference on Learning Representations (ICLR) (2025)
Preview abstract The core of the scientific problem-solving process involves synthesizing information while applying expert knowledge. Large Language Models (LLMs) have the potential to accelerate this process due to their extensive knowledge across a variety of domains. Recent advancements have also made it possible for LLMs to handle very long "in-context" content. However, existing evaluations of long-context LLMs have focused on assessing their ability to summarize or retrieve information within the given context, primarily in generalist tasks that do not require deep scientific expertise. To facilitate analogous assessments of domain-specific tasks, we introduce the scientific long-Context Understanding and Reasoning Inference Evaluations (CURIE) benchmark. This benchmark provides a set of 8 challenging tasks, derived from around 250 scientific research papers, requiring domain expertise, comprehension of long in-context information, and multi-step reasoning that tests the ability of LLMs to assist scientists in realistic workflows. Tasks in CURIE have been collected from experts in six disciplines - materials science, theoretical condensed matter physics, quantum computing, geospatial analysis, biodiversity, and protein sequencing - covering both experimental and theoretical workflows in science. We evaluate a range of closed and open LLMs on these tasks. Additionally, we propose strategies for task decomposition, which allow for a more nuanced evaluation of the models and facilitate staged multi-step assessments. We hope that insights gained from CURIE can guide the future development of LLMs. View details
Benchmarking and improving algorithms for attributing satellite-observed contrails to flights
Aaron Sarna
Vincent Rudolf Meijer
Rémi Chevallier
Allie Duncan
Kyle McConnaughay
Scott Geraedts
Kevin McCloskey
Atmospheric Measurement Techniques, 18 (2025), pp. 3495-3532
Preview abstract Condensation trail (contrail) cirrus clouds cause a substantial fraction of aviation's climate impact. One proposed method for the mitigation of this impact involves modifying flight paths to avoid particular regions of the atmosphere that are conducive to the formation of persistent contrails, which can transform into contrail cirrus. Determining the success of such avoidance maneuvers can be achieved by ascertaining which flight formed each nearby contrail observed in satellite imagery. The same process can be used to assess the skill of contrail forecast models. The problem of contrail-to-flight attribution is complicated by several factors, such as the time required for a contrail to become visible in satellite imagery, high air traffic densities, and errors in wind data. Recent work has introduced automated algorithms for solving the attribution problem, but it lacks an evaluation against ground-truth data. In this work, we present a method for producing synthetic contrail detections with predetermined contrail-to-flight attributions that can be used to evaluate – or “benchmark” – and improve such attribution algorithms. The resulting performance metrics can be employed to understand the implications of using these observational data in downstream tasks, such as forecast model evaluation and the analysis of contrail avoidance trials, although the metrics do not directly quantify real-world performance. We also introduce a novel, highly scalable contrail-to-flight attribution algorithm that leverages the characteristic compounding of error induced by simulating contrail advection using numerical weather models. The benchmark shows an improvement of approximately 25 % in precision versus previous contrail-to-flight attribution algorithms, without compromising recall. View details
Neural general circulation models for weather and climate
Dmitrii Kochkov
Janni Yuval
Ian Langmore
Peter Norgaard
Jamie Smith
Griffin Mooers
Milan Kloewer
James Lottes
Stephan Rasp
Peter Dueben
Samuel Hatfield
Peter Battaglia
Alvaro Sanchez
Matthew Willson
Michael Brenner
Stephan Hoyer
Nature, 632 (2024), pp. 1060-1066
Preview abstract General circulation models (GCMs) are the foundation of weather and climate prediction. GCMs are physics-based simulators that combine a numerical solver for large-scale dynamics with tuned representations for small-scale processes such as cloud formation. Recently, machine-learning models trained on reanalysis data have achieved comparable or better skill than GCMs for deterministic weather forecasting. However, these models have not demonstrated improved ensemble forecasts, or shown sufficient stability for long-term weather and climate simulations. Here we present a GCM that combines a differentiable solver for atmospheric dynamics with machine-learning components and show that it can generate forecasts of deterministic weather, ensemble weather and climate on par with the best machine-learning and physics-based methods. NeuralGCM is competitive with machine-learning models for one- to ten-day forecasts, and with the European Centre for Medium-Range Weather Forecasts ensemble prediction for one- to fifteen-day forecasts. With prescribed sea surface temperature, NeuralGCM can accurately track climate metrics for multiple decades, and climate forecasts with 140-kilometre resolution show emergent phenomena such as realistic frequency and trajectories of tropical cyclones. For both weather and climate, our approach offers orders of magnitude computational savings over conventional GCMs, although our model does not extrapolate to substantially different future climates. Our results show that end-to-end deep learning is compatible with tasks performed by conventional GCMs and can enhance the large-scale physical simulations that are essential for understanding and predicting the Earth system. View details
A petavoxel fragment of human cerebral cortex reconstructed at nanoscale resolution
Alex Shapson-Coe
Michal Januszewski
Daniel R. Berger
Art Pope
Yuelong Wu
Tim Blakely
Richard L. Schalek
Peter Hawley Li
Shuohong Wang
Jeremy Maitin-Shepard
Neha Karlupia
Sven Dorkenwald
Evelina Sjostedt
Laramie J Leavitt
Dongil Lee
Luke Bailey
Angerica Fitzmaurice
Rohin Kar
Benjamin Field
Hank Wu
Julian Wagner-Carena
David Aley
Joanna Lau
Zudi Lin
Donglai Wei
Hanspeter Pfister
Adi Peleg
Viren Jain
Jeff W. Lichtman
Science (2024)
Preview abstract To fully understand how the human brain works, knowledge of its structure at high resolution is needed. Presented here is a computationally intensive reconstruction of the ultrastructure of a cubic millimeter of human temporal cortex that was surgically removed to gain access to an underlying epileptic focus. It contains about 57,000 cells, about 230 millimeters of blood vessels, and about 150 million synapses and comprises 1.4 petabytes. Our analysis showed that glia outnumber neurons 2:1, oligodendrocytes were the most common cell, deep layer excitatory neurons could be classified on the basis of dendritic orientation, and among thousands of weak connections to each neuron, there exist rare powerful axonal inputs of up to 50 synapses. Further studies using this resource may bring valuable insights into the mysteries of the human brain. View details
CO2 capture by pumping surface acidity to the deep ocean
Christopher H Van Arsdale
John Platt
Mike Tyka
Energy and Environmental Science (EES) (2022)
Preview abstract The majority of IPCC scenarios call for active CO2 removal (CDR) to remain below 2ºC of warming. On geological timescales, ocean uptake regulates atmospheric CO2 concentration, with two homeostats driving sequestration: dissolution of deep ocean calcite deposits and terrestrial weathering of silicate rocks, acting on 1ka to 100ka timescales. Many current ocean-based CDR proposals effectively act to accelerate the latter. Here we present a method which relies purely on the redistribution and dilution of acidity from a thin layer of the surface ocean to a thicker layer of deep ocean, with the aim of accelerating the former carbonate homeostasis. This downward transport could be seen analogous to the action of the natural biological carbon pump. The method offers advantages over other ocean CDR methods and direct air capture approaches (DAC): the conveyance of mass is minimized (acidity is pumped in situ to depth), and expensive mining, grinding and distribution of alkaline material is eliminated. No dilute substance needs to be concentrated, avoiding the Sherwood’s Rule costs typically encountered in DAC. Finally, no terrestrial material is added to the ocean, avoiding significant alteration of seawater ion concentrations and issues with heavy metal toxicity encountered in mineral-based alkalinity schemes. The artificial transport of acidity accelerates the natural deep ocean invasion and subsequent compensation by calcium carbonate. It is estimated that the total compensation capacity of the ocean is on the order of 1500GtC. We show through simulation that pumping of ocean acidity could remove up to 150GtC from the atmosphere by 2100 without excessive increase of local ocean pH. For an acidity release below 2000m, the relaxation half time of CO2 return to the atmosphere was found to be ~2500 years (~1000yr without accounting for carbonate dissolution), with ~85% retained for at least 300 years. The uptake efficiency and residence time were found to vary with the location of acidity pumping, and optimal areas were calculated. Requiring only local resources (ocean water and energy), this method could be uniquely suited to utilize otherwise-stranded open ocean energy sources at scale. We examine technological pathways that could be used to implement it and present a brief techno-economic estimate of 130-250$/tCO2 at current prices and as low as 86$/tCO2 under modest learning-curve assumptions. View details
Next Day Wildfire Spread: A Machine Learning Dataset to Predict Wildfire Spreading From Remote-Sensing Data
Fantine Huot
Lily Hu
Nita Goyal
Tharun Pratap Sankar
Matthias Ihme
Yi-fan Chen
IEEE Transactions on Geoscience and Remote Sensing, 60 (2022), pp. 1-13
Preview abstract Predicting wildfire spread is critical for land management and disaster preparedness. To this end, we present “Next Day Wildfire Spread,” a curated, large-scale, multivariate dataset of historical wildfires aggregating nearly a decade of remote-sensing data across the United States. In contrast to existing fire datasets based on Earth observation satellites, our dataset combines 2-D fire data with multiple explanatory variables (e.g., topography, vegetation, weather, drought index, and population density) aligned over 2-D regions, providing a feature-rich dataset for machine learning. To demonstrate the usefulness of this dataset, we implement a neural network that takes advantage of the spatial information of these data to predict wildfire spread. We compare the performance of the neural network with other machine learning models: logistic regression and random forest. This dataset can be used as a benchmark for developing wildfire propagation models based on remote-sensing data for a lead time of one day. View details
View more publications

Highlighted work

Some of our locations

Some of our people

Follow us

×