High-dimensional imaging using combinatorial channel multiplexing and deep learning

Raz Ben-Uri; Lior Ben Shabat; Dana Shainshein; Omer Bar-Tal; Yuval Bussi; Noa Maimon; Tal Keidar Haran; Idan Milo; Inna Goliand; Yoseph Addadi; Tomer Meir Salame; Alexander Rochwarger; Christian M. Schürch; Shai Bagon; Ofer Elhanani; Leeat Keren

doi:10.1038/s41587-025-02585-0

High-dimensional imaging using combinatorial channel multiplexing and deep learning

Raz Ben-Uri, Lior Ben Shabat, Dana Shainshein, Omer Bar-Tal, Yuval Bussi, Noa Maimon, Tal Keidar Haran, Idan Milo, Inna Goliand, Yoseph Addadi, Tomer Meir Salame, Alexander Rochwarger, Christian M. Schürch, Shai Bagon, Ofer Elhanani, Leeat Keren

Research output: Contribution to journal › Article › peer-review

1 Scopus citations

Abstract

Understanding tissue structure and function requires tools that quantify the expression of multiple proteins at single-cell resolution while preserving spatial information. Current imaging technologies use a separate channel for each protein, limiting throughput and scalability. Here, we present combinatorial multiplexing (CombPlex), a combinatorial staining platform coupled with an algorithmic framework to exponentially increase the number of measured proteins. Every protein can be imaged in several channels and every channel contains agglomerated images of several proteins. These combinatorically compressed images are then decompressed to individual protein images using deep learning. We achieve accurate reconstruction when compressing the stains of 22 proteins to five imaging channels. We demonstrate the approach both in fluorescence microscopy and in mass-based imaging and show successful application across multiple tissues and cancer types. CombPlex can escalate the number of proteins measured by any imaging modality, without the need for specialized instrumentation.

Original language	English
Article number	eaar7042
Journal	Nature Biotechnology
DOIs	https://doi.org/10.1038/s41587-025-02585-0
State	Accepted/In press - 1 Jan 2025
Externally published	Yes

ASJC Scopus subject areas

Biotechnology
Bioengineering
Applied Microbiology and Biotechnology
Molecular Medicine
Biomedical Engineering

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1038/s41587-025-02585-0

Cite this

Ben-Uri, R., Ben Shabat, L., Shainshein, D., Bar-Tal, O., Bussi, Y., Maimon, N., Keidar Haran, T., Milo, I., Goliand, I., Addadi, Y., Salame, T. M., Rochwarger, A., Schürch, C. M., Bagon, S., Elhanani, O., & Keren, L. (Accepted/In press). High-dimensional imaging using combinatorial channel multiplexing and deep learning. Nature Biotechnology, Article eaar7042. https://doi.org/10.1038/s41587-025-02585-0

@article{41525a82d19746ffbe79b4e4c4de8287,

title = "High-dimensional imaging using combinatorial channel multiplexing and deep learning",

abstract = "Understanding tissue structure and function requires tools that quantify the expression of multiple proteins at single-cell resolution while preserving spatial information. Current imaging technologies use a separate channel for each protein, limiting throughput and scalability. Here, we present combinatorial multiplexing (CombPlex), a combinatorial staining platform coupled with an algorithmic framework to exponentially increase the number of measured proteins. Every protein can be imaged in several channels and every channel contains agglomerated images of several proteins. These combinatorically compressed images are then decompressed to individual protein images using deep learning. We achieve accurate reconstruction when compressing the stains of 22 proteins to five imaging channels. We demonstrate the approach both in fluorescence microscopy and in mass-based imaging and show successful application across multiple tissues and cancer types. CombPlex can escalate the number of proteins measured by any imaging modality, without the need for specialized instrumentation.",

author = "Raz Ben-Uri and \{Ben Shabat\}, Lior and Dana Shainshein and Omer Bar-Tal and Yuval Bussi and Noa Maimon and \{Keidar Haran\}, Tal and Idan Milo and Inna Goliand and Yoseph Addadi and Salame, \{Tomer Meir\} and Alexander Rochwarger and Sch{\"u}rch, \{Christian M.\} and Shai Bagon and Ofer Elhanani and Leeat Keren",

note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive licence to Springer Nature America, Inc. 2025.",

year = "2025",

month = jan,

day = "1",

doi = "10.1038/s41587-025-02585-0",

language = "English",

journal = "Nature Biotechnology",

issn = "1087-0156",

publisher = "Nature Research",

}

TY - JOUR

T1 - High-dimensional imaging using combinatorial channel multiplexing and deep learning

AU - Ben-Uri, Raz

AU - Ben Shabat, Lior

AU - Shainshein, Dana

AU - Bar-Tal, Omer

AU - Bussi, Yuval

AU - Maimon, Noa

AU - Keidar Haran, Tal

AU - Milo, Idan

AU - Goliand, Inna

AU - Addadi, Yoseph

AU - Salame, Tomer Meir

AU - Rochwarger, Alexander

AU - Schürch, Christian M.

AU - Bagon, Shai

AU - Elhanani, Ofer

AU - Keren, Leeat

PY - 2025/1/1

Y1 - 2025/1/1

N2 - Understanding tissue structure and function requires tools that quantify the expression of multiple proteins at single-cell resolution while preserving spatial information. Current imaging technologies use a separate channel for each protein, limiting throughput and scalability. Here, we present combinatorial multiplexing (CombPlex), a combinatorial staining platform coupled with an algorithmic framework to exponentially increase the number of measured proteins. Every protein can be imaged in several channels and every channel contains agglomerated images of several proteins. These combinatorically compressed images are then decompressed to individual protein images using deep learning. We achieve accurate reconstruction when compressing the stains of 22 proteins to five imaging channels. We demonstrate the approach both in fluorescence microscopy and in mass-based imaging and show successful application across multiple tissues and cancer types. CombPlex can escalate the number of proteins measured by any imaging modality, without the need for specialized instrumentation.

AB - Understanding tissue structure and function requires tools that quantify the expression of multiple proteins at single-cell resolution while preserving spatial information. Current imaging technologies use a separate channel for each protein, limiting throughput and scalability. Here, we present combinatorial multiplexing (CombPlex), a combinatorial staining platform coupled with an algorithmic framework to exponentially increase the number of measured proteins. Every protein can be imaged in several channels and every channel contains agglomerated images of several proteins. These combinatorically compressed images are then decompressed to individual protein images using deep learning. We achieve accurate reconstruction when compressing the stains of 22 proteins to five imaging channels. We demonstrate the approach both in fluorescence microscopy and in mass-based imaging and show successful application across multiple tissues and cancer types. CombPlex can escalate the number of proteins measured by any imaging modality, without the need for specialized instrumentation.

UR - https://www.scopus.com/pages/publications/105001013845

U2 - 10.1038/s41587-025-02585-0

DO - 10.1038/s41587-025-02585-0

M3 - Article

C2 - 40133518

AN - SCOPUS:105001013845

SN - 1087-0156

JO - Nature Biotechnology

JF - Nature Biotechnology

M1 - eaar7042

ER -

High-dimensional imaging using combinatorial channel multiplexing and deep learning

Abstract

ASJC Scopus subject areas

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this