Leeat Keren
@leeat-keren.bsky.social
Assistant Prof. at Weizmann Institute of Science. Spatial Proteomics, Immunology, Cancer, Imaging, CompBio.
Finally, we identify inter-crypt heterogeneity in deterioration with spatially confined immune microenvironments. Deteriorated crypts may neighbor healthy, suggesting local processes driving this compartmentalization and underscoring the importance of spatially-resolved analyses.
November 6, 2025 at 5:51 AM
Finally, we identify inter-crypt heterogeneity in deterioration with spatially confined immune microenvironments. Deteriorated crypts may neighbor healthy, suggesting local processes driving this compartmentalization and underscoring the importance of spatially-resolved analyses.
To examine this we need to distinguish immune cells from the host and donor. How? Cool trick! We imaged the X and Y chromosomes in sex-mismatched samples. Turns out that host T and plasma cells dominate the gut in the first month following transplantation and persist for months!
November 6, 2025 at 5:51 AM
To examine this we need to distinguish immune cells from the host and donor. How? Cool trick! We imaged the X and Y chromosomes in sex-mismatched samples. Turns out that host T and plasma cells dominate the gut in the first month following transplantation and persist for months!
But why do some patients have more CD8s and others more macs? Turns out that time after tranplantation is a major correlate for immune composition. Could it be that immune reconstitution from the hematopoietic stem cell transplant differs across immune cell types?
November 6, 2025 at 5:51 AM
But why do some patients have more CD8s and others more macs? Turns out that time after tranplantation is a major correlate for immune composition. Could it be that immune reconstitution from the hematopoietic stem cell transplant differs across immune cell types?
How does this change in GVHD? We know that donor T cells attack the recipient, but is this the whole story? No! CD8s were only enriched in some patients. We suggest a major role for plasma cells, macs and neutrophils, which correlate with worse clinical manifestation.
November 6, 2025 at 5:51 AM
How does this change in GVHD? We know that donor T cells attack the recipient, but is this the whole story? No! CD8s were only enriched in some patients. We suggest a major role for plasma cells, macs and neutrophils, which correlate with worse clinical manifestation.
We used MIBI to profile 59 diagnostic biopsies from patients with GI GVHD and 18 healthy controls. Healthy gut was similar in its organization across individuals: CD4s in the base of the crypts, plasma in the bottom of the villi and macs at the tip, with some differences between colon and duodenum.
November 6, 2025 at 5:51 AM
We used MIBI to profile 59 diagnostic biopsies from patients with GI GVHD and 18 healthy controls. Healthy gut was similar in its organization across individuals: CD4s in the base of the crypts, plasma in the bottom of the villi and macs at the tip, with some differences between colon and duodenum.
To make it accessible, we built CellTune into an intuitive software, adding advanced capabilities for visualization, gating, annotation and more. CellTune was designed specifically for spatial proteomics and has been a game-changer for our lab – we’ve now used it in 6 projects. 5/7
May 15, 2025 at 1:59 PM
To make it accessible, we built CellTune into an intuitive software, adding advanced capabilities for visualization, gating, annotation and more. CellTune was designed specifically for spatial proteomics and has been a game-changer for our lab – we’ve now used it in 6 projects. 5/7
To benchmark accuracy, we generated gold standard consensus labels from 3 manual annotators. CellTune outperformed every existing method in both precision and recall! It also identified more granular cell types! 4/7
May 15, 2025 at 1:59 PM
To benchmark accuracy, we generated gold standard consensus labels from 3 manual annotators. CellTune outperformed every existing method in both precision and recall! It also identified more granular cell types! 4/7
CellTune’s core innovation is an optimized human-in-the-loop workflow. It trains a model and iteratively refines it by prioritizing information-rich cells for human curation. This approach improves classification accuracy and uncovers rare or novel cell types. 3/7
May 15, 2025 at 1:59 PM
CellTune’s core innovation is an optimized human-in-the-loop workflow. It trains a model and iteratively refines it by prioritizing information-rich cells for human curation. This approach improves classification accuracy and uncovers rare or novel cell types. 3/7
New tools for cell classification in spatial proteomics emerge frequently—but without reliable labels, we’re stuck in a loop comparing inaccurate new predictions to inaccurate old ones.
CellTune delivers accurate results by learning from you! 2/7
CellTune delivers accurate results by learning from you! 2/7
May 15, 2025 at 1:59 PM
New tools for cell classification in spatial proteomics emerge frequently—but without reliable labels, we’re stuck in a loop comparing inaccurate new predictions to inaccurate old ones.
CellTune delivers accurate results by learning from you! 2/7
CellTune delivers accurate results by learning from you! 2/7
🚨Preprint Alert!🚨
Cell classification is one of the most difficult tasks in analyzing spatial data. We present CellTune - a powerful toolkit for accelerating biological discovery in spatial proteomics.
📄 www.biorxiv.org/content/10.1...
👇🧵1/7
Cell classification is one of the most difficult tasks in analyzing spatial data. We present CellTune - a powerful toolkit for accelerating biological discovery in spatial proteomics.
📄 www.biorxiv.org/content/10.1...
👇🧵1/7
May 15, 2025 at 1:59 PM
🚨Preprint Alert!🚨
Cell classification is one of the most difficult tasks in analyzing spatial data. We present CellTune - a powerful toolkit for accelerating biological discovery in spatial proteomics.
📄 www.biorxiv.org/content/10.1...
👇🧵1/7
Cell classification is one of the most difficult tasks in analyzing spatial data. We present CellTune - a powerful toolkit for accelerating biological discovery in spatial proteomics.
📄 www.biorxiv.org/content/10.1...
👇🧵1/7
We then tested it experimentally. We devised protocols to perform combinatorial staining of proteins, both for fluorescence microscopy and for mass-based imaging. It worked! CombPlex accurately reconstructed the single-protein images of 22 proteins from 5 channels.
March 26, 2025 at 4:12 AM
We then tested it experimentally. We devised protocols to perform combinatorial staining of proteins, both for fluorescence microscopy and for mass-based imaging. It worked! CombPlex accurately reconstructed the single-protein images of 22 proteins from 5 channels.
We first tested this idea in simulations in silico. We used publicly-available CODEX data to simulate an experiment in which 22 proteins are measured in 5 imaging channels. It worked! CombPlex accurately reconstructed the single-protein images.
March 26, 2025 at 4:12 AM
We first tested this idea in simulations in silico. We used publicly-available CODEX data to simulate an experiment in which 22 proteins are measured in 5 imaging channels. It worked! CombPlex accurately reconstructed the single-protein images.
In CombPlex, every protein is 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.
March 26, 2025 at 4:12 AM
In CombPlex, every protein is 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.
TL;DR Multiplexed imaging methods use a separate channel for each protein, inherently limiting their scalability. We devised CombPlex (COMBinatorial multiPLEXing) to exponentially increase the number of proteins that can be measured using any imaging modality from C up to ~2^C.
March 26, 2025 at 4:12 AM
TL;DR Multiplexed imaging methods use a separate channel for each protein, inherently limiting their scalability. We devised CombPlex (COMBinatorial multiPLEXing) to exponentially increase the number of proteins that can be measured using any imaging modality from C up to ~2^C.
@yardenasamuels.bsky.social opening #MICC2025 in Athens - the birthplace of Democracy. Looking forward to an amazing conference!
March 10, 2025 at 7:06 AM
@yardenasamuels.bsky.social opening #MICC2025 in Athens - the birthplace of Democracy. Looking forward to an amazing conference!
We integrated these features into a machine learning model to predict the development of distant metastases. In cross-validation settings, our model achieved an AUC of 93% and 79% for involved and clean LNs, respectively, with potential implications for clinical diagnosis.
November 27, 2024 at 2:58 PM
We integrated these features into a machine learning model to predict the development of distant metastases. In cross-validation settings, our model achieved an AUC of 93% and 79% for involved and clean LNs, respectively, with potential implications for clinical diagnosis.
In patients with involved LNs, T cell responses correlated with absence of recurrence. In non-involved LNs, protection from metastases was linked to expanded sinuses colocalized with plasmablasts whereas CCR7+ cells and Tregs predicted future development of distant metastases.
November 27, 2024 at 2:58 PM
In patients with involved LNs, T cell responses correlated with absence of recurrence. In non-involved LNs, protection from metastases was linked to expanded sinuses colocalized with plasmablasts whereas CCR7+ cells and Tregs predicted future development of distant metastases.
We generated 254 images of ~2.5M cells. We integrated protein and mRNA data to comprehensively map cell types and states in sLNs and developed a novel deep-learning algorithm to map recurrent microenvironments across patients. This is a one-in-a-kind dataset!
November 27, 2024 at 2:58 PM
We generated 254 images of ~2.5M cells. We integrated protein and mRNA data to comprehensively map cell types and states in sLNs and developed a novel deep-learning algorithm to map recurrent microenvironments across patients. This is a one-in-a-kind dataset!
To address this, we assembled a unique cohort of clean and metastatic sLNs from 69 melanoma patients. Half of the patients in each group developed distant metastases within five years. We performed high-res spatial proteomics and spatial transcriptomics to map the MEs in the LNs.
November 27, 2024 at 2:58 PM
To address this, we assembled a unique cohort of clean and metastatic sLNs from 69 melanoma patients. Half of the patients in each group developed distant metastases within five years. We performed high-res spatial proteomics and spatial transcriptomics to map the MEs in the LNs.
LNs play a dual role in tumor immunity. They drive anti-tumor immunity, but can also promote systemic tolerance. Clinically, sLNs are routinely examined to look for metastases. Could we do more with them? Can we identify pro- and anti-tumorigenic LN states in human patients?
November 27, 2024 at 2:58 PM
LNs play a dual role in tumor immunity. They drive anti-tumor immunity, but can also promote systemic tolerance. Clinically, sLNs are routinely examined to look for metastases. Could we do more with them? Can we identify pro- and anti-tumorigenic LN states in human patients?