PNNL researchers Daniel Deng and Yehia Ibrahim join the 2025 class of National Academy of Inventors Fellows.

In a recent study, PNNL researchers found that noncrystalline mineral surfaces can bind microbial residues to build soil organic matter.

PNNL researchers are opening up a new front in the fight to stop viral pathogens, protecting people against multiple viruses.

An award-winning computing method called poly-streaming accelerates the analysis of massive datasets while using limited memory.

This achievement earned DIMPLES a Best Paper award at the ACM International Conference on Supercomputing 2025.

As Laboratory Director Steve Ashby pens his last monthly column, he highlights PNNL’s accomplishments with pride and gratitude.

Three research teams led by PNNL organizers achieve journal publications following the Mathematics Research Communities program.

PNNL showcased its expertise in AI, machine learning, and related domains at the International Conference on Machine Learning.

Dušan Veličković, a PNNL mass spectrometry imaging scientist received a $2.1 million grant to develop techniques to understand how changes in carbohydrate structure affect human health.

Expanding the nation’s largest wind power plant time series database; helping grid planners navigate a resilient energy future with better data.

Two new publications provide emergency response agencies with critical insights into commercially available unmanned ground vehicles used for hazardous materials response.

Jonathan Barr, senior systems engineer at PNNL, was recently invited to co-present on a panel at the Texas Department of Emergency Management Annual Conference.

PNNL has created and licensed a system that sniffs out trace levels of fentanyl, other drugs and explosives through the air.

The Wildfire Mitigation Plan Database was built to support electric utilities, state governments, policymakers, and regulators in understanding and improving wildfire risk and resilience strategies.

PNNL researchers have published their paper, “Introducing Molecular Hypernetworks for Discovery in Multidimensional Metabolomics Data,” in the Journal of Proteome Research.

Researchers can now quantify uncertainty in trained AI models aimed at solving difficult chemistry, biology and materials science problems.

Researchers have created an algorithm that slashes the calculations necessary to prepare and customize data for quantum computing.

Over the next four years, PNNL and University of Arizona will develop open-source computational tools to better identify and characterize the viruses associated with the human microbiome.

Scientists are exploring how web search engine technology might also keep the lights on, the water running and the trains moving.

Researchers developed a robust, cost-effective, and easy-to-use cap-based technique for spatial proteome mapping, addressing the lack of accessible proteomics technologies for studying tissue heterogeneity and microenvironments.