Stefan Schrod
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Stefan Schrod has a total of 86 co-authors. Here are the top co-authors by number of shared publications:
- Michael Altenbuchinger
Department of Medical Bioinformatics, University Medical Center Göttingen, Göttingen, Germany
9 shared publications - Helena U. Zacharias
Peter L. Reichertz Institute for Medical Informatics of TU Braunschweig and Hannover Medical School, Hannover Medical School, 30625 Hannover, Germany
5 shared publications - Tim Beißbarth
Department of Medical Bioinformatics, University Medical Center Göttingen, Göttingen, Germany
5 shared publications - Andreas Schäfer
University of Regensburg
4 shared publications - Anagha Joshi
Department of Clinical Science, Computational Biology Unit, University of Bergen, Bergen, Norway
2 shared publications - Fabian Hubert Sinz
Department of Neuroscience and Center for Neuroscience and Artificial Intelligence, Baylor College of Medicine
2 shared publications
Latest Publications
- Open Access
- 7 July 2026
- Published by Springer Nature in Leukemia
- Vol. 40 (9) , 1961-1967
Mantle cell lymphoma artificial intelligence prognostic index using hematoxylin and eosin histology
The clinical course of Mantle Cell Lymphoma (MCL) varies between individual patients. Early detection of risk is crucial to assign MCL patients to novel treatment strategies. Most of the established biomarkers of outcome require specifically trained pathologists or molecular analysis. Here we introduce MAIPI (MCL Artificial Intelligence Prognostic Index), a deep learning algorithm trained only on Hematoxylin and Eosin (H&E) images of diagnostic biopsies of n = 428 MCL patients from clinical trials to assess prognosis. The capability of MAIPI to predict disease outcome was validated in an independent cohort of n = 140 patients treated with immunochemotherapy with and without ibrutinib. MAIPI selects areas of interest by itself and provides prognostic information independent of the MCL International Prognostic Index (MIPI) and Ki67 and without the need of molecular testing or expert pathologists evaluation.
- 7 July 2026
- Published by Springer Nature in Leukemia
- Vol. 40 (9) , 1961-1967
Open AccessMantle cell lymphoma artificial intelligence prognostic index using hematoxylin and eosin histology
The clinical course of Mantle Cell Lymphoma (MCL) varies between individual patients. Early detection of risk is crucial to assign MCL patients to novel treatment strategies. Most of the established biomarkers of outcome require specifically trained pathologists or molecular analysis. Here we introduce MAIPI (MCL Artificial Intelligence Prognostic Index), a deep learning algorithm trained only on Hematoxylin and Eosin (H&E) images of diagnostic biopsies of n = 428 MCL patients from clinical trials to assess prognosis. The capability of MAIPI to predict disease outcome was validated in an independent cohort of n = 140 patients treated with immunochemotherapy with and without ibrutinib. MAIPI selects areas of interest by itself and provides prognostic information independent of the MCL International Prognostic Index (MIPI) and Ki67 and without the need of molecular testing or expert pathologists evaluation.
- Open Access
- 15 April 2026
- Published by Springer Nature in Machine Learning
- Vol. 115 (4) , 96
Adversarial Distribution Balancing for Counterfactual Reasoning
The development of causal prediction models is challenged by the fact that the outcome is only observable for the applied (factual) intervention and not for its alternatives (the so-called counterfactuals); in medicine we only know patients’ survival for the administered drug and not for other therapeutic options. Machine learning approaches for counterfactual reasoning have to deal with both unobserved outcomes and distributional differences due to non-random treatment administration. Unsupervised domain adaptation (UDA) addresses similar issues; one has to deal with unobserved outcomes—the labels of the target domain—and distributional differences between source and target domain. We propose Adversarial Distribution Balancing for Counterfactual Reasoning (ADBCR), which directly uses potential outcome estimates of the counterfactuals to remove spurious causal relations. We show that ADBCR outcompetes state-of-the-art methods on three benchmark datasets, and demonstrate that ADBCR’s performance can be further improved if unlabeled validation data are included in the training procedure to better adapt the model to the validation domain.
- 15 April 2026
- Published by Springer Nature in Machine Learning
- Vol. 115 (4) , 96
Open AccessAdversarial Distribution Balancing for Counterfactual Reasoning
The development of causal prediction models is challenged by the fact that the outcome is only observable for the applied (factual) intervention and not for its alternatives (the so-called counterfactuals); in medicine we only know patients’ survival for the administered drug and not for other therapeutic options. Machine learning approaches for counterfactual reasoning have to deal with both unobserved outcomes and distributional differences due to non-random treatment administration. Unsupervised domain adaptation (UDA) addresses similar issues; one has to deal with unobserved outcomes—the labels of the target domain—and distributional differences between source and target domain. We propose Adversarial Distribution Balancing for Counterfactual Reasoning (ADBCR), which directly uses potential outcome estimates of the counterfactuals to remove spurious causal relations. We show that ADBCR outcompetes state-of-the-art methods on three benchmark datasets, and demonstrate that ADBCR’s performance can be further improved if unlabeled validation data are included in the training procedure to better adapt the model to the validation domain.
- Open Access
Genome-scale mapping of variant, enhancer and gene function in primary human CD4+ T cells
CD4+ T cells harbor a disproportionate enrichment of immune disease risk loci and represent the primary cellular context for immune disease biology, yet the genes and regulatory programs these variants affect remain largely unknown. We combined targeted Perturb-seq of 1,032 cis -regulatory elements (CREs) overlapping 4,724 variants across 14 immune diseases with genome-wide Perturb-seq of all expressed genes in primary human CD4+ T cells, spanning 4.1 million cells. We identified 626 CRE-gene pairs, and connected CRE targets to downstream regulatory cascades. At the TYK2 and DEXI /CLEC16A loci, we resolved target genes and linked noncoding variants to inflammatory and metabolic programs. Across diseases, we revealed that dispersed variants converged on shared and disease-specific programs. Our work provides a framework for tracing variant-to-CRE-to-gene-to-network in disease-relevant primary cells.
Open AccessGenome-scale mapping of variant, enhancer and gene function in primary human CD4+ T cells
CD4+ T cells harbor a disproportionate enrichment of immune disease risk loci and represent the primary cellular context for immune disease biology, yet the genes and regulatory programs these variants affect remain largely unknown. We combined targeted Perturb-seq of 1,032 cis -regulatory elements (CREs) overlapping 4,724 variants across 14 immune diseases with genome-wide Perturb-seq of all expressed genes in primary human CD4+ T cells, spanning 4.1 million cells. We identified 626 CRE-gene pairs, and connected CRE targets to downstream regulatory cascades. At the TYK2 and DEXI /CLEC16A loci, we resolved target genes and linked noncoding variants to inflammatory and metabolic programs. Across diseases, we revealed that dispersed variants converged on shared and disease-specific programs. Our work provides a framework for tracing variant-to-CRE-to-gene-to-network in disease-relevant primary cells.
- Top Cited Papers
- 2 January 2026
- Published by Springer Nature in Nature Reviews Genetics
- Vol. 27 (5) , 349-370
Interpretation, extrapolation and perturbation of single cells
Single-cell analyses have transitioned from descriptive atlasing towards inferring causal effects and mechanistic relationships that capture cellular logic. Technological advances and the growing scale of observational and interventional datasets have fuelled the development of machine learning methods aimed at identifying such dependencies and extrapolating perturbation effects. Here, we review and connect these approaches according to their modelling concepts (including representation learning, causal inference, mechanistic discovery, disentanglement and population tracing), underlying assumptions and downstream tasks. We propose a unifying ontology to guide practitioners in selecting the most suitable methods for a given biological question, with detailed technical descriptions provided in an online resource. Finally, we identify promising computational directions and underexplored data properties that could pave the way for future developments.
- 2 January 2026
- Published by Springer Nature in Nature Reviews Genetics
- Vol. 27 (5) , 349-370
Top Cited PapersInterpretation, extrapolation and perturbation of single cells
Single-cell analyses have transitioned from descriptive atlasing towards inferring causal effects and mechanistic relationships that capture cellular logic. Technological advances and the growing scale of observational and interventional datasets have fuelled the development of machine learning methods aimed at identifying such dependencies and extrapolating perturbation effects. Here, we review and connect these approaches according to their modelling concepts (including representation learning, causal inference, mechanistic discovery, disentanglement and population tracing), underlying assumptions and downstream tasks. We propose a unifying ontology to guide practitioners in selecting the most suitable methods for a given biological question, with detailed technical descriptions provided in an online resource. Finally, we identify promising computational directions and underexplored data properties that could pave the way for future developments.
- Open Access
- 10 May 2025
- Published by Elsevier in Knowledge-Based Systems
- Vol. 320
FACT: Federated Adaptive Cross Training
- 10 May 2025
- Published by Elsevier in Knowledge-Based Systems
- Vol. 320
Open AccessFACT: Federated Adaptive Cross Training
- Open Access
- 4 September 2024
- Published by Cold Spring Harbor Laboratory in Genome Research
- Vol. 34 (9) , 1371-1383
Spatial Cellular Networks from omics data with SpaCeNet
Advances in omics technologies have allowed spatially resolved molecular profiling of single cells, providing a window not only into the diversity and distribution of cell types within a tissue, but also into the effects of interactions between cells in shaping the transcriptional landscape. Cells send chemical and mechanical signals which are received by other cells, where they can subsequently initiate context-specific gene regulatory responses. These interactions and their responses shape the individual molecular phenotype of a cell in a given microenvironment. RNAs or proteins measured in individual cells, together with the cells' spatial distribution, provide invaluable information about these mechanisms and the regulation of genes beyond processes occurring independently in each individual cell. “SpaCeNet” is a method designed to elucidate both the intracellular molecular networks (how molecular variables affect each other within the cell) and the intercellular molecular networks (how cells affect molecular variables in their neighbors). This is achieved by estimating conditional independence (CI) relations between captured variables within individual cells and by disentangling these from CI relations between variables of different cells.
- 4 September 2024
- Published by Cold Spring Harbor Laboratory in Genome Research
- Vol. 34 (9) , 1371-1383
Open AccessSpatial Cellular Networks from omics data with SpaCeNet
Advances in omics technologies have allowed spatially resolved molecular profiling of single cells, providing a window not only into the diversity and distribution of cell types within a tissue, but also into the effects of interactions between cells in shaping the transcriptional landscape. Cells send chemical and mechanical signals which are received by other cells, where they can subsequently initiate context-specific gene regulatory responses. These interactions and their responses shape the individual molecular phenotype of a cell in a given microenvironment. RNAs or proteins measured in individual cells, together with the cells' spatial distribution, provide invaluable information about these mechanisms and the regulation of genes beyond processes occurring independently in each individual cell. “SpaCeNet” is a method designed to elucidate both the intracellular molecular networks (how molecular variables affect each other within the cell) and the intercellular molecular networks (how cells affect molecular variables in their neighbors). This is achieved by estimating conditional independence (CI) relations between captured variables within individual cells and by disentangling these from CI relations between variables of different cells.
- Open Access
- 28 June 2024
- Published by Oxford University Press (OUP) in Bioinformatics
- Vol. 40 (Supplement) , i100-i109
Adaptive digital tissue deconvolution
Motivation: The inference of cellular compositions from bulk and spatial transcriptomics data increasingly complements data analyses. Multiple computational approaches were suggested and recently, machine learning techniques were developed to systematically improve estimates. Such approaches allow to infer additional, less abundant cell types. However, they rely on training data which do not capture the full biological diversity encountered in transcriptomics analyses; data can contain cellular contributions not seen in the training data and as such, analyses can be biased or blurred. Thus, computational approaches have to deal with unknown, hidden contributions. Moreover, most methods are based on cellular archetypes which serve as a reference; e.g. a generic T-cell profile is used to infer the proportion of T-cells. It is well known that cells adapt their molecular phenotype to the environment and that pre-specified cell archetypes can distort the inference of cellular compositions. Results: We propose Adaptive Digital Tissue Deconvolution (ADTD) to estimate cellular proportions of pre-selected cell types together with possibly unknown and hidden background contributions. Moreover, ADTD adapts prototypic reference profiles to the molecular environment of the cells, which further resolves cell-type specific gene regulation from bulk transcriptomics data. We verify this in simulation studies and demonstrate that ADTD improves existing approaches in estimating cellular compositions. In an application to bulk transcriptomics data from breast cancer patients, we demonstrate that ADTD provides insights into cell-type specific molecular differences between breast cancer subtypes. Availability and implementation: A python implementation of ADTD and a tutorial are available at Gitlab and zenodo (doi:10.5281/zenodo.7548362).
- 28 June 2024
- Published by Oxford University Press (OUP) in Bioinformatics
- Vol. 40 (Supplement) , i100-i109
Open AccessAdaptive digital tissue deconvolution
Motivation: The inference of cellular compositions from bulk and spatial transcriptomics data increasingly complements data analyses. Multiple computational approaches were suggested and recently, machine learning techniques were developed to systematically improve estimates. Such approaches allow to infer additional, less abundant cell types. However, they rely on training data which do not capture the full biological diversity encountered in transcriptomics analyses; data can contain cellular contributions not seen in the training data and as such, analyses can be biased or blurred. Thus, computational approaches have to deal with unknown, hidden contributions. Moreover, most methods are based on cellular archetypes which serve as a reference; e.g. a generic T-cell profile is used to infer the proportion of T-cells. It is well known that cells adapt their molecular phenotype to the environment and that pre-specified cell archetypes can distort the inference of cellular compositions. Results: We propose Adaptive Digital Tissue Deconvolution (ADTD) to estimate cellular proportions of pre-selected cell types together with possibly unknown and hidden background contributions. Moreover, ADTD adapts prototypic reference profiles to the molecular environment of the cells, which further resolves cell-type specific gene regulation from bulk transcriptomics data. We verify this in simulation studies and demonstrate that ADTD improves existing approaches in estimating cellular compositions. In an application to bulk transcriptomics data from breast cancer patients, we demonstrate that ADTD provides insights into cell-type specific molecular differences between breast cancer subtypes. Availability and implementation: A python implementation of ADTD and a tutorial are available at Gitlab and zenodo (doi:10.5281/zenodo.7548362).
- Open Access
- 28 June 2024
- Published by Oxford University Press (OUP) in Bioinformatics
- Vol. 40 (Supplement) , i91-i99
CODEX: COunterfactual Deep learning for the in silico EXploration of cancer cell line perturbations
High-throughput screens (HTS) provide a powerful tool to decipher the causal effects of chemical and genetic perturbations on cancer cell lines. Their ability to evaluate a wide spectrum of interventions, from single drugs to intricate drug combinations and CRISPR-interference, has established them as an invaluable resource for the development of novel therapeutic approaches. Nevertheless, the combinatorial complexity of potential interventions makes a comprehensive exploration intractable. Hence, prioritizing interventions for further experimental investigation becomes of utmost importance. We propose CODEX (COunterfactual Deep learning for the in silico EXploration of cancer cell line perturbations) as a general framework for the causal modeling of HTS data, linking perturbations to their downstream consequences. CODEX relies on a stringent causal modeling strategy based on counterfactual reasoning. As such, CODEX predicts drug-specific cellular responses, comprising cell survival and molecular alterations, and facilitates the in silico exploration of drug combinations. This is achieved for both bulk and single-cell HTS. We further show that CODEX provides a rationale to explore complex genetic modifications from CRISPR-interference in silico in single cells. Our implementation of CODEX is publicly available at https://github.com/sschrod/CODEX. All data used in this article are publicly available.
- 28 June 2024
- Published by Oxford University Press (OUP) in Bioinformatics
- Vol. 40 (Supplement) , i91-i99
Open AccessCODEX: COunterfactual Deep learning for the in silico EXploration of cancer cell line perturbations
High-throughput screens (HTS) provide a powerful tool to decipher the causal effects of chemical and genetic perturbations on cancer cell lines. Their ability to evaluate a wide spectrum of interventions, from single drugs to intricate drug combinations and CRISPR-interference, has established them as an invaluable resource for the development of novel therapeutic approaches. Nevertheless, the combinatorial complexity of potential interventions makes a comprehensive exploration intractable. Hence, prioritizing interventions for further experimental investigation becomes of utmost importance. We propose CODEX (COunterfactual Deep learning for the in silico EXploration of cancer cell line perturbations) as a general framework for the causal modeling of HTS data, linking perturbations to their downstream consequences. CODEX relies on a stringent causal modeling strategy based on counterfactual reasoning. As such, CODEX predicts drug-specific cellular responses, comprising cell survival and molecular alterations, and facilitates the in silico exploration of drug combinations. This is achieved for both bulk and single-cell HTS. We further show that CODEX provides a rationale to explore complex genetic modifications from CRISPR-interference in silico in single cells. Our implementation of CODEX is publicly available at https://github.com/sschrod/CODEX. All data used in this article are publicly available.
- 17 May 2024
- Published by Springer Nature
- p. 344-347
SpaCeNet: Spatial Cellular Networks from Omics Data
- 17 May 2024
- Published by Springer Nature
- p. 344-347
SpaCeNet: Spatial Cellular Networks from Omics Data
- Open Access
CODEX: COunterfactual Deep learning for the in-silico EXploration of cancer cell line perturbations
Motivation: High-throughput screens (HTS) provide a powerful tool to decipher the causal effects of chemical and genetic perturbations on cancer cell lines. Their ability to evaluate a wide spectrum of interventions, from single drugs to intricate drug combinations and CRISPR-interference, has established them as an invaluable resource for the development of novel therapeutic approaches. Nevertheless, the combinatorial complexity of potential interventions makes a comprehensive exploration intractable. Hence, prioritizing interventions for further experimental investigation becomes of utmost importance. Results: We propose CODEX as a general framework for the causal modeling of HTS data, linking perturbations to their downstream consequences. CODEX relies on a stringent causal modeling strategy based on counterfactual reasoning. As such, CODEX predicts drug-specific cellular responses, comprising cell survival and molecular alterations, and facilitates the in-silico exploration of drug combinations. This is achieved for both bulk and single-cell HTS. We further show that CODEX provides a rationale to explore complex genetic modifications from CRISPR-interference in silico in single cells. Availability and Implementation: Our implementation of CODEX is publicly available at https://github.com/sschrod/CODEX . All data used in this article are publicly available.
Open AccessCODEX: COunterfactual Deep learning for the in-silico EXploration of cancer cell line perturbations
Motivation: High-throughput screens (HTS) provide a powerful tool to decipher the causal effects of chemical and genetic perturbations on cancer cell lines. Their ability to evaluate a wide spectrum of interventions, from single drugs to intricate drug combinations and CRISPR-interference, has established them as an invaluable resource for the development of novel therapeutic approaches. Nevertheless, the combinatorial complexity of potential interventions makes a comprehensive exploration intractable. Hence, prioritizing interventions for further experimental investigation becomes of utmost importance. Results: We propose CODEX as a general framework for the causal modeling of HTS data, linking perturbations to their downstream consequences. CODEX relies on a stringent causal modeling strategy based on counterfactual reasoning. As such, CODEX predicts drug-specific cellular responses, comprising cell survival and molecular alterations, and facilitates the in-silico exploration of drug combinations. This is achieved for both bulk and single-cell HTS. We further show that CODEX provides a rationale to explore complex genetic modifications from CRISPR-interference in silico in single cells. Availability and Implementation: Our implementation of CODEX is publicly available at https://github.com/sschrod/CODEX . All data used in this article are publicly available.