Spatial transcriptomics(ST)has become a key technology for interrogating gene expression within spatial context,providing spatially resolved insights into tissue architecture and microenvironmental organization.Rapid ...Spatial transcriptomics(ST)has become a key technology for interrogating gene expression within spatial context,providing spatially resolved insights into tissue architecture and microenvironmental organization.Rapid advances in experimental platforms and analytical methods,however,have resulted in challenges for technology selection,methodological comparison,and data interpretation.In this review,we present a systematic summary of 594 ST analysis tools spanning 77 ST technologies(as of September 2025).We outline the complete analytical workflow and discuss major analytical tasks,including data preprocessing,denoising and imputation,spatial pattern and domain identification,cellular composition,trajectory analysis,cell-cell communication,and spatial multi-omics integration.For each task,we summarize representative methodological principles and emphasize platform-dependent considerations arising from differences in spatial resolution and detection efficiency.We further highlight how analytical applications of ST data have enabled biomedical discoveries by revealing spatial heterogeneity,tissue organization,and context-dependent cellular interactions.Furthermore,we develop SpatialToolDB(https://www.spatialtooldb.yelab.site/),a systematically curated,categorized,and continuously updated platform that integrates the ST technologies,analytical methods,and related databases covered in this review,facilitating informed tool selection and method comparison.We also discuss development trends and future directions of spatial-omics technologies and analytical tools,including advances in spatial technologies,AI-driven computation,benchmarking and standardization,and improved experimental validation for mechanistic and predictive spatial biology.Together,this review and SpatialToolDB provide a data-driven foundation for selecting ST platforms and analytical strategies tailored to diverse biological and translational research applications.展开更多
Micro-/nano-motors(MNMs)or swimmers are minuscule machines that can convert various forms of energy,such as chemical,electrical,or magnetic energy,into motion.These devices have attracted significant attention owing t...Micro-/nano-motors(MNMs)or swimmers are minuscule machines that can convert various forms of energy,such as chemical,electrical,or magnetic energy,into motion.These devices have attracted significant attention owing to their potential application in a wide range of fields such as drug delivery,sensing,and microfabrication.However,owing to their diverse shapes,sizes,and structural/chemical compositions,the development of MNMs faces several challenges,such as understanding their structure-function relationships,which is crucial for achieving precise control over their motion within complex environments.In recent years,machine learning techniques have shown promise in addressing these challenges and improving the performance of MNMs.Machine learning techniques can analyze large amounts of data,learn from patterns,and make predictions,thereby enabling MNMs to navigate complex environments,avoid obstacles,and perform tasks with higher efficiency and reliability.This review introduces the current state-of-the-art machine learning techniques in MNM research,with a particular focus on employing machine learning to understand and manipulate the navigation and locomotion of MNMs.Finally,we discuss the challenges and opportunities in this field and suggest future research directions.展开更多
基金supported by the National Key Research and Development Program(2024YFC3407700)the National Natural Science Foundation of China(82073145 and 3247040729)+2 种基金Shanghai Jiao Tong University 2030 Initiative(WH510363003/018)Fundamental and Interdisciplinary Disciplines Breakthrough Plan of the Ministry of Education of China(JYB2025XDXM604 and JYB2025XDXM611)the Research Funds of Centre for Leading Medicine and Advanced Technologies of IHM(2023IHM01032).
文摘Spatial transcriptomics(ST)has become a key technology for interrogating gene expression within spatial context,providing spatially resolved insights into tissue architecture and microenvironmental organization.Rapid advances in experimental platforms and analytical methods,however,have resulted in challenges for technology selection,methodological comparison,and data interpretation.In this review,we present a systematic summary of 594 ST analysis tools spanning 77 ST technologies(as of September 2025).We outline the complete analytical workflow and discuss major analytical tasks,including data preprocessing,denoising and imputation,spatial pattern and domain identification,cellular composition,trajectory analysis,cell-cell communication,and spatial multi-omics integration.For each task,we summarize representative methodological principles and emphasize platform-dependent considerations arising from differences in spatial resolution and detection efficiency.We further highlight how analytical applications of ST data have enabled biomedical discoveries by revealing spatial heterogeneity,tissue organization,and context-dependent cellular interactions.Furthermore,we develop SpatialToolDB(https://www.spatialtooldb.yelab.site/),a systematically curated,categorized,and continuously updated platform that integrates the ST technologies,analytical methods,and related databases covered in this review,facilitating informed tool selection and method comparison.We also discuss development trends and future directions of spatial-omics technologies and analytical tools,including advances in spatial technologies,AI-driven computation,benchmarking and standardization,and improved experimental validation for mechanistic and predictive spatial biology.Together,this review and SpatialToolDB provide a data-driven foundation for selecting ST platforms and analytical strategies tailored to diverse biological and translational research applications.
基金supported by the Australian Research Council(DP210100422 and FT220100479)National Breast Cancer Foundation,Australia(IIRS-22–104)Scientia Program at UNSW,Sydney。
文摘Micro-/nano-motors(MNMs)or swimmers are minuscule machines that can convert various forms of energy,such as chemical,electrical,or magnetic energy,into motion.These devices have attracted significant attention owing to their potential application in a wide range of fields such as drug delivery,sensing,and microfabrication.However,owing to their diverse shapes,sizes,and structural/chemical compositions,the development of MNMs faces several challenges,such as understanding their structure-function relationships,which is crucial for achieving precise control over their motion within complex environments.In recent years,machine learning techniques have shown promise in addressing these challenges and improving the performance of MNMs.Machine learning techniques can analyze large amounts of data,learn from patterns,and make predictions,thereby enabling MNMs to navigate complex environments,avoid obstacles,and perform tasks with higher efficiency and reliability.This review introduces the current state-of-the-art machine learning techniques in MNM research,with a particular focus on employing machine learning to understand and manipulate the navigation and locomotion of MNMs.Finally,we discuss the challenges and opportunities in this field and suggest future research directions.
