Spatial Arrangement Information Research Articles

VASE: A High-Entropy Alloy Short-Range Order Structural Descriptor for Machine Learning.

The short-range order (SRO) structure in high-entropy alloys (HEAs) is closely associated with many properties, which can be studied through density functional theory (DFT) calculations. Atomic-scale modeling and calculations require substantial computational resources, and machine learning can provide rapid estimations of DFT results. To describe SRO information in HEAs, a new descriptor based on Voronoi Analysis and Shannon Entropy (VASE) is proposed. Based on Voronoi analysis, the Shannon entropy is introduced to directly characterize atomic spatial arrangement information except for composition and atomic interactions, which is necessary for describing the disorder atomic occupancy in HEAs. The new descriptor is used for predicting the formation energy of FeCoNiAlTiCu system based on machine learning model, which is more accurate than other descriptors (Coulomb matrices, partial radial distribution functions, and Voronoi analysis). Moreover, the model trained based on VASE descriptors exhibits the best predictive performance for unrelaxed structures (24.06 meV/atom). The introduction of Shannon entropy provides an effective representation of atomic arrangement information in HEAs, which is a powerful tool for investigating the SRO phenomena.

Importance of Terrain and Climate for Predicting Soil Organic Carbon Is Highly Variable across Local to Continental Scales.

Soil organic carbon (SOC) plays a vital role in global carbon cycling and sequestration, underpinning the need for a comprehensive understanding of its distribution and controls. This study explores the importance of various covariates on SOC spatial distribution at both local (up to 1.25 km) and continental (USA) scales using a deep learning approach. Our findings highlight the significant role of terrain attributes in predicting SOC concentration distribution with terrain, contributing approximately one-third of the overall prediction at the local scale. At the continental scale, climate is only 1.2 times more important than terrain in predicting SOC distribution, whereas at the local scale, the structural pattern of terrain is 14 and 2 times more important than climate and vegetation, respectively. We underscore that terrain attributes, while being integral to the SOC distribution at all scales, are stronger predictors at the local scale with explicit spatial arrangement information. While this observational study does not assess causal mechanisms, our analysis nonetheless presents a nuanced perspective about SOC spatial distribution, which suggests disparate predictors of SOC at local and continental scales. The insights gained from this study have implications for improved SOC mapping, decision support tools, and land management strategies, aiding in the development of effective carbon sequestration initiatives and enhancing climate mitigation efforts.

Computational methods for alignment and integration of spatially resolved transcriptomics data

Most of the complex biological regulatory activities occur in three dimensions (3D). To better analyze biological processes, it is essential not only to decipher the molecular information of numerous cells but also to understand how their spatial contexts influence their behavior. With the development of spatially resolved transcriptomics (SRT) technologies, SRT datasets are being generated to simultaneously characterize gene expression and spatial arrangement information within tissues, organs or organisms. To fully leverage spatial information, the focus extends beyond individual two-dimensional (2D) slices. Two tasks known as slices alignment and data integration have been introduced to establish correlations between multiple slices, enhancing the effectiveness of downstream tasks. Currently, numerous related methods have been developed. In this review, we first elucidate the details and principles behind several representative methods. Then we report the testing results of these methods on various SRT datasets, and assess their performance in representative downstream tasks. Insights into the strengths and weaknesses of each method and the reasons behind their performance are discussed. Finally, we provide an outlook on future developments. The codes and details of experiments are now publicly available at https://github.com/YangLabHKUST/SRT_alignment_and_integration.

Multidimensional immune profiling in Gastric Cancer Multiplex Immunohistochemistry Atlas from Peking University Cancer Hospital project informs PD-1/PD-L1 blockade efficacy

BackgroundImmunotherapy has resulted in impressive objective response rates and durable tumour remission, but only in a subset of gastric cancer (GC) patients. The PD-L1 combined positive score is the most widely used tissue-based biomarker for anti-PD-1/PD-L1 therapy; however, this unidimensional method has limitations. Next-generation exploration of tissue-based biomarkers for GC requires characterisation of various cellular markers and key immunoregulatory molecule expression in situ. Thus, a complete, stepwise solution covering the entire process from staining samples to cross-site utilisation of pathomics data is urgently needed. MethodsWith the advanced multispectral imaging analysis method, web-based data repository, and interactive sharing technology, we conducted a project entitled Gastric Cancer Multiplex Immunohistochemistry Atlas from Peking University Cancer Hospital (GMAP). We propose a standard pipeline covering sample collection, staining, scanning multispectral images, constructing a spectral library, identifying and phenotyping cells, positioning each element, and quantitatively extracting immune features. We designed an open-access relational database to explore tissue-based biomarkers to determine PD-1/PD-L1 blockade efficacy. ResultsThe GMAP project detected the functional status and spatial location of more than 50 million cells using 15 markers in 80 GC patients, based on which billions of cell pairs were recognised, highlighting the rich spatial arrangement information and the fine tumour microenvironment structure. We generated a tumour-immune atlas using the count and spatial features of 65 immune cell types. We eventually selected the indicators and built a comprehensive risk-scoring system. Patients with higher risk score showed superior immunotherapy-related progression-free survival (irPFS) (hazard ratio [HR]: 3.19; P < 0.001; median irPFS: 4.87 versus 19.87months, respectively) and immunotherapy-related overall survival (HR: 3.10; P = 0.001; median irPFS: 10.03 versus 24.87months, respectively) compared with lower risk patients, demonstrating their potential for guiding anti-PD-1/PD-L1-based immunotherapy. Importantly, an easy-to-use and versatile web server was built to promote tissue-based biomarker exploration in GC. ConclusionThe GMAP project highlighted the clinical value of tissue-based immune features as biomarkers for immunotherapeutic decision-making. We present a well-designed, detailed workflow for the orderly generation and use of a high-quality, spatially resolved pathological database.

A Bayesian characterization of urban land use configurations from VHR remote sensing images

The composition and arrangement of spatial entities, i.e., land cover objects, play a key role in distinguishing land use types from very high resolution (VHR) remote sensing images, in particular in urban environments. This paper presents a new method to characterize the spatial arrangement for urban land use extraction using VHR images. We derive an adjacency unit matrix to represent the spatial arrangement of land cover objects obtained from a VHR image, and use a graph convolutional network to quantify the spatial arrangement by extracting hidden features from adjacency unit matrices. The distribution of the spatial arrangement variables, i.e., hidden features, and the spatial composition variables, i.e., widely used land use indicators, are then estimated. We use a Bayesian method to integrate the variables of spatial arrangement and composition for urban land use extraction. Experiments were conducted using three VHR images acquired in two urban areas: a Pleiades image in Wuhan in 2013, a Superview image in Wuhan in 2019, and a GeoEye image in Oklahoma City in 2012. Our results show that the proposed method provides an effective means to characterize the spatial arrangement of land cover objects, and produces urban land use extractions with overall accuracies (i.e., 86% and 93%) higher than existing methods (i.e., 83% and 88%) that use spatial arrangement information based on building types on the Pleiades and GeoEye datasets. Moreover, it is unnecessary to further categorize the dominant land cover type into finer types for the characterization of spatial arrangement. We conclude that the proposed method has a high potential for the characterization of urban structure using different VHR images, and for the extraction of urban land use in different urban areas.

Extracting spatial arrangement of structural textures using projection information

A new algorithm is proposed which extracts the spatial arrangement information of texture elements in structural textures. In the proposed algorithm, using projection information in several directions obtained from the texture image, two directions which determine the texture structure and parallelogram grid which isolates texture elements are extracted accurately. The isolated texture elements are analyzed and used to synthesize a texture image. Computer simulation shows that the proposed method can extract the correct spatial structure of the structural texture even when the texture image is highly corrupted by additive noise. Furthermore, the selected spatial arrangement information is shown to be a very effective feature in texture classification.

Extraction of periodicity vectors from structural textures using projection information

In this paper we propose a new algorithm which extracts spatial arrangement information of texture elements in structural textures. In the proposed algorithm, by the use of projection information in several directions obtained from the texture image, two directions which determine the texture structure are selected. The selected two directions represent correctly the structure of regularly arranged textures and simplify the calculations required to extract an arrangement rule.