{"database":"biostudies-literature","file_versions":[],"scores":null,"additional":{"omics_type":["Unknown"],"submitter":["Wu E"],"funding":["NCI NIH HHS"],"pubmed_abstract":["Hematoxylin and eosin (H&E) is a common and inexpensive histopathology assay. Though widely used and information-rich, it cannot directly inform about specific molecular markers, which require additional experiments to assess. To address this gap, we present <i>ROSIE,</i> a deep-learning framework that computationally imputes the expression and localization of dozens of proteins from H&E images. Our model is trained on a dataset of over 1000 paired and aligned H&E and multiplex immunofluorescence (mIF) samples from 20 tissues and disease conditions, spanning over 16 million cells. Validation of our <i>in silico mIF</i> staining method on held-out H&E samples demonstrates that the predicted biomarkers are effective in identifying cell phenotypes, particularly distinguishing lymphocytes such"],"journal":["bioRxiv : the preprint server for biology"],"pagination":["2024.11.10.622859"],"full_dataset_link":["https://www.ebi.ac.uk/biostudies/studies/S-EPMC11601356"],"repository":["biostudies-literature"],"pubmed_title":["&lt;i&gt;ROSIE&lt;/i&gt;: AI generation of multiplex immunofluorescence staining from histopathology images."],"pmcid":["PMC11601356"],"funding_grant_id":["P20 CA252733","P50 CA285275","R01 CA280639"],"pubmed_authors":["Bieniosek M","Zou J","Wu E","Charville GW","Li L","Wu Z","Giba H","Schurch C","Huyghe JR","Mayer AT","Peters U","Li CI","Makky A","Behera V","Raman A","Thakkar N","Trevino AE"],"additional_accession":[]},"is_claimable":false,"name":"&lt;i&gt;ROSIE&lt;/i&gt;: AI generation of multiplex immunofluorescence staining from histopathology images.","description":"Hematoxylin and eosin (H&E) is a common and inexpensive histopathology assay. Though widely used and information-rich, it cannot directly inform about specific molecular markers, which require additional experiments to assess. To address this gap, we present <i>ROSIE,</i> a deep-learning framework that computationally imputes the expression and localization of dozens of proteins from H&E images. Our model is trained on a dataset of over 1000 paired and aligned H&E and multiplex immunofluorescence (mIF) samples from 20 tissues and disease conditions, spanning over 16 million cells. Validation of our <i>in silico mIF</i> staining method on held-out H&E samples demonstrates that the predicted biomarkers are effective in identifying cell phenotypes, particularly distinguishing lymphocytes such","dates":{"release":"2024-01-01T00:00:00Z","publication":"2024 Nov","modification":"2026-04-07T20:19:10.377Z","creation":"2025-04-03T23:57:13.862Z"},"accession":"S-EPMC11601356","cross_references":{"pubmed":["39605711"],"doi":["10.1101/2024.11.10.622859"]}}