[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-350027-105":3,"detail-sidebar-cat-0-en-105":80,"doc-detail-350027-en":130},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":73,"head_meta":75,"extra_data":77,"updated_unix":79},105,"en","spatialfusion-a-lightweight-multimodal-foundation-model-for-pathway-informed-spatial-niche-mapping","SpatialFusion - A lightweight multimodal foundation model for pathway-informed spatial niche mapping","","Foundation models enable knowledge transfer across data modalities and tasks, yet spatial biology foundation models remain early, often encoding single-cell representations without fully integrating transcriptomic and morphological information to define functional niches. SpatialFusion is a lightweight multimodal foundation model that identifies coherent microenvironments using distinct pathway activation patterns rather than spatial proximity alone. It unifies histopathology, gene expression, and inferred pathway activity, competes with specialist methods, resolves fine-grained niches, and reveals pre-malignant and stage-predictive malignant environments across cohorts.",{"@graph":14,"@context":72},[15,34,55],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/document/","Document",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/document/research-report/","Research & Report",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/spatialfusion-a-lightweight-multimodal-foundation-model-for-pathway-informed-spatial-niche-mapping/350027/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/spatialfusion-a-lightweight-multimodal-foundation-model-for-pathway-informed-spatial-niche-mapping/350027.png","ImageObject",300,407,{"name":42,"@type":43},"\tJames","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-09-23","2026-09-22",true,{"@type":52,"interactionType":53,"userInteractionCount":22},"InteractionCounter",{"@type":54},"ViewAction",{"@type":56,"mainEntity":57},"FAQPage",[58,64,68],{"name":59,"@type":60,"acceptedAnswer":61},"What does SpatialFusion aim to improve in spatial biology modeling?","Question",{"text":62,"@type":63},"It integrates transcriptomic and morphological information instead of relying mainly on single-cell representations in spatial context. SpatialFusion focuses on defining functional niches using pathway activation patterns.","Answer",{"name":65,"@type":60,"acceptedAnswer":66},"How does SpatialFusion define spatial niches?",{"text":67,"@type":63},"SpatialFusion identifies biologically coherent microenvironments based on distinct pathway activation patterns rather than spatial proximity alone. It integrates paired histopathology, gene expression, and inferred pathway activity into one representation.",{"name":69,"@type":60,"acceptedAnswer":70},"What biological findings were reported from applying the model?",{"text":71,"@type":63},"Applied to two Visium HD cohorts, it uncovered a pre-malignant niche in morphologically normal mucosa adjacent to colorectal tumors. It also revealed distinct malignant microenvironments in non-small cell lung cancer that were predictive of tumor stage.","https://schema.org",{"og:url":32,"og:type":74,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":76,"canonical":32},"index,follow",{"doc_id":78,"site_id":7},350027,1790146280,{"code":4,"msg":81,"data":82},"success",[83,87,91,95,100,105,110,114,119,122,126],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":84,"show_sort_weight":85,"slug":86},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":88,"show_sort_weight":89,"slug":90},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":92,"show_sort_weight":93,"slug":94},"Exam",70,"exam",{"id":96,"doc_module":4,"doc_module_name":25,"category_name":97,"show_sort_weight":98,"slug":99},5,"Comic",60,"comic",{"id":101,"doc_module":4,"doc_module_name":25,"category_name":102,"show_sort_weight":103,"slug":104},6,"Technology",50,"technology",{"id":106,"doc_module":4,"doc_module_name":25,"category_name":107,"show_sort_weight":108,"slug":109},7,"Healthcare",40,"healthcare",{"id":111,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":112,"slug":113},8,30,"research-report",{"id":115,"doc_module":4,"doc_module_name":25,"category_name":116,"show_sort_weight":117,"slug":118},9,"Religion & Spirituality",20,"religion-spirituality",{"id":117,"doc_module":4,"doc_module_name":25,"category_name":120,"show_sort_weight":117,"slug":121},"World Cup","world-cup",{"id":123,"doc_module":4,"doc_module_name":25,"category_name":124,"show_sort_weight":123,"slug":125},10,"Lifestyle","lifestyle",{"id":127,"doc_module":4,"doc_module_name":25,"category_name":128,"show_sort_weight":96,"slug":129},19,"General","general",{"code":4,"msg":81,"data":131},{"doc_id":78,"user_id":132,"nickname":42,"user_avatar":133,"doc_module":4,"category_id":111,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":22,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":139,"language":140,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":12,"update_tm":144,"read_time":145},2336474466412,"https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d","bioRxiv preprint doi: [https://doi.org/10.64898/2026.03.16.712056](https://doi.org/10.64898/2026.03.16.712056); this version posted March 18, 2026. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made  \navailable under aCC-BY-NC-ND 4.0 International license.  \nSpatialFusion: A lightweight multimodal foundation model for pathway-informed spatial niche mapping  \nJosephine Yates1,2 , Mitra Shavakhi2 , Toni K. Choueiri2 , Eliezer M. Van Allen2,3,4,5*, Caroline Uhler1,6*  \n*: these authors contributed equally  \nAffiliations:  \n1: Eric and Wendy Schmidt Center at the Broad Institute of MIT and Harvard, Cambridge, USA, 02142  \n2: Department of Medical Oncology, Dana Farber Cancer Institute, Boston, MA and Harvard Medical School  \n3: Cancer Program, Broad Institute of MIT and Harvard, Cambridge, MA, USA  \n4: Division of Medical Sciences, Harvard University, Boston, MA, USA  \n5: Parker Institute for Cancer Immunotherapy, Dana-Farber Cancer Institute, Boston, MA, USA  \n6: Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, MA, USA  \nAbstract  \nFoundation models enable knowledge transfer across data modalities and tasks, yet foundation models for spatial biology remain in their early stages, largely centered on encoding single-cell representations in spatial context without fully integrating transcriptomic and morphological information to delineate functional niches. Here we introduce SpatialFusion, a lightweight multimodal foundation model that identifies biologically coherent microenvironments defined by distinct pathway activation patterns rather than spatial proximity alone. SpatialFusion integrates paired histopathology, gene expression, and inferred pathway activity into a unified representation. Compared with two specialist niche-detection methods and four spatial foundation models, SpatialFusion performs competitively and consistently resolves fine-grained spatial niches with unique pathway-level signatures. Applying the model to two Visium HD cohorts uncovered a pre-malignant niche in morphologically normal mucosa adjacent to colorectal tumors and revealed distinct malignant microenvironments in non-small cell lung cancer that were predictive of tumor stage. Overall, SpatialFusion offers a versatile framework for multimodal spatial analysis, enabling the discovery of new morpho-molecular niches with significant biological and clinical relevance.  \nbioRxiv preprint doi: [https://doi.org/10.64898/2026.03.16.712056](https://doi.org/10.64898/2026.03.16.712056); this version posted March 18, 2026. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made  \navailable under aCC-BY-NC-ND 4.0 International license.  \nIntroduction  \nFoundation models have recently transformed computational biology in domains like protein structure prediction and single-cell biology by enabling the transfer of knowledge across diverse data modalities and tasks. Trained on large and heterogeneous datasets, these deep learning architectures capture high-dimensional biological structure, often exhibiting emergent capabilities such as batch correction, without explicit supervision. In single-cell genomics, models such as scGPT 1 and Geneformer2 have demonstrated powerful generalization across cell types and tissues, while in digital pathology, foundation models including UNI3 and H-optimus ([https://huggingface.co/bioptimus/H-optimus-1](https://huggingface.co/bioptimus/H-optimus-1)) have shown similar versatility in the analysis of hematoxylin and eosin (H&E)–stained images.  \nThe advent of spatially resolved transcriptomic technologies has introduced a new dimension to molecular profiling by coupling gene expression measurements with the spatial a","cbCaihYG8Q6nB6tk","https://ap.wps.com/l/cbCaihYG8Q6nB6tk","pdf",16637700,46,"English","# Abstract\n# Introduction","[{\"question\":\"What does SpatialFusion aim to improve in spatial biology modeling?\",\"answer\":\"It integrates transcriptomic and morphological information instead of relying mainly on single-cell representations in spatial context. SpatialFusion focuses on defining functional niches using pathway activation patterns.\"},{\"question\":\"How does SpatialFusion define spatial niches?\",\"answer\":\"SpatialFusion identifies biologically coherent microenvironments based on distinct pathway activation patterns rather than spatial proximity alone. It integrates paired histopathology, gene expression, and inferred pathway activity into one representation.\"},{\"question\":\"What biological findings were reported from applying the model?\",\"answer\":\"Applied to two Visium HD cohorts, it uncovered a pre-malignant niche in morphologically normal mucosa adjacent to colorectal tumors. It also revealed distinct malignant microenvironments in non-small cell lung cancer that were predictive of tumor stage.\"}]","SpatialFusion - A lightweight multimodal foundation model for pathway-informed spatial niche mapping | PDF",1790086819,116]