Nature Computational Science
A multidisciplinary journal that focuses on the development and use of computational techniques and mathematical models, as well as their application to address complex problems across a range of scientific disciplines.
With the rapid generation of spatial transcriptomics (ST) data, integrative analysis of multiple ST datasets can provide more comprehensive characterizations of spatial tissue structures. We develop a powerful artificial intelligence method for integrating multiple spatial transcriptomics datasets.
The strength information of atmospheric turbulence is hidden in its infrared imaging effects. We have developed a deep learning method to both extract the 2D atmospheric turbulence strength fields and clear and stable images from turbulence-distorted infrared images.
Broad materials screening for defect properties requires fast methods that circumvent the need for first-principles supercell calculations of a vast number of possible defect configurations. We construct and train a defect graph neural network to screen oxides for clean energy applications.
Best of both worlds: Large, cheap data sets enhance machine learning of high-value properties of ordered and disordered materials