Louis Benedikt Kuemmerle
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Louis Benedikt Kuemmerle has a total of 90 co-authors. Here are the top co-authors by number of shared publications:
- Fabian Joachim Theis
Wellcome Sanger Institute, Wellcome Genome Campus, Cambridge, UK
8 shared publications - Malte Daniel Luecken
Institute of Computational Biology, Helmholtz Munich, Neuherberg, Germany
5 shared publications - Mats Nilsson
Science for Life Laboratory, Department of Biochemistry and Biophysics, Stockholm University, Solna, Sweden
3 shared publications - Naveed Ishaque
Center of Digital Health, Berlin Institute of Health at Charité – Universitätsmedizin Berlin, Berlin, Germany
3 shared publications - Sergio Marco Salas
Science for Life Laboratory, Department of Biochemistry and Biophysics, Stockholm University, Stockholm, Sweden
3 shared publications - Alexandros Sountoulidis
Science for Life Laboratory, Department of Molecular Biosciences, Wenner-Gren Institute, Stockholm University, Stockholm, Sweden
2 shared publications
Latest Publications
- Open Access
Benchmarking gene expression reconstruction from single-cell latent representations
Single-cell transcriptomics is typically modeled in low-dimensional latent representations that improve the signal-to-noise ratio of the data. Such representations underpin data integration, cell state discovery, and perturbation prediction, with applications ranging from large-scale organ atlases to latent trajectory modeling. Recent virtual cell approaches further leverage these representations to predict cellular responses as distributional shifts in latent space. Each of these applications ultimately requires faithful gene expression reconstruction from latent spaces for biological interpretation, enabling gene-level analysis of predicted perturbed or batch-corrected cells. Yet representation choice is typically treated as an implementation detail rather than a primary modeling decision, with no systematic evaluation of how well latent representations support gene expression reconstruction. Here, we introduce ReconEval, a benchmark for evaluating gene expression reconstruction from single-cell latent spaces. We benchmark two classes of latent representations: end-to-end trained models such as PCA, autoencoders, and variational autoencoders, and pretrained single-cell foundation model embeddings coupled to newly trained decoders. Reconstruction is evaluated both directly and after latent-space perturbation prediction. Across perturbational and observational datasets totaling over 100 million cells, our metric suite quantifies statistical fidelity; biological signal preservation, including differential expression, coexpression, cell-cycle structure, cytokine response and pathway activity; and perturbation-specific effects. We find that autoencoders achieve the strongest stand-alone reconstruction at low dimensionality, while variational regularization does not improve generalization in reconstruction. Frozen foundation model embeddings retain recoverable gene-level information, with reconstruction quality depending strongly on decoder architecture and pretraining objective. In latent perturbation modeling, high-dimensional PCA matches foundation model embeddings, while low-dimensional AE embeddings are optimal for flow-based generative models. Overall, reconstruction depends critically on the interplay between representation and downstream model, and simpler representations can outperform complex alternatives given appropriate capacity. Our benchmark establishes reconstruction as a critical evaluation axis for single-cell foundation models. We envision it improving the biological interpretability of latent-space modeling, a prerequisite for future virtual cell models to be validated by domain experts and grounded in biology.
Open AccessBenchmarking gene expression reconstruction from single-cell latent representations
Single-cell transcriptomics is typically modeled in low-dimensional latent representations that improve the signal-to-noise ratio of the data. Such representations underpin data integration, cell state discovery, and perturbation prediction, with applications ranging from large-scale organ atlases to latent trajectory modeling. Recent virtual cell approaches further leverage these representations to predict cellular responses as distributional shifts in latent space. Each of these applications ultimately requires faithful gene expression reconstruction from latent spaces for biological interpretation, enabling gene-level analysis of predicted perturbed or batch-corrected cells. Yet representation choice is typically treated as an implementation detail rather than a primary modeling decision, with no systematic evaluation of how well latent representations support gene expression reconstruction. Here, we introduce ReconEval, a benchmark for evaluating gene expression reconstruction from single-cell latent spaces. We benchmark two classes of latent representations: end-to-end trained models such as PCA, autoencoders, and variational autoencoders, and pretrained single-cell foundation model embeddings coupled to newly trained decoders. Reconstruction is evaluated both directly and after latent-space perturbation prediction. Across perturbational and observational datasets totaling over 100 million cells, our metric suite quantifies statistical fidelity; biological signal preservation, including differential expression, coexpression, cell-cycle structure, cytokine response and pathway activity; and perturbation-specific effects. We find that autoencoders achieve the strongest stand-alone reconstruction at low dimensionality, while variational regularization does not improve generalization in reconstruction. Frozen foundation model embeddings retain recoverable gene-level information, with reconstruction quality depending strongly on decoder architecture and pretraining objective. In latent perturbation modeling, high-dimensional PCA matches foundation model embeddings, while low-dimensional AE embeddings are optimal for flow-based generative models. Overall, reconstruction depends critically on the interplay between representation and downstream model, and simpler representations can outperform complex alternatives given appropriate capacity. Our benchmark establishes reconstruction as a critical evaluation axis for single-cell foundation models. We envision it improving the biological interpretability of latent-space modeling, a prerequisite for future virtual cell models to be validated by domain experts and grounded in biology.
- 10 February 2026
- Published by Springer Nature in Nature Biotechnology
Identifying 3D signal overlaps in spatial transcriptomics data with ovrlpy
Imaging-based spatially resolved transcriptomics can localize transcripts within tissue sections in three dimensions. However, cell segmentation, which assigns transcripts to cells, is usually performed in two dimensions and spatial doublets in the vertical dimension result in segmented cells containing transcripts originating from multiple cell types. Here we present a computational tool called ovrlpy that identifies overlapping cells, tissue folds and inaccurate cell segmentation by analyzing transcript localization in three dimensions.
- 10 February 2026
- Published by Springer Nature in Nature Biotechnology
Identifying 3D signal overlaps in spatial transcriptomics data with ovrlpy
Imaging-based spatially resolved transcriptomics can localize transcripts within tissue sections in three dimensions. However, cell segmentation, which assigns transcripts to cells, is usually performed in two dimensions and spatial doublets in the vertical dimension result in segmented cells containing transcripts originating from multiple cell types. Here we present a computational tool called ovrlpy that identifies overlapping cells, tissue folds and inaccurate cell segmentation by analyzing transcript localization in three dimensions.
- Open Access
Brain Region-Specific Oligodendrocyte States Highlight Mitochondrial Gene Upregulation and Loss of Canonical Identity Signatures in Alzheimer’s Disease
Oligodendrocyte dysfunction and heterogeneity are emerging as key contributors to Alzheimer’s disease (AD) pathology, though cellular mechanisms remain unclear. Using single-nucleus RNA sequencing across three cortical regions from the same individuals, we identified region- and disease-specific transcriptomic changes, including increased mitochondrial genes and loss of cell type signatures. These results highlight the importance of biologically-informed quality control and underscore the value of multi-regional analysis for understanding AD pathogenesis and progression.
Open AccessBrain Region-Specific Oligodendrocyte States Highlight Mitochondrial Gene Upregulation and Loss of Canonical Identity Signatures in Alzheimer’s Disease
Oligodendrocyte dysfunction and heterogeneity are emerging as key contributors to Alzheimer’s disease (AD) pathology, though cellular mechanisms remain unclear. Using single-nucleus RNA sequencing across three cortical regions from the same individuals, we identified region- and disease-specific transcriptomic changes, including increased mitochondrial genes and loss of cell type signatures. These results highlight the importance of biologically-informed quality control and underscore the value of multi-regional analysis for understanding AD pathogenesis and progression.
- Top Cited PapersOpen Access
- 13 March 2025
- Published by Springer Nature in Nature Methods
- Vol. 22 (4) , 813-823
Optimizing Xenium In Situ data utility by quality assessment and best-practice analysis workflows
The Xenium In Situ platform is a new spatial transcriptomics product commercialized by 10x Genomics, capable of mapping hundreds of genes in situ at subcellular resolution. Given the multitude of commercially available spatial transcriptomics technologies, recommendations in choice of platform and analysis guidelines are increasingly important. Herein, we explore 25 Xenium datasets generated from multiple tissues and species, comparing scalability, resolution, data quality, capacities and limitations with eight other spatially resolved transcriptomics technologies and commercial platforms. In addition, we benchmark the performance of multiple open-source computational tools, when applied to Xenium datasets, in tasks including preprocessing, cell segmentation, selection of spatially variable features and domain identification. This study serves as an independent analysis of the performance of Xenium, and provides best practices and recommendations for analysis of such datasets.
- 13 March 2025
- Published by Springer Nature in Nature Methods
- Vol. 22 (4) , 813-823
Top Cited PapersOpen AccessOptimizing Xenium In Situ data utility by quality assessment and best-practice analysis workflows
The Xenium In Situ platform is a new spatial transcriptomics product commercialized by 10x Genomics, capable of mapping hundreds of genes in situ at subcellular resolution. Given the multitude of commercially available spatial transcriptomics technologies, recommendations in choice of platform and analysis guidelines are increasingly important. Herein, we explore 25 Xenium datasets generated from multiple tissues and species, comparing scalability, resolution, data quality, capacities and limitations with eight other spatially resolved transcriptomics technologies and commercial platforms. In addition, we benchmark the performance of multiple open-source computational tools, when applied to Xenium datasets, in tasks including preprocessing, cell segmentation, selection of spatially variable features and domain identification. This study serves as an independent analysis of the performance of Xenium, and provides best practices and recommendations for analysis of such datasets.
- Open Access
- 18 November 2024
- Published by Springer Nature in Nature Methods
- Vol. 21 (12) , 2260-2270
Probe set selection for targeted spatial transcriptomics
Targeted spatial transcriptomic methods capture the topology of cell types and states in tissues at single-cell and subcellular resolution by measuring the expression of a predefined set of genes. The selection of an optimal set of probed genes is crucial for capturing the spatial signals present in a tissue. This requires selecting the most informative, yet minimal, set of genes to profile (gene set selection) for which it is possible to build probes (probe design). However, current selections often rely on marker genes, precluding them from detecting continuous spatial signals or new states. We present Spapros, an end-to-end probe set selection pipeline that optimizes both gene set specificity for cell type identification and within-cell type expression variation to resolve spatially distinct populations while considering prior knowledge as well as probe design and expression constraints. We evaluated Spapros and show that it outperforms other selection approaches in both cell type recovery and recovering expression variation beyond cell types. Furthermore, we used Spapros to design a single-cell resolution in situ hybridization on tissues (SCRINSHOT) experiment of adult lung tissue to demonstrate how probes selected with Spapros identify cell types of interest and detect spatial variation even within cell types.
- 18 November 2024
- Published by Springer Nature in Nature Methods
- Vol. 21 (12) , 2260-2270
Open AccessProbe set selection for targeted spatial transcriptomics
Targeted spatial transcriptomic methods capture the topology of cell types and states in tissues at single-cell and subcellular resolution by measuring the expression of a predefined set of genes. The selection of an optimal set of probed genes is crucial for capturing the spatial signals present in a tissue. This requires selecting the most informative, yet minimal, set of genes to profile (gene set selection) for which it is possible to build probes (probe design). However, current selections often rely on marker genes, precluding them from detecting continuous spatial signals or new states. We present Spapros, an end-to-end probe set selection pipeline that optimizes both gene set specificity for cell type identification and within-cell type expression variation to resolve spatially distinct populations while considering prior knowledge as well as probe design and expression constraints. We evaluated Spapros and show that it outperforms other selection approaches in both cell type recovery and recovering expression variation beyond cell types. Furthermore, we used Spapros to design a single-cell resolution in situ hybridization on tissues (SCRINSHOT) experiment of adult lung tissue to demonstrate how probes selected with Spapros identify cell types of interest and detect spatial variation even within cell types.
- Open Access
Optimizing Xenium In Situ data utility by quality assessment and best practice analysis workflows
The Xenium In Situ platform is a new spatial transcriptomics product commercialized by 10X Genomics capable of mapping hundreds of transcripts in situ at a subcellular resolution. Given the multitude of commercially available spatial transcriptomics technologies, recommendations in choice of platform and analysis guidelines are increasingly important. Herein, we explore eight preview Xenium datasets of the mouse brain and two of human breast cancer by comparing scalability, resolution, data quality, capacities and limitations with eight other spatially resolved transcriptomics technologies. In addition, we benchmarked the performance of multiple open source computational tools when applied to Xenium datasets in tasks including cell segmentation, segmentation-free analysis, selection of spatially variable genes and domain identification, among others. This study serves as the first independent analysis of the performance of Xenium, and provides best-practices and recommendations for analysis of such datasets.
Open AccessOptimizing Xenium In Situ data utility by quality assessment and best practice analysis workflows
The Xenium In Situ platform is a new spatial transcriptomics product commercialized by 10X Genomics capable of mapping hundreds of transcripts in situ at a subcellular resolution. Given the multitude of commercially available spatial transcriptomics technologies, recommendations in choice of platform and analysis guidelines are increasingly important. Herein, we explore eight preview Xenium datasets of the mouse brain and two of human breast cancer by comparing scalability, resolution, data quality, capacities and limitations with eight other spatially resolved transcriptomics technologies. In addition, we benchmarked the performance of multiple open source computational tools when applied to Xenium datasets in tasks including cell segmentation, segmentation-free analysis, selection of spatially variable genes and domain identification, among others. This study serves as the first independent analysis of the performance of Xenium, and provides best-practices and recommendations for analysis of such datasets.
- Open Access
Probe set selection for targeted spatial transcriptomics
Targeted spatial transcriptomics methods capture the topology of cell types and states in tissues at single cell- and subcellular resolution by measuring the expression of a predefined set of genes. The selection of an optimal set of probed genes is crucial for capturing and interpreting the spatial signals present in a tissue. However, current selections often rely on marker genes, precluding them from detecting continuous spatial signals or novel states. We present Spapros, an end-to-end probe set selection pipeline that optimizes both probe set specificity for cell type identification and within-cell-type expression variation to resolve spatially distinct populations while taking into account prior knowledge, as well as probe design and expression constraints. To facilitate data analysis and interpretation, Spapros also provides rules for cell type identification. We evaluated Spapros by selecting probes on 6 different data sets and built an evaluation pipeline with 12 quality metrics to find that Spapros outperforms other selection approaches in both cell type recovery and recovering expression variation beyond cell types. Furthermore, we used Spapros to design a SCRINSHOT experiment of adult lung tissue to demonstrate how probes selected with Spapros identify cell types of interest and detect spatial variation even within cell types. Spapros enables optimal probe set selection, probe set evaluation, and probe design, as a freely available Python package.
Open AccessProbe set selection for targeted spatial transcriptomics
Targeted spatial transcriptomics methods capture the topology of cell types and states in tissues at single cell- and subcellular resolution by measuring the expression of a predefined set of genes. The selection of an optimal set of probed genes is crucial for capturing and interpreting the spatial signals present in a tissue. However, current selections often rely on marker genes, precluding them from detecting continuous spatial signals or novel states. We present Spapros, an end-to-end probe set selection pipeline that optimizes both probe set specificity for cell type identification and within-cell-type expression variation to resolve spatially distinct populations while taking into account prior knowledge, as well as probe design and expression constraints. To facilitate data analysis and interpretation, Spapros also provides rules for cell type identification. We evaluated Spapros by selecting probes on 6 different data sets and built an evaluation pipeline with 12 quality metrics to find that Spapros outperforms other selection approaches in both cell type recovery and recovering expression variation beyond cell types. Furthermore, we used Spapros to design a SCRINSHOT experiment of adult lung tissue to demonstrate how probes selected with Spapros identify cell types of interest and detect spatial variation even within cell types. Spapros enables optimal probe set selection, probe set evaluation, and probe design, as a freely available Python package.
- Top Cited PapersOpen Access
- 31 January 2022
- Published by Springer Nature in Nature Methods
- Vol. 19 (2) , 171-178
Squidpy: a scalable framework for spatial omics analysis
Spatial omics data are advancing the study of tissue organization and cellular communication at an unprecedented scale. Flexible tools are required to store, integrate and visualize the large diversity of spatial omics data. Here, we present Squidpy, a Python framework that brings together tools from omics and image analysis to enable scalable description of spatial molecular data, such as transcriptome or multivariate proteins. Squidpy provides efficient infrastructure and numerous analysis methods that allow to efficiently store, manipulate and interactively visualize spatial omics data. Squidpy is extensible and can be interfaced with a variety of already existing libraries for the scalable analysis of spatial omics data.
- 31 January 2022
- Published by Springer Nature in Nature Methods
- Vol. 19 (2) , 171-178
Top Cited PapersOpen AccessSquidpy: a scalable framework for spatial omics analysis
Spatial omics data are advancing the study of tissue organization and cellular communication at an unprecedented scale. Flexible tools are required to store, integrate and visualize the large diversity of spatial omics data. Here, we present Squidpy, a Python framework that brings together tools from omics and image analysis to enable scalable description of spatial molecular data, such as transcriptome or multivariate proteins. Squidpy provides efficient infrastructure and numerous analysis methods that allow to efficiently store, manipulate and interactively visualize spatial omics data. Squidpy is extensible and can be interfaced with a variety of already existing libraries for the scalable analysis of spatial omics data.
- Open Access
Squidpy: a scalable framework for spatial single cell analysis
Spatial omics data are advancing the study of tissue organization and cellular communication at an unprecedented scale. Here, we present Squidpy, a Python framework that brings together tools from omics and image analysis to enable scalable description of spatial molecular data, such as transcriptome or multivariate proteins. Squidpy provides both infrastructure and numerous analysis methods that allow to efficiently store, manipulate and interactively visualize spatial omics data.
Open AccessSquidpy: a scalable framework for spatial single cell analysis
Spatial omics data are advancing the study of tissue organization and cellular communication at an unprecedented scale. Here, we present Squidpy, a Python framework that brings together tools from omics and image analysis to enable scalable description of spatial molecular data, such as transcriptome or multivariate proteins. Squidpy provides both infrastructure and numerous analysis methods that allow to efficiently store, manipulate and interactively visualize spatial omics data.