[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-82230-en":3,"doc-seo-82230-105":29,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},82230,962075114765,"Quinn","https://ap-avatar.wpscdn.com/davatar_a8503ba1806abce46bf441b54a3ca4cd",8,"Research & Report","COAST Context Aware Differential Learning for Gene Expression Prediction in Spatial Transcriptomics","Spatial transcriptomics profiles gene expression in native tissue space but is limited by cost, resolution, and throughput, motivating prediction from widely available H&E histopathology images. COAST is a context-aware differential learning framework that captures fine-grained local patterns and slide-level structure by modulating local and global context features and aggregating target and context tokens with a Transformer encoder. Training uses a joint objective combining absolute expression regression with signed differential supervision between target and context spots.","arXiv :2607 .09 166v 1 [ cs .LG] 10 Jul 2026  \nCOAST: Context-Aware Differential Learning for Gene Expression Prediction in Spatial Transcriptomics  \nKeunho Byeon 1 , Sunhong Park 1 , Jeewoo Lim 1 , and Jin Tae Kwak 1  \nSchool of Electrical Engineering, Korea University, Seoul 02841, Republic of Korea {bkh5922,sunhongpapa,jeewoolim,[jkwak}@korea.ac.kr](jkwak}@korea.ac.kr)  \nAbstract. Spatial transcriptomics enables profiling of spatial gene expression but is limited by high cost and low throughput, motivating prediction from H&E histopathology images. Existing context-aware methods mainly supervise absolute expression, while relative expression relationships between spots are rarely used explicitly. We propose COAST, a context-aware differential learning framework for spatial gene expression prediction. COAST conditions the local and global context features with type-specific modulation and aggregates the target and context spot tokens using a Transformer encoder to capture both fine-grained local patterns and slide-level structure. It is trained with a joint objective that combines absolute expression regression with signed differential regression between the target and context spots. Experiments on multiple spatial transcriptomics datasets show consistent improvements in correlation-and distribution-based metrics, demonstrating the effectiveness of context-aware differential learning for histology-based spatial gene expression prediction.  \nKeywords: Spatial Transcriptomics · Differential Learning · ContextAware  \n1 Introduction  \nSpatial transcriptomics (ST) has emerged as a transformative technology that bridges morphological observations with molecular profiling by providing gene expression measurements within their native spatial context. Despite its great potential, current ST platforms, such as 10x Genomics Visium and Slide-seq, are constrained by high experimental costs, limited resolution, and relatively low throughput. These practical barriers severely restrict their widespread adoption in routine clinical workflows. Consequently, there is growing interest in developing and deploying computational approaches capable of inferring spatial gene expression directly from universally available and cost-effective hematoxylin and eosin (H&E) stained histopathology images. Inferring gene expression from histology is inherently challenging due to the complex and highly non-linear relationship between tissue morphology and transcriptomic states. Early approaches framed this as an independent patch-level regression problem, focusing on local  \n2 K. Byeon et al.  \nfeature representations extracted from individual image patches and regressing expression values at each spot in isolation. However, gene expression is fundamentally shaped by spatial context. Morphologically similar regions can exhibit distinct expression levels depending on their specific microenvironment, structural compartment, and global tissue composition. To address this, recent methods introduced neighbor aggregation, graph-based propagation, or attentionbased mechanisms to incorporate spatial context from adjacent regions.  \nDespite these advancements, two critical limitations persist. First, most existing models are optimized to regress the absolute expression magnitude at each spot, making predictions sensitive to slide-level variability and protocoldependent shifts. Second, although spatial context is often provided as input, models are rarely supervised to explicitly preserve relative spatial relationships. In biological tissues, spatial gradients and relative changes between regions (such as tumor-stroma boundaries) frequently encode more stable and structurally meaningful signals than absolute counts.  \nTo address these limitations, we propose COAST, a context-aware differential learning framework for spatial gene expression prediction from H&E histopathology images. COAST seamlessly integrates heterogeneous contextual information through type-specific c","cbCaikxHEVBkZEnV","https://ap.wps.com/l/cbCaikxHEVBkZEnV","pdf",868541,1,10,"English","en",105,"# Introduction\n## Problem background and motivation\n## Limitations of existing context-aware methods\n## COAST approach\n# Related Works\n## Histology-based gene expression prediction","[{\"question\":\"What problem does COAST address in spatial transcriptomics?\",\"answer\":\"COAST targets the difficulty of predicting spatial gene expression when experimental platforms are costly and limited, by inferring expression from H\\u0026E histopathology images.\"},{\"question\":\"How does COAST differ from prior context-aware methods?\",\"answer\":\"COAST jointly supervises absolute expression regression and signed differential relationships between a target spot and its contextual spots, explicitly preserving relative spatial signals.\"},{\"question\":\"What model components enable COAST to use local and global spatial context?\",\"answer\":\"COAST uses type-specific modulation of local and global context features and aggregates target/context spot tokens with a Transformer encoder to capture local patterns and slide-level structure.\"}]",1784178994,25,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":27},"coast-context-aware-differential-learning-for-gene-expression-prediction-in-spatial-transcriptomics","",{"@graph":35,"@context":85},[36,53,68],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/coast-context-aware-differential-learning-for-gene-expression-prediction-in-spatial-transcriptomics/82230/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-21","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does COAST address in spatial transcriptomics?","Question",{"text":75,"@type":76},"COAST targets the difficulty of predicting spatial gene expression when experimental platforms are costly and limited, by inferring expression from H&E histopathology images.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does COAST differ from prior context-aware methods?",{"text":80,"@type":76},"COAST jointly supervises absolute expression regression and signed differential relationships between a target spot and its contextual spots, explicitly preserving relative spatial signals.",{"name":82,"@type":73,"acceptedAnswer":83},"What model components enable COAST to use local and global spatial context?",{"text":84,"@type":76},"COAST uses type-specific modulation of local and global context features and aggregates target/context spot tokens with a Transformer encoder to capture local patterns and slide-level structure.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,134],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":45,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":45,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":45,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":45,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":45,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":21,"doc_module":4,"doc_module_name":45,"category_name":132,"show_sort_weight":21,"slug":133},"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":45,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]