Kunlun Wang, Kaoutar Ait-Ahmad, Sam Kupp, Zachary Sims, Eric Cramer, Zeynep Sayar, Jessica Yu, Melissa H Wong, Gordon B Mills, S Ece Eksi, Young Hwan Chang
{"title":"统一组织成像的通用免疫荧光归一化。","authors":"Kunlun Wang, Kaoutar Ait-Ahmad, Sam Kupp, Zachary Sims, Eric Cramer, Zeynep Sayar, Jessica Yu, Melissa H Wong, Gordon B Mills, S Ece Eksi, Young Hwan Chang","doi":"10.1016/j.crmeth.2025.101172","DOIUrl":null,"url":null,"abstract":"<p><p>We present UniFORM, a non-parametric, Python-based pipeline for normalizing multiplex tissue imaging (MTI) data at both the feature and pixel levels. UniFORM employs an automated rigid landmark registration method tailored to the distributional characteristics of MTI, with UniFORM operating without prior distributional assumptions and handling both unimodal and bimodal patterns. By aligning the biologically invariant negative populations, UniFORM removes technical variation while preserving tissue-specific expression patterns in positive populations. Benchmarked on three MTI platforms, UniFORM consistently outperforms existing methods in mitigating batch effects while maintaining biological signal fidelity. This is evidenced by improved marker distribution alignment and positive population preservation, enhanced k-nearest neighbor batch effect test (kBET) and silhouette scores, and more coherent downstream analyses, such as uniform manifold approximation and projection (UMAP) visualizations and Leiden clustering. UniFORM also offers an optional guided fine-tuning mode for complex or heterogeneous datasets. While optimized for fluorescence-based MTI, its scalable design supports broad applications for MTI data normalization, enabling accurate and biologically meaningful interpretations.</p>","PeriodicalId":29773,"journal":{"name":"Cell Reports Methods","volume":" ","pages":"101172"},"PeriodicalIF":4.5000,"publicationDate":"2025-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Toward universal immunofluorescence normalization for multiplex tissue imaging with UniFORM.\",\"authors\":\"Kunlun Wang, Kaoutar Ait-Ahmad, Sam Kupp, Zachary Sims, Eric Cramer, Zeynep Sayar, Jessica Yu, Melissa H Wong, Gordon B Mills, S Ece Eksi, Young Hwan Chang\",\"doi\":\"10.1016/j.crmeth.2025.101172\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>We present UniFORM, a non-parametric, Python-based pipeline for normalizing multiplex tissue imaging (MTI) data at both the feature and pixel levels. UniFORM employs an automated rigid landmark registration method tailored to the distributional characteristics of MTI, with UniFORM operating without prior distributional assumptions and handling both unimodal and bimodal patterns. By aligning the biologically invariant negative populations, UniFORM removes technical variation while preserving tissue-specific expression patterns in positive populations. Benchmarked on three MTI platforms, UniFORM consistently outperforms existing methods in mitigating batch effects while maintaining biological signal fidelity. This is evidenced by improved marker distribution alignment and positive population preservation, enhanced k-nearest neighbor batch effect test (kBET) and silhouette scores, and more coherent downstream analyses, such as uniform manifold approximation and projection (UMAP) visualizations and Leiden clustering. UniFORM also offers an optional guided fine-tuning mode for complex or heterogeneous datasets. While optimized for fluorescence-based MTI, its scalable design supports broad applications for MTI data normalization, enabling accurate and biologically meaningful interpretations.</p>\",\"PeriodicalId\":29773,\"journal\":{\"name\":\"Cell Reports Methods\",\"volume\":\" \",\"pages\":\"101172\"},\"PeriodicalIF\":4.5000,\"publicationDate\":\"2025-09-15\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Cell Reports Methods\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1016/j.crmeth.2025.101172\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"2025/9/8 0:00:00\",\"PubModel\":\"Epub\",\"JCR\":\"Q1\",\"JCRName\":\"BIOCHEMICAL RESEARCH METHODS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Cell Reports Methods","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1016/j.crmeth.2025.101172","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2025/9/8 0:00:00","PubModel":"Epub","JCR":"Q1","JCRName":"BIOCHEMICAL RESEARCH METHODS","Score":null,"Total":0}
Toward universal immunofluorescence normalization for multiplex tissue imaging with UniFORM.
We present UniFORM, a non-parametric, Python-based pipeline for normalizing multiplex tissue imaging (MTI) data at both the feature and pixel levels. UniFORM employs an automated rigid landmark registration method tailored to the distributional characteristics of MTI, with UniFORM operating without prior distributional assumptions and handling both unimodal and bimodal patterns. By aligning the biologically invariant negative populations, UniFORM removes technical variation while preserving tissue-specific expression patterns in positive populations. Benchmarked on three MTI platforms, UniFORM consistently outperforms existing methods in mitigating batch effects while maintaining biological signal fidelity. This is evidenced by improved marker distribution alignment and positive population preservation, enhanced k-nearest neighbor batch effect test (kBET) and silhouette scores, and more coherent downstream analyses, such as uniform manifold approximation and projection (UMAP) visualizations and Leiden clustering. UniFORM also offers an optional guided fine-tuning mode for complex or heterogeneous datasets. While optimized for fluorescence-based MTI, its scalable design supports broad applications for MTI data normalization, enabling accurate and biologically meaningful interpretations.