Biology subjects

Kreutz, C.

Publications and source records attributed to Kreutz, C..

At least 19 recordsLinked to original sources

bioRxiv2026

1100 Synthetic Benchmark Problems for Dynamic Modeling of Cellular Processes

MotivationSystems biology strives to unravel the complex dynamics of cellular processes, often with the help of ordinary differential equations (ODEs). However, the sparsity of measured data and the strong non-linearity of common ODEs introduce severe numerical problems in typical modeling tasks. This gave rise to the development of many computational algorithms that must be systematically evaluated to ensure optimal method choices. Currently, the amount of well curated models for such benchmarking efforts is insufficient, as building and calibrating biologically reasonable models based on experiments requires years of work. ResultsWe present a large-scale collection of 1100 synthetic modeling problems, generated based on the ODE systems and experimental designs of 22 published modeling problems. This is achieved by extending a recent method for simulation of time-course data for randomly generated observation functions to also include realistic measurement patterns across multiple experimental conditions. By analyzing data and model characteristics, optimization performance and parameter identifiability, we show that the synthetic problems provide both a realistic and diverse extension of the existing problem space. Hence, the synthetic collection provides a valuable resource for benchmarking in dynamic modeling. Availability and ImplementationBenchmark problems and algorithm are publicly available at https://github.com/niklasneubrand/1100SyntheticBenchmarksODE and https://zenodo.org/records/14008247.

systems biology↗

bioRxiv2025

mmContext: an open framework for multimodal contrastive learning of omics and text data

SummaryMultimodal approaches are increasingly leveraged for integrating omics data with textual biological knowledge. Yet there is still no accessible, standardized framework that enables systematic comparison of omics representations with different text encoders within a unified workflow. We present mmContext, a lightweight and extensible multimodal embedding framework built on top of the open-source Sentence Transformers library. The software allows researchers to train or apply models that jointly embed omics and text data using any numeric representation stored in an AnnData .obsm layer and any text encoder available in Hugging Face. mmContext supports integration of diverse biological text sources and provides pipelines for training, evaluation, and data preparation. We train and evaluate models for a RNA-Seq and text integration task, and demonstrate their utility through zero-shot classification of cell types and diseases across four independent datasets. By releasing all models, datasets, and tutorials openly, mmContext enables reproducible and accessible multimodal learning for omics-text integration. Availability and implementationPretrained checkpoints and full source code for our custom MMContextEncoder are available on Hugging Face huggingface.co/jo-mengr. The Python package github.com/mengerj/mmcontext provides the model implementation and training and evaluation scripts for custom training.

bioinformatics↗

bioRxiv2025

NEK7 accelerates NLRP3 inflammasome activation

The NLRP3 inflammasome is a major driver of immunopathology, making it a sought-after drug target. In spite of two decades of intense research, its precise activation mechanism remains elusive, impeding inhibitor design. NEK7 was reported as essential for NLRP3 activation, and several newly identified inhibitors were suggested to act by interfering with their interaction. Here we report that NEK7 accelerates, but is in principle dispensable for NLRP3 activation. The onset of inflammasome activation was unaltered in the absence of NEK7, yet the rate of cells to undergo inflammasome formation and subsequent pyroptosis was approximately 4-fold reduced. Therefore, therapeutic targeting of the NEK7-NLRP3 interaction might have an incomplete effect, which should be considered for drug development. We confirmed entrectinib as a NEK7-dependent inhibitor, while other published compounds turned out not to rely on it. Our results support two possible scenarios for the role of NEK7 in NLRP3 activation: either, NEK7 accelerates one unique pathway of NLRP3 activation, or it is essential for a fast pathway, while being dispensable for a second, slower mode of NLRP3 activation.

immunology↗

bioRxiv2025

PEtab Select: specification standard and supporting software for automated model selection

A central question in mathematical modeling of biological systems is determining which processes are most relevant and how they can be described. There are often competing hypotheses, which yield different models. Model comparison requires parameter optimization and sampling methods. Yet, standards for the specification of model selection problems and the swift evaluation of a broad spectrum of approaches are not available. PEtab Select addresses this challenge by providing a concise, standardized specification of model selection and its associated calibration problems through a new file format standard and software package. The standard facilitates the compact representation of even very large model selection problems; in one example, billions of model alternatives. PEtab Select builds on the PEtab standard for the specification of parameter estimation problems, and enables the use of state-of-the-art modelling and calibration workflows utilizing COPASI, Data2Dynamics, PEtab.jl, and pyPESTO. PEtab Select supports common model selection criteria (e.g., Akaike and Bayesian information criteria) and can be easily extended to use others. To ensure flexibility, PEtab Select implements several model space exploration approaches, including basic brute-force, forward, and backward selection, and also advanced, flexible selection methods. PEtab Select introduces the first standardization of model selection tasks, filling a critical gap in existing computational pipelines. It constitutes an essential contribution to FAIR research software in systems biology by promoting interoperability and reusability in model selection. Author summaryModel selection is a crucial step in mathematical modeling, guiding the choice of components to include in a model. PEtab Select automates this process across diverse modeling frameworks and programming languages via (1) a new interoperable, language-agnostic standard for specifying large-scale model selection problems, and (2) a comprehensive software package that implements these selection methods. Developed through a community effort, PEtab Select has been integrated into multiple modeling frameworks and is accessible to users of COPASI, Julia, MATLAB, and Python.

systems biology↗

bioRxiv2025

Tissue imprinting defines functional mosaic of dermal macrophages

Dermal macrophages (macs) protect the skin from invading pathogens. They are derived from embryonic as well as hematopoietic progenitors. However, the functional impact of their diverse origin and the control networks defining different subsets remain unclear. Here, using multidimensional analysis of dermal macs, we reveal that the absence of circulating monocytes in interferon regulatory factor 8 (Irf8) deficient mice delays mac renewal during the steady state. Yet, the functional mosaic of dermal macs remains largely intact, i.e., major dermal mac subsets develop independently of monocyte replenishment. Thus, the tissue microenvironment is sufficient to induce alternative differentiation pathways and functional specialization of resident cells. Mycobacterial skin infection induces a steep increase in mac density due to monocyte-derived macs which execute urgent antibacterial functions and differentiate into site-adapted mac subsets in wildtype but not Irf8-/-mice, while long-term resident macs are required to initiate a tissue repair program already in early stages of infection. In summary, we introduce a model, where an intricate network of specialized mac subsets develops to meet microanatomical needs and external cellular input is required only during immunological emergency situations. HighlightsO_LIIrf8-/--driven monocytopenia has negligible impact on homeostatic dermal macrophage diversity. C_LIO_LIResident dermal macrophages have diverse specializations but remain flexible to adapt to challenges such as lacking monocyte influx C_LIO_LIBone marrow-derived macrophages differentiate into specialized resident cells, with microenvironmental cues overriding origin-dependent programming. C_LIO_LIIn chronic bacterial infections, distinct specialized bone-marrow-derived macrophages mount the defense, while resident macrophages activate a tissue-modifying program from early on. C_LI

immunology↗

bioRxiv2024

Foxi1 regulates multiple steps of mucociliary development and ionocyte specification through transcriptional and epigenetic mechanisms

Foxi1 is a master regulator of ionocytes (ISCs / INCs) across species and organs. Two subtypes of ISCs exist, and both -and {beta}-ISCs regulate pH-and ion-homeostasis in epithelia. Gain and loss of FOXI1 function are associated with human diseases, including Pendred syndrome, male infertility, renal acidosis and cancers. Foxi1 was predominantly studied in the context of ISC specification, however, reports indicate additional functions in early and ectodermal development. Here, we re-investigated the functions of Foxi1 in Xenopus laevis embryonic mucociliary epidermis development and found a novel function for Foxi1 in the generation of Notch-ligand expressing mucociliary multipotent progenitors (MPPs). We demonstrate that Foxi1 has multiple concentration-dependent functions: At low levels, Foxi1 maintains ectodermal competence in MPPs through transcriptional and epigenetic mechanisms, while at high levels, Foxi1 induces a multi-step process of ISC specification and differentiation in cooperation with Ubp1 and Dmrt2. We further describe how foxi1 expression is affected through auto-and Notch-regulation, and how this developmental program affects mucociliary patterning. Together, we reveal novel functions for Foxi1 in Xenopus mucociliary epidermis formation, relevant to our understanding of vertebrate development and human disease.

developmental biology↗

bioRxiv2024

RTF: An R package for modelling time course data

SummaryThe retarded transient function (RTF) approach has been introduced as a complementary approach to employing ordinary differential equations (ODEs) for modelling dynamics typically observed for cellular signalling processes. Here, we introduce the R package to the RTF approach, which has originally been implemented within the MATLAB-based Data2Dynamics modelling framework. The package enables modelling of time and dose dependencies and includes model reduction to minimize overfitting. It can be applied not only to experimental data but also to trajectories of ODE models, to characterize the dynamics or to identify major targets of experimental perturbations in the low-dimensional representations calculated by the package. Availability and ImplementationThe R package RTF is available at https://github.com/kreutz-lab/RTF. Contactclemens.kreutz@uniklinik-freiburg.de

bioinformatics↗

bioRxiv2024

Dynamic modelling of signalling pathways when ODEsare not feasible

MotivationMathematical modelling plays a crucial role in understanding inter- and intracellular signalling processes. Currently, ordinary differential equations (ODEs) are the predominant approach in systems biology for modelling such pathways. While ODE models offer mechanistic interpretability, they also suffer from limitations, including the need to consider all relevant compounds, resulting in large models difficult to handle numerically and requiring extensive data. ResultsIn previous work, we introduced the retarded transient function (RTF) as an alternative method for modelling temporal responses of signalling pathways. Here, we extend the RTF approach to integrate concentration or dose-dependencies into the modelling of dynamics. With this advancement, RTF modelling now fully encompasses the application range of ordinary differential equation (ODE) models, which comprises predictions in both time and concentration domains. Moreover, characterizing dose-dependencies provides an intuitive way to investigate and characterize signalling differences between biological conditions or cell-types based on their response to stimulating inputs. To demonstrate the applicability of our extended approach, we employ data from time- and dose-dependent inflammasome activation in bone-marrow derived macrophages (BMDMs) treated with nigericin sodium salt. Our results show the effectiveness of the extended RTF approach as a generic framework for modelling dose-dependent kinetics in cellular signalling. The approach results in intuitively interpretable parameters that describe signal dynamics and enables predictive modelling of time- and dose-dependencies even if only individual cellular components are quantified. AvailabilityThe presented approach is available within the MATLAB-based Data2Dynamics modelling toolbox at https://github.com/Data2Dynamics and https://zenodo.org/records/14008247 and as R code at https://github.com/kreutz-lab/RTF. Contactclemens.kreutz@uniklinik-freiburg.de

systems biology↗

bioRxiv2023

Likelihood-ratio test statistic for the finite-sample case in nonlinear ordinary differential equation models

Likelihood ratios are frequently utilized as basis for statistical tests, for model selection criteria and for assessing parameter and prediction uncertainties, e.g. using the profile likelihood. However, translating these likelihood ratios into p-values or confidence intervals requires the exact form of the test statistics distribution. The lack of knowledge about this distribution for nonlinear ordinary differential equation (ODE) models requires an approximation which assumes the so-called asymptotic setting, i.e. a sufficiently large amount of data. Since the amount of data from quantitative molecular biology is typically limited in applications, this finite-sample case regularly occurs for mechanistic models of dynamical systems, e.g. biochemical reaction networks or infectious disease models. Thus, it is unclear whether the standard approach of using statistical thresholds derived for the asymptotic large-sample setting in realistic applications results in valid conclusions. In this study, empirical likelihood ratios for parameters from 19 published nonlinear ODE benchmark models are investigated using a resampling approach for the original data designs. Their distributions are compared to the asymptotic approximation and statistical thresholds are checked for conservativeness. It turns out, that corrections of the likelihood ratios in such finite-sample applications are required in order to avoid anti-conservative results. Author summaryStatistical methods based on the likelihood ratio are ubiquitous in mathematical modelling in systems biology. For example confidence intervals of estimated parameters rely on the statistical properties of the likelihood-ratio test. However, it is often overlooked that these intervals sizes rely on assumptions on the amounts of data, which are regularly violated in typical applications in systems biology. By checking the appropriateness of these assumptions in models from the literature, this study shows that in a surprisingly large fraction confidence intervals might be too small. Using a geometric interpretation of parameter estimation in the so-called data space, it is motivated why these issues appear and how they depend on the identifiability of the model parameters. In order to avoid such problematic situations, this work makes suggestions on how to adapt the statistical threshold values for likelihood-ratio test. By this, it can be assured that valid statistical conclusions are drawn from the analysis, also in situations where only smaller data sets are available. Such corrections yield for example more conservative confidence interval sizes and thus decrease a potential underestimation of the parameter uncertainty.

systems biology↗

bioRxiv2023

Temporal Notch signaling regulates mucociliary cell fates through Hes-mediated competitive de-repression

Mucociliary epithelia are found across different organs in animals, where they release bioactive substances and generate extracellular fluid flows. One key function of mucociliary epithelia is the clearance of pathogens, e.g. in the vertebrate lung and the epidermis of amphibian tadpoles. Mucociliary clearance relies on the correct balance between secretory cells that release mucus, ciliated cells that generate fluid flow as well as specialized cell types, including pH-regulating ionocytes and basal stem cells. Notch signaling, Hes repressors and cell type-inducing transcription factors (e.g. Foxi1, Mcidas, Spdef and Tp63) regulate cell fates and cell type proportions across mucociliary systems. Lateral inhibition was proposed to control mucociliary cell fates, but current models cannot explain how more than two cell types are generated and how Hes genes are employed as mediators during patterning. Using the Xenopus tadpole epidermis, we addressed these open questions in mucociliary biology through a combination of in vivo and organoid experiments, time-resolved transcriptomic studies and mathematical modeling. This revealed that ionocytes, ciliated cells, secretory cells and basal cells are preferentially specified at different time points and Notch signaling levels via sequentially expressed Hes factors. We termed this mode of patterning "competitive de-repression", because cell fates are selected by suppression of alternative fate choices, and demonstrate that this relies on differential active repression of cell fate transcription factors. Mathematical modeling further indicated the need for a positively Notch-regulated patterning factor, and we provide evidence that Spdef mediates Notch input for secretory and basal cell specification. Collectively, this work presents a coherent model for Notch- and Hes-mediated mucociliary cell fate specification in a vertebrate tissue, which allows for the specification of more than two cell fates within the Notch lateral-inhibition paradigm.

developmental biology↗

bioRxiv2023

Realistic simulation of time-course measurements in systems biology

MotivationIn systems biology, the analysis of complex nonlinear systems faces many methodological challenges. However, the performance evaluation of competing methods is limited by the small amount of publicly available data from biological experiments. Therefore, simulation studies with a realistic representation of the data are a promising alternative and bring the advantage of knowing the ground truth. ResultsWe present an approach for designing a realistic simulation study. Based on 19 published systems biology models with experimental data, we assess typical measurement characteristics such as observables, observation type, measurement error, and observation times. For the latter, we estimate typical time features by fitting a transient response function. We demonstrate the approach on a meal model of the glucose insulin system, a mitogen-activated protein-kinase cascade and a model for the epidermal growth factor signaling. The performance of the realistic design is validated on 9 systems biology models in terms of optimization, integration and identifiability. For any dynamic model downloaded from an online database, our algorithm analyzes the model dynamics and specifies a realistic experimental design. The approach is specifically suited for systematic benchmarking of methods for timecourse data in the context of systems biology. In particular, various application settings such as number of parameters, initial conditions, error model etc. can be tested. AvailabilityThe approach is implemented in the MATLAB-based modelling toolbox Data2Dynamics and available at https://github.com/Data2Dynamics/d2d.

systems biology↗

bioRxiv2022

Conformational dynamics and target-dependent myristoyl switch of calcineurin B homologous protein 3

Calcineurin B homologous protein 3 (CHP3) is an EF-hand Ca2+-binding protein involved in regulation of cancerogenesis, cardiac hypertrophy and neuronal development via interactions with sodium/proton exchangers (NHEs) and signalling proteins. CHP3 binds Ca2+ with micromolar affinity providing the basis to respond to intracellular Ca2+ signals. Ca2+ binding and myristoylation are important for CHP3 function but the underlying molecular mechanism remained elusive. Here, we show that Ca2+ binding and myristoylation independently affect conformational dynamics and functions of human CHP3. Ca2+ binding increased flexibility and hydrophobicity of CHP3 indicative of an open conformation. CHP3 in open Ca2+-bound conformation had higher affinity for NHE1 and associated stronger with lipid membranes compared to the closed Mg2+-bound conformation. Myristoylation enhanced flexibility of CHP3 and decreased its affinity to NHE1 independently of the bound ion, but did not affect its binding to lipid membranes. The data exclude the proposed Ca2+-myristoyl switch for CHP3. Instead, they document a Ca2+-independent exposure of the myristoyl moiety induced by binding of the target peptide to CHP3 enhancing its association to lipid membranes. We name this novel regulatory mechanism "target-dependent myristoyl switch". Taken together, the interplay of Ca2+ binding, myristoylation and target binding allows for a context-specific regulation of CHP3 functions.

biochemistry↗

bioRxiv2022

The performance of deep generative models for learning joint embeddings of single-cell multi-omics data

Recent extensions of single-cell studies to multiple data modalities raise new questions regarding experimental design. For example, the challenge of sparsity in single-omics data might be partly resolved by compensating for missing information across modalities. In particular, deep learning approaches, such as deep generative models (DGMs), can potentially uncover complex patterns via a joint embedding. Yet, this also raises the question of sample size requirements for identifying such patterns from single-cell multi-omics data. Here, we empirically examine the quality of DGM-based integrations for varying sample sizes. We first review the existing literature and give a short overview of deep learning methods for multi-omics integration. Next, we consider eight popular tools in more detail and examine their robustness to different cell numbers, covering two of the most common multi-omics types currently favored. Specifically, we use data featuring simultaneous gene expression measurements at the RNA level and protein abundance measurements for cell surface proteins (CITE-seq), as well as data where chromatin accessibility and RNA expression are measured in thousands of cells (10x Multiome). We examine the ability of the methods to learn joint embeddings based on biological and technical metrics. Finally, we provide recommendations for the design of multi-omics experiments and discuss potential future developments.

bioinformatics↗

bioRxiv2021

Enhanced FcγRIII/CD16 activation by discrete ligands as independent correlates of disease severity in COVID-19 patients

A dysregulated immune response with high levels of SARS-CoV-2 specific IgG antibodies characterizes patients with severe or critical COVID-19. Although a robust IgG response is traditionally considered to be protective, excessive triggering of activating Fc-gamma-receptors (Fc{gamma}Rs) could be detrimental and cause immunopathology. Here, we document that patients who develop soluble circulating IgG immune complexes (sICs) during infection are subject to enhanced immunopathology driven by Fc{gamma}R activation. Utilizing cell-based reporter systems we provide evidence that sICs are predominantly formed prior to a specific humoral response against SARS-CoV-2. sIC formation, together with increased afucosylation of SARS-CoV-2 specific IgG eventually leads to an enhanced CD16 (Fc{gamma}RIII) activation of immune cells reaching activation levels comparable active systemic lupus erythematosus (SLE) disease. Our data suggest a vicious cycle of escalating immunopathology driven by an early formation of sICs in predisposed patients. These findings reconcile the seemingly paradoxical findings of high antiviral IgG responses and systemic immune dysregulation in severe COVID-19. Clinical implicationsThe identification of sICs as drivers of an escalating immunopathology in predisposed patients opens new avenues regarding intervention strategies to alleviate critical COVID-19 progression. Graphical abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=78 SRC="FIGDIR/small/449893v4_ufig1.gif" ALT="Figure 1"> View larger version (17K): org.highwire.dtl.DTLVardef@b4616corg.highwire.dtl.DTLVardef@682d1aorg.highwire.dtl.DTLVardef@16946cborg.highwire.dtl.DTLVardef@a6ef7d_HPS_FORMAT_FIGEXP M_FIG C_FIG A vicious cycle of immunopathology in COVID-19 patients is driven by soluble multimeric immune complexes (sICs). SARS-CoV-2 infection triggers sIC formation in prone individuals. Activation of Fc{gamma}RIII/CD16 expressing immune cells by sICs precedes a humoral response to SARS-CoV2 infection. sICs and infection add to IgG afucosylation, further enhancing Fc{gamma}RIII/CD16 activation by opsonized targets. High inflammation induces further sIC mediated immune cell activation ultimately leading to an escalating immunopathology.

immunology↗

bioRxiv2020

A quantitative model predicts how m6A reshapes the kinetic landscape of nucleic acid hybridization and conformational transitions

N6-methyladenosine (m6A) is a post-transcriptional modification that controls gene expression by recruiting proteins to RNA sites. The modification also slows biochemical processes through mechanisms that are not understood. Using NMR relaxation dispersion, we show that m6A pairs with uridine with the methylamino group in the anti conformation to form a Watson-Crick base pair that transiently exchanges on the millisecond timescale with a singly hydrogen-bonded low-populated (1%) mismatch-like conformation in which the methylamino group is syn. This ability to rapidly interchange between Watson-Crick or mismatch-like forms, combined with different syn:anti isomer preferences when paired (~1:100) versus unpaired (~10:1), explains how m6A robustly slows duplex annealing without affecting melting via two pathways in which isomerization occurs before or after duplex annealing. Our model quantitatively predicts how m6A reshapes the kinetic landscape of nucleic acid hybridization and conformational transitions, and provides an explanation for why the modification robustly slows diverse cellular processes.

biochemistry↗

bioRxiv2020

DIMA: Data-driven selection of a suitable imputation algorithm

MotivationImputation is a prominent strategy when dealing with missing values (MVs) in proteomics data analysis pipelines. However, the performance of different imputation methods is difficult to assess and varies strongly depending on data characteristics. To overcome this issue, we present the concept of a data-driven selection of a suitable imputation algorithm (DIMA). ResultsThe performance and broad applicability of DIMA is demonstrated on 121 quantitative proteomics data sets from the PRIDE database and on simulated data consisting of 5 - 50% MVs with different proportions of missing not at random and missing completely at random values. DIMA reliably suggests a high-performing imputation algorithm which is always among the three best algorithms and results in a root mean square error difference ({Delta}RMSE) [≤] 10% in 84% of the cases. Availability and ImplementationSource code is freely available for download at github.com/clemenskreutz/OmicsData.

bioinformatics↗

bioRxiv2020

2'-O-methylation alters the RNA secondary structural ensemble

2-O-methyl (Nm) is a highly abundant post-transcriptional RNA modification that plays important biological roles through mechanisms that are not entirely understood. There is evidence that Nm can alter the biological activities of RNAs by biasing the ribose sugar pucker equilibrium toward the C3-endo conformation formed in canonical duplexes. However, little is known about how Nm might more broadly alter the dynamic ensembles of non-canonical RNA motifs. Here, using NMR and the HIV-1 transactivation response (TAR) element as a model system, we show that Nm preferentially stabilizes alternative secondary structures in which the Nm-modified nucleotides are paired, increasing both the abundance and lifetime of a low-populated short-lived excited state by up to 10-fold. The extent of stabilization increased with number of Nm modifications and was also dependent on Mg2+. Through phi ({Phi}) value analysis, the Nm modification also provided rare insights into the structure of the transition state for conformational exchange. Our results suggest that Nm could alter the biological activities of Nm-modified RNAs by modulating their secondary structural ensembles as well as establish the utility of Nm as a tool for the discovery and characterization of RNA excited state conformations.

biochemistry↗

bioRxiv2020

Tail-Robust Quantile Normalization

High-throughput biological data - such as mass spectrometry-based proteomics data - suffer from systematic non-biological variance, which is introduced by systematic errors such as batch effects. This hinders the estimation of real biological signals and, thus, decreases the power of statistical tests and biases the identification of differentially expressed sample classes. To remove such unintended variation, while retaining the biological signal of interest, the analysis workflows for mass spectrometry-based quantification typically comprises normalization steps prior to the statistical analysis of the data. Several normalization methods, such as quantile normalization, have originally been developed for microarray data. However, unlike microarray data, proteomics data may contain features, in the form of protein intensities, that are consistently highly abundant across experimental conditions and, hence, are encountered in the tails of the protein intensity distribution. If such proteins are present, statistical inferences of the intensity profiles of the normalized features are impeded through the increased number of false positive findings due to the biased estimation of the variance of the data. Thus, we developed a, freely available, novel approach: tail-robust quantile normalization. It extends the traditional quantile normalization to preserve the biological signals of features in the tails of the distribution over experimental conditions and to account for sample-dependent missing values.

bioinformatics↗