{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,4]],"date-time":"2026-06-04T18:12:34Z","timestamp":1780596754015,"version":"3.54.1"},"update-to":[{"DOI":"10.1371\/journal.pcbi.1013581","type":"new_version","label":"New version","source":"publisher","updated":{"date-parts":[[2026,6,4]],"date-time":"2026-06-04T00:00:00Z","timestamp":1780531200000}}],"reference-count":37,"publisher":"Public Library of Science (PLoS)","issue":"5","license":[{"start":{"date-parts":[[2026,5,26]],"date-time":"2026-05-26T00:00:00Z","timestamp":1779753600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100025294","name":"Research Ireland","doi-asserted-by":"crossref","award":["21\/RI\/9787"],"award-info":[{"award-number":["21\/RI\/9787"]}],"id":[{"id":"10.13039\/501100025294","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/501100025294","name":"Research Ireland","doi-asserted-by":"crossref","award":["13\/RC\/2106_P2"],"award-info":[{"award-number":["13\/RC\/2106_P2"]}],"id":[{"id":"10.13039\/501100025294","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/501100001592","name":"Higher Education Authority","doi-asserted-by":"publisher","award":["AICRIstart"],"award-info":[{"award-number":["AICRIstart"]}],"id":[{"id":"10.13039\/501100001592","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100025294","name":"Research Ireland","doi-asserted-by":"crossref","award":["13\/CDA\/2183"],"award-info":[{"award-number":["13\/CDA\/2183"]}],"id":[{"id":"10.13039\/501100025294","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/501100025294","name":"Research Ireland","doi-asserted-by":"crossref","award":["20\/FFP-P\/8884"],"award-info":[{"award-number":["20\/FFP-P\/8884"]}],"id":[{"id":"10.13039\/501100025294","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/100010661","name":"Horizon 2020 Framework Programme","doi-asserted-by":"publisher","award":["754923"],"award-info":[{"award-number":["754923"]}],"id":[{"id":"10.13039\/100010661","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["www.ploscompbiol.org"],"crossmark-restriction":false},"short-container-title":["PLoS Comput Biol"],"abstract":"<jats:p>Identifying image features that associate strongly with diagnostic or prognostic classes in large-scale, multi-channel spatial imaging is challenging without pixel-level annotations. We present DREAMER-S, an attention-based multiple-instance learning (MIL) framework that, using only image- or slide-level labels, learns spatial features within 3D imaging hypercubes that are most informative for downstream classification. We demonstrate DREAMER-S on Quantum Cascade Laser infrared (QCL-IR) tissue imaging, where attention weights are rendered spatially to highlight class-relevant spectral instances without manual annotation. Because the MIL attention layer assigns interpretable importances to spatial instances, the method is broadly transferable to spatial-biology applications that require instance-level filtering to focus towards salient regions of interest in high-content datasets. We further evaluate DREAMER-S on a chemotherapy-response task in a colorectal cancer patient-derived xenograft (PDX) model. After tuning, DREAMER-S separated spectral instances from a chemo-sensitive PDX (CRC0344) and a less responsive PDX (CRC0076) with an F1 score of ~0.95. To validate explainability, we linked model saliency to cellular physiology, observing that, (i) unsupervised UMAP embeddings of high-attention spectra stratified samples by treatment (chemotherapy, apoptosis sensitizer, combination, vehicle), and (ii) selected spectral markers correlated with pro-apoptotic proteins measured independently in the same PDX system. Together, these results support a mechanistic link between spectral signals and apoptosis pathways and position DREAMER-S as an efficient, interpretable approach for analysing high-content spatial-biology imaging datasets.<\/jats:p>","DOI":"10.1371\/journal.pcbi.1013581","type":"journal-article","created":{"date-parts":[[2026,5,26]],"date-time":"2026-05-26T17:40:35Z","timestamp":1779817235000},"page":"e1013581","update-policy":"https:\/\/doi.org\/10.1371\/journal.pcbi.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["DREAMER-S: Deep leaRning-Enabled Attention-based Multiple-instance approaches with Explainable Representations for Spatial biology"],"prefix":"10.1371","volume":"22","author":[{"given":"M. Rifqi","family":"Rafsanjani","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Alison","family":"Dooney","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rahul","family":"Suresh","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Alice C.","family":"O\u2019Farrell","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Monika A.","family":"Jarzabek","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Liam","family":"Shiels","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Annette T.","family":"Byrne","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jochen H. M.","family":"Prehn","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0353-9768","authenticated-orcid":true,"given":"Aidan D.","family":"Meade","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,5,26]]},"reference":[{"issue":"1","key":"pcbi.1013581.ref001","doi-asserted-by":"crossref","first-page":"76","DOI":"10.1109\/TKDE.2009.58","article-title":"MILD: Multiple-instance learning via disambiguation","volume":"22","author":"W-J Li","year":"2010","journal-title":"IEEE Trans Knowl Data Eng"},{"key":"pcbi.1013581.ref002","doi-asserted-by":"crossref","first-page":"4277","DOI":"10.1109\/CVPR.2015.7299056","article-title":"Multiple instance learning for soft bags via top instances","author":"W Li","year":"2015","journal-title":"2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)"},{"key":"pcbi.1013581.ref003","doi-asserted-by":"crossref","first-page":"1508","DOI":"10.1109\/ICPR.2014.268","article-title":"Classification of COPD with Multiple Instance Learning","author":"V Cheplygina","year":"2014","journal-title":"2014 22nd International Conference on Pattern Recognition"},{"key":"pcbi.1013581.ref004","doi-asserted-by":"crossref","first-page":"78409","DOI":"10.1109\/ACCESS.2024.3403538","article-title":"Multiple instance learning in medical images: A systematic review","volume":"12","author":"D Barbosa","year":"2024","journal-title":"IEEE Access"},{"key":"pcbi.1013581.ref005","first-page":"48","article-title":"Multi-instance multi-label learning for whole slide breast histopathology","author":"C Mercan","year":"2016","journal-title":"Medical Imaging 2016: Digital Pathology"},{"issue":"1","key":"pcbi.1013581.ref006","doi-asserted-by":"crossref","first-page":"89","DOI":"10.1039\/C2AY25544H","article-title":"Diagnostic segregation of human brain tumours using Fourier-transform infrared and\/or Raman spectroscopy coupled with discriminant analysis","volume":"5","author":"K Gajjar","year":"2013","journal-title":"Anal Methods"},{"issue":"10","key":"pcbi.1013581.ref007","doi-asserted-by":"crossref","first-page":"2047","DOI":"10.1039\/c0an00972e","article-title":"Fourier-transform infrared spectroscopy discriminates a spectral signature of endometriosis independent of inter-individual variation","volume":"136","author":"KT Cheung","year":"2011","journal-title":"Analyst"},{"issue":"13","key":"pcbi.1013581.ref008","doi-asserted-by":"crossref","first-page":"6666","DOI":"10.1021\/ac501395g","article-title":"Raman imaging providing insights into chemical composition of lipid droplets of different size and origin: In hepatocytes and endothelium","volume":"86","author":"K Majzner","year":"2014","journal-title":"Anal Chem"},{"issue":"8","key":"pcbi.1013581.ref009","doi-asserted-by":"crossref","first-page":"1771","DOI":"10.1038\/nprot.2014.110","article-title":"Using Fourier transform IR spectroscopy to analyze biological materials","volume":"9","author":"MJ Baker","year":"2014","journal-title":"Nat Protoc"},{"issue":"10","key":"pcbi.1013581.ref010","doi-asserted-by":"crossref","first-page":"2978","DOI":"10.3390\/cancers12102978","article-title":"Implementing systems modelling and molecular imaging to predict the efficacy of BCL-2 inhibition in colorectal cancer patient-derived xenograft models","volume":"12","author":"AC O\u2019Farrell","year":"2020","journal-title":"Cancers (Basel)"},{"key":"pcbi.1013581.ref011","unstructured":"Krizhevsky A, Hinton G. Learning multiple layers of features from tiny images. 2009."},{"issue":"2","key":"pcbi.1013581.ref012","doi-asserted-by":"crossref","first-page":"025002","DOI":"10.1117\/1.JBO.27.2.025002","article-title":"Saliva-based detection of COVID-19 infection in a real-world setting using reagent-free Raman spectroscopy and machine learning","volume":"27","author":"K Ember","year":"2022","journal-title":"J Biomed Opt"},{"issue":"16","key":"pcbi.1013581.ref013","doi-asserted-by":"crossref","first-page":"3817","DOI":"10.1039\/D3AN00618B","article-title":"Weakly supervised anomaly detection coupled with Fourier transform infrared (FT-IR) spectroscopy for the identification of non-normal tissue","volume":"148","author":"D Ferguson","year":"2023","journal-title":"Analyst"},{"issue":"29","key":"pcbi.1013581.ref014","doi-asserted-by":"crossref","first-page":"10957","DOI":"10.1021\/acs.analchem.3c00979","article-title":"Weakly supervised identification and localization of drug fingerprints based on label-free hyperspectral CARS microscopy","volume":"95","author":"J Shi","year":"2023","journal-title":"Anal Chem"},{"key":"pcbi.1013581.ref015","doi-asserted-by":"crossref","first-page":"163786","DOI":"10.1016\/j.scitotenv.2023.163786","article-title":"Exploiting weak supervision to facilitate segmentation, classification, and analysis of microplastics (<100 \u03bcm) using Raman microspectroscopy images","volume":"886","author":"S Phan","year":"2023","journal-title":"Sci Total Environ"},{"key":"pcbi.1013581.ref016","doi-asserted-by":"crossref","first-page":"103","DOI":"10.1016\/j.eswa.2018.09.049","article-title":"Multiple instance learning for histopathological breast cancer image classification","volume":"117","author":"PJ Sudharshan","year":"2019","journal-title":"Expert Systems with Applications"},{"key":"pcbi.1013581.ref017","article-title":"Incorporating hierarchical information into multiple instance learning for patient phenotype prediction with single-cell RNA-sequencing data","volume":"41","author":"C Do","year":"2025","journal-title":"Bioinformatics"},{"key":"pcbi.1013581.ref018","doi-asserted-by":"crossref","unstructured":"Grouard S, Esposito C, Khoury JE, Ducret V, Thiriez C, Herpin L. Multiple instance learning with spatial transcriptomics for interpretable patient-level predictions: application in glioblastoma. 2025. 2025.10.13.682206. https:\/\/doi.org\/10.1101\/2025.10.13.682206","DOI":"10.1101\/2025.10.13.682206"},{"issue":"6","key":"pcbi.1013581.ref019","doi-asserted-by":"crossref","first-page":"508","DOI":"10.1158\/2159-8290.CD-11-0109","article-title":"A molecularly annotated platform of patient-derived xenografts (\u201cxenopatients\u201d) identifies HER2 as an effective therapeutic target in cetuximab-resistant colorectal cancer","volume":"1","author":"A Bertotti","year":"2011","journal-title":"Cancer Discov"},{"issue":"1","key":"pcbi.1013581.ref020","doi-asserted-by":"crossref","first-page":"7495","DOI":"10.1038\/s41467-024-51909-2","article-title":"XENTURION is a population-level multidimensional resource of xenografts and tumoroids from metastatic colorectal cancer patients","volume":"15","author":"SM Leto","year":"2024","journal-title":"Nat Commun"},{"key":"pcbi.1013581.ref021","unstructured":"Erb D. pybaselines: A Python library of algorithms for the baseline correction of experimental data. https:\/\/doi.org\/10.5281\/zenodo.5608581"},{"issue":"85","key":"pcbi.1013581.ref022","first-page":"2825","article-title":"Scikit-learn: Machine learning in python","volume":"12","author":"F Pedregosa","year":"2011","journal-title":"J Mach Learn Res"},{"key":"pcbi.1013581.ref023","article-title":"Pytorch: An imperative style, high-performance deep learning library","volume":"32","author":"A Paszke","year":"2019","journal-title":"Adv Neural Inf Process Syst"},{"key":"pcbi.1013581.ref024","doi-asserted-by":"crossref","first-page":"770","DOI":"10.1109\/CVPR.2016.90","article-title":"Deep Residual Learning for Image Recognition","author":"K He","year":"2016","journal-title":"2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)"},{"key":"pcbi.1013581.ref025","author":"M Ilse","year":"2018"},{"key":"pcbi.1013581.ref026","article-title":"A unified approach to interpreting model predictions","volume-title":"Advances in Neural Information Processing Systems","author":"SM Lundberg","year":"2017"},{"key":"pcbi.1013581.ref027","author":"L McInnes","year":"2020"},{"key":"pcbi.1013581.ref028","first-page":"2837","article-title":"Information theoretic measures for clusterings comparison: Variants, properties, normalization and correction for chance","volume":"11","author":"NX Vinh","year":"2010","journal-title":"J Mach Learn Res"},{"key":"pcbi.1013581.ref029","first-page":"583","article-title":"Cluster ensembles --- a knowledge reuse framework for combining multiple partitions","author":"A Strehl","year":"2003","journal-title":"J Mach Learn Res"},{"issue":"9","key":"pcbi.1013581.ref030","doi-asserted-by":"crossref","first-page":"1144","DOI":"10.1038\/s41416-018-0211-x","article-title":"Metabolite and lipoprotein responses and prediction of weight gain during breast cancer treatment","volume":"119","author":"TS Madssen","year":"2018","journal-title":"Br J Cancer"},{"issue":"1","key":"pcbi.1013581.ref031","doi-asserted-by":"crossref","first-page":"91","DOI":"10.1186\/s12944-018-0745-1","article-title":"Status of lipid and lipoprotein in female breast cancer patients at initial diagnosis and during chemotherapy","volume":"17","author":"X Li","year":"2018","journal-title":"Lipids Health Dis"},{"key":"pcbi.1013581.ref032","first-page":"47","article-title":"Misfolded proteins: from little villains to little helpers in the fight against cancer","volume":"5","author":"A Br\u00fcning","year":"2015","journal-title":"Front Oncol"},{"issue":"5","key":"pcbi.1013581.ref033","doi-asserted-by":"crossref","first-page":"556","DOI":"10.1177\/0003702818819857","article-title":"Digital staining of high-definition FTIR images using deep learning","volume":"73","author":"M Lotfollahi","year":"2019","journal-title":"Appl Spectrosc"},{"key":"pcbi.1013581.ref034","unstructured":"Sukumaran S. Protein secondary structure elucidation using FTIR spectroscopy. 2017 [cited 2025 Aug 7]. https:\/\/gammadata.se\/wp-content\/uploads\/2024\/01\/TFS-Assets_MSD_Application-Notes_AN52985-protein-secondary-structure-elucidation-using-ftir-spectroscopy.pdf"},{"issue":"5","key":"pcbi.1013581.ref035","doi-asserted-by":"crossref","first-page":"1717","DOI":"10.1007\/s00216-006-0876-5","article-title":"Growth substrate induced functional changes elucidated by FTIR and Raman spectroscopy in in-vitro cultured human keratinocytes","volume":"387","author":"AD Meade","year":"2007","journal-title":"Anal Bioanal Chem"},{"issue":"2","key":"pcbi.1013581.ref036","doi-asserted-by":"crossref","first-page":"477","DOI":"10.1016\/S0301-4622(03)00035-8","article-title":"A library of IR bands of nucleic acids in solution","volume":"104","author":"M Banyay","year":"2003","journal-title":"Biophys Chem"},{"issue":"4","key":"pcbi.1013581.ref037","doi-asserted-by":"crossref","first-page":"2743","DOI":"10.1021\/acs.analchem.5c04765","article-title":"Spatial-spectral deep learning for prostate cancer tissue classification in infrared spectroscopy","volume":"98","author":"L O\u2019Leary","year":"2026","journal-title":"Anal Chem"}],"updated-by":[{"DOI":"10.1371\/journal.pcbi.1013581","type":"new_version","label":"New version","source":"publisher","updated":{"date-parts":[[2026,6,4]],"date-time":"2026-06-04T00:00:00Z","timestamp":1780531200000}}],"container-title":["PLOS Computational Biology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pcbi.1013581","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,4]],"date-time":"2026-06-04T17:40:56Z","timestamp":1780594856000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pcbi.1013581"}},"subtitle":[],"editor":[{"given":"Jinshan","family":"Xu","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,5,26]]},"references-count":37,"journal-issue":{"issue":"5","published-online":{"date-parts":[[2026,5,26]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pcbi.1013581","relation":{},"ISSN":["1553-7358"],"issn-type":[{"value":"1553-7358","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,26]]}}}