Biology subjects

Lauffenburger, D. A.

Publications and source records attributed to Lauffenburger, D. A..

At least 19 recordsLinked to original sources

bioRxiv2026

Interactome-based deep learning and path-level decomposition reveal TGFβ-IFNα competition in MASH

Dysregulated signaling drives Metabolic Dysfunction-Associated Steatohepatitis (MASH) through a complex crosstalk between pro- and anti-fibrotic cues. We previously observed potential competition between IFN-mediated JAK-STAT signaling and TGF{beta} in modulating MASH fibrosis, but the precise mechanism of that competition, and whether it can act directly on hepatocytes, is yet unknown. We used DT-LEMBAS, an interactome-based deep learning framework, training an ensemble of 50 models on drug-induced transcriptomics from primary human hepatocytes. The ensemble captured transcriptional changes induced by simulated receptor knockouts and aligned predicted profiles with known TGF{beta}- and IFN-induced phenotypes. Systematic simulations demonstrate their competitive modulation of MASH phenotype scores benchmarked against human cohorts. To extract mechanistic insights, we developed a gradient-based approach, where we computed the Jacobian of the used recurrent signaling network and applied chain rule decomposition to identify a four-node module of SMAD3, TP63, STAT2, and STAT3. Mechanistic hypotheses involving these canonical and non-canonical TF mediators were experimentally validated in primary hepatocyte spheroids. This work establishes a generalizable and scalable approach to decode large signaling neural networks into interpretable biological subnetworks.

systems biology↗

bioRxiv2026

An IgG-Optimized Enzyme-Linked Lectin Assay (ELLA) for Quantitative Analysis of Immunoglobulin Glycosylation

Antibody Fc glycosylation is modulated in a variety of disease and immune response contexts, altering downstream functional responses including antibody-dependent cellular cytotoxicity through modified immune cell Fc receptor binding. Accessible, high-throughput glycosylation assays such as enzyme-linked lectin assays (ELLAs) are essential to advance understanding of glycosylation regulation and function. However, current ELLA protocols lack standardization and optimization, and results are reported out in arbitrary absorbance units, limiting reproducibility and cross-study comparability. We developed an optimized multi-lectin parallel ELLA with three specific improvements: systematic optimization of incubation times and reagent concentrations; incorporation of Protein A for IgG specificity; and use of commercially available bovine fetuin B as a quantitative surrogate standard for cross-study reproducibility. Our panel of 8 lectins, SNA, RCA, LCA, PHA-E, PHA-L, MAL-I, WGA, and DSL, cover the major IgG glycoforms. We demonstrate that our ELLA panel can reveal biologically relevant cytokine-induced plasticity of IgG glycosylation profiles in immortalized B cells.

bioengineering↗

bioRxiv2026

FlowTransOP: Distributional Translation of Omics Signatures via Constrained Deep Flow Matching

Observations from pre-clinical models rarely generalize to human patients, leading to many failures in clinical trials. Most existing methods cannot handle domains with non-overlapping features and no paired samples. Here, we developed FlowTransOP to translate biological observations across such domains without requiring 1-to-1 feature mappings and paired data, while providing a guideline for model selection across four translational regimes. We use flow matching to align full domain distributions in a pre-aligned latent space, with a structural regularization term that keeps similar conditions proximate after transformation. FlowTransOP remains competitive with gold-standard approaches requiring paired samples, but outperforms them when pairs become scarce (<35 pairs) or when cross-domain features are only moderately correlated (r<=0.58). Overall, FlowTransOP can translate perturbations between pre-clinical models and patients when direct correspondences are unavailable, enabling reliable therapeutic inference. As a proof-of-concept, we trained a foundational mouse-human transcriptomic map on ARCHS4 and applied it to liver disease predictions.

bioinformatics↗

bioRxiv2025

Biologically informed neural network models are robust to spurious interactions via self-pruning

Computational models of cellular networks hold promise to uncover disease mechanisms and guide therapeutic strategies. Biology-informed neural networks (BINNs) is an emerging approach to create such models by combining the predictive power of deep learning with prior knowledge, a vital aspect of biological research. The architectures of BINNs enforces a network structure from which mechanism can ideally be inferred. However, a key challenge is to evaluate the reliability of these models, as cells are inherently complex, involving intricate and sometimes unknown interactions. Currently, analysis mainly focuses on selected pathways rather than a more comprehensive perspective. In this work we demonstrate an alternative holistic approach: we measure to which extent purposefully introduced spurious interactions are down-weighted by a BINN during training (self-pruning). The metric suggested RRA (Relative Residual Area) allows for direct distribution comparison with perfect self-pruning achieved at zero and a failure to self-prune if above one. To enable rapid testing, we updated LEMBAS (Large-scale knowledge-EMBedded Artificial Signaling-networks), our recurrent neural network framework for intracellular signaling dynamics, with full GPU acceleration. Our implementation achieves a >7-fold speedup compared to the original while preserving predictive accuracy. We evaluated self-pruning in 3 different datasets and found that when spurious interactions are introduced at random, the model prunes these to a larger extent than those from the prior knowledge network (PKN), provided the model is regularized with a sufficiently large L2 norm. This suggests that BINNs can be robust to uncertainty in the PKN. Implementation and applicationOur implementation of LEMBAS is freely available under a MIT license at https://github.com/AvlantNilssonLab/LEMBAS_GPU. The models and results to generate the figures can be downloaded through https://zenodo.org/records/17425598.

systems biology↗

bioRxiv2025

Multivariate analysis of glycogenes reveals coordinated regulation of immunoglobulin glycosylation in an immortalized human B cell system

While neutralizing ability has traditionally been considered the most important antibody function, appreciation has grown for Fc-mediated extra-neutralizing functions, which are shaped by IgG glycosylation. However, there remain fundamental questions as to how B lymphocytes induce and regulate antibody glycosylation and thus functional capability. Understanding how transcriptional and cell state regulation shape glycosylation could reveal levers to tune protective humoral profiles in a disease- and antigen-specific manner. Prior studies have explored a limited panel of glycogenes and measured bulk glycosylation changes. Here, employing an in vitro antigen-specific B cell culture system, we systematically characterize transcriptional and humoral responses to cytokine perturbations. After exposure to a broad panel of cytokines (IL-4, IL-6, IL-10, IL-17, TNFa, IFNg, APRIL, and BAFF) across multiple concentrations and timepoints, transcriptomic profiling and lectin-based IgG glycome assays are employed to associate cytokine stimuli with both glycogene expression and IgG glycosylation. Supervised and unsupervised machine learning models identify cytokine-specific glycogene "signatures" as well as distinct immunoglobulin glycosylation profiles. We find that cytokines induce rapid transcriptional responses, with glycogene signatures outperforming single-gene changes in distinguishing stimulation conditions. We further demonstrate the ability to induce both pro- and anti-inflammatory IgG glycosylation profiles, particularly in terms of IgG galactosylation. This work demonstrates the utility of this system to parse the cytokine-driven regulation of B lymphocyte glycogenes, establishing a framework for dissecting how environmental cues shape antibody glycosylation, with relevance for autoimmune disease, infection, and vaccine responses.

systems biology↗

bioRxiv2025

Cross-species blood transcriptional correlates of BCG-mediated protection against tuberculosis include innate and adaptive immune processes

The immune mechanisms induced by the Bacillus Calmette-Guerin (BCG) vaccine, and the subset of which mediate protection against tuberculosis (TB), remain poorly understood. This is further complicated by difficulties to verify vaccine-induced protection in humans. Although research in animal models, namely mice and non-human primates (NHPs), has begun to close this knowledge gap, discrepancies in the relative importance of biological pathways across species limit the utility of animal model-derived biological insights in humans. To address these challenges, we applied a systems modeling framework, Translatable Components Regression (TransCompR), to identify human blood transcriptional variability which could predict Mtb challenge outcomes in BCG-vaccinated NHPs. These protection-associated pathways included both innate and adaptive immune activation mechanisms, along with signaling via type I interferons and anti-mycobacterial T helper cytokines. We further partially validated the associations between these mechanisms and protection in humans using publicly available microarray data collected from BCG-vaccinated infants who either developed TB or remained healthy during two years of follow-up. Overall, our work demonstrates how species translation modeling can leverage animal studies to generate hypotheses about the mechanisms that underlie human infectious disease and vaccination outcomes, which may be difficult or impossible to ascertain using human data alone.

systems biology↗

bioRxiv2025

Systems biology framework for rational design of operational conditions for in vitro / in vivo translation of microphysiological systems

Preclinical models are used extensively to study diseases and potential therapeutic treatments. Complex in vitro platforms incorporating human cellular components, known as microphysiological systems (MPS), can model cellular and microenvironmental features of diseased tissues. However, determining experimental conditions -- particularly biomolecular cues such as growth factors, cytokines, and matrix proteins -- providing the most effective translatability of MPS-generated information to in vivo human subject contexts is a major challenge. Here, using metabolic dysfunction-associated fatty liver disease (MAFLD) studied using the CNBio PhysioMimix as a case study, we developed a machine learning framework called Latent In Vitro to In Vivo Translation (LIV2TRANS) to ascertain how MPS data map to in vivo data, first sharpening translation insights and consequently elucidating experimental conditions that can further enhance translation capability. Our findings in this case study highlight TGF{beta} as a crucial cue for MPS translatability and indicate that adding JAK-STAT pathway perturbations via interferon stimuli could increase the predictive performance of this MPS in MAFLD studies. Finally, we developed an optimization approach that identified androgen and EGFR signaling as key for maximizing the capacity of this MPS to capture in vivo human biological information germane to MAFLD. More broadly, this work establishes a mathematically principled approach for identifying experimental conditions most beneficially capturing in vivo human-relevant molecular pathways and processes, generalizable to preclinical studies for a wide range of diseases and potential treatments.

systems biology↗

bioRxiv2025

A Vascularized Liver Microphysiological System Captures Key Features of Hepatic Insulin Resistance and Monocyte Infiltration

Three-dimensional in vitro liver models are a promising means to recapitulate key aspects of human liver disease pathologies, thereby aiding therapeutic development. Spheroidal aggregates of hepatocytes, sometimes including non-parenchymal cells, are an established approach for modeling facets of metabolism and drug responses, yet these models often lack dynamic interactions with vascular and immune cells that also contribute to disease development and progression. To address this, we developed a microphysiological system (MPS) that integrates multicellular human hepatic spheroids with self-organized microvascular networks. We show extensive interaction between primary human spheroids and functional vasculature while maintaining key hepatic functions. We demonstrate the utility of this MPS by modeling an insulin resistance state through chronic exposure to disease-mimetic media conditions. This disease model displays altered hepatocyte metabolism, dysregulated vascular features, and increased inflammation state. Enabled by the functional vasculature, we further extend this disease model to capture changes in immune cell recruitment. When culturing CD14+ monocytes in our liver MPS, a subset of monocytes extravasate, localize to hepatic spheroids, and begin differentiating into CD163+ macrophages. These cells infiltrate with greater frequency in insulin resistant samples, consistent with known clinical findings. All together, this vascularized MPS model captures disease-relevant liver biology including inflammatory features.

bioengineering↗

bioRxiv2024

Systematic deconstruction of myeloid cell signaling in tuberculosis granulomas reveals IFN-γ, TGF-β, and time are associated with conserved myeloid diversity

Myeloid cells are key constituents of tuberculosis (TB) granulomas. They are the major target of pathogen infection and play central roles in pathogen control, antigen presentation, adaptive immune cell recruitment, and tissue homeostasis. However, the role of myeloid cells in TB has been studied largely through ex vivo experimental approaches that do not capture the dynamic phenotypic and functional states of these cells in the disease environment. To address this gap, we used a combination of bulk and single-cell RNA sequencing (scRNA-seq), computational modeling, and imaging to define the molecular diversity of myeloid cells in granulomas from Mycobacterium tuberculosis-infected nonhuman primates. We observed an increase in myeloid cell diversity in granulomas compared to non-granulomatous lung tissue. This increased transcriptional diversity is defined by a continuum of macrophage differentiation-, metabolism-, and cytokine-regulated transcriptional programs. In vitro experimental modeling of monocyte-to-macrophage differentiation in defined cytokine environments implicates differentiation time, IFN-{gamma}, and TGF-{beta} signaling as candidate drivers of macrophage diversity. We next examined the conservation of these populations across additional experimental models of Mtb infection and found myeloid cell subsets enriched across the TB disease spectrum. To further contextualize these responses, we constructed an atlas of myeloid cells across diverse human lung pathologies, finding myeloid cell subpopulations that were similar between TB and other lung pathologies as well as subpopulations that distinguish between diseases. Collectively, this study identifies points of integration between myeloid cell biology in TB granulomas and other lung diseases that can be used for defining the signals that instruct myeloid cell behavior in TB and other diseases, as well as advance myeloid cell-targeted therapies.

immunology↗

bioRxiv2024

Library-based single-cell analysis of CAR signaling reveals drivers of in vivo persistence

The anti-tumor function of engineered T cells expressing chimeric antigen receptors (CARs) is dependent on signals transduced through intracellular signaling domains (ICDs). Different ICDs are known to drive distinct phenotypes, but systematic investigations into how ICD architectures direct T cell function--particularly at the molecular level--are lacking. Here, we use single-cell sequencing to map diverse signaling inputs to transcriptional outputs, focusing on a defined library of clinically relevant ICD architectures. Informed by these observations, we functionally characterize transcriptionally distinct ICD variants across various contexts to build comprehensive maps from ICD composition to phenotypic output. We identify a unique tonic signaling signature associated with a subset of ICD architectures that drives durable in vivo persistence and efficacy in liquid, but not solid, tumors. Our findings work toward decoding CAR signaling design principles, with implications for the rational design of next-generation ICD architectures optimized for in vivo function.

immunology↗

bioRxiv2024

Cross species modeling reveals a role for the unfolded protein response in shaping the transcriptional reaction to Mycobacterium tuberculosis infection

Numerous blood mRNA signatures have been developed to diagnose tuberculosis (TB) disease. The utility of these signatures in diverse populations depends on the inclusion of ubiquitously expressed features, such as type 1 interferon (IFN) production and innate immune cell activities. As a result, these signatures are generally insensitive to heterogeneous responses between individuals. Designing more effective therapies will require understanding the diverse mechanisms underlying pathogenesis by associating them with appropriate preclinical animal models. To address this critical animal-to-human gap, we applied a modeling framework, Translatable Components Regression, which is designed to account for biological heterogeneity by identifying multiple orthogonal axes of variation that are common to humans and animal models. Our framework was capable of distinguishing human active TB from latent TB infection using a model derived from murine data. This discrimination was based on differential expression of numerous biological pathways in addition to the common IFN and neutrophil signatures. Prominent among these predictive pathways was protein translation, which we show is a feature of the Mtb infection-induced Unfolded Protein Response (UPR) in macrophages. We show that this cellular stress pathway controls a variety of immune-related functions in Mtb-infected mouse macrophages, suggesting a possible causative role during the development of TB disease. Significance StatementDespite tuberculosis being one of the top causes of global mortality, the mechanisms that control the progression of disease are still not fully understand. Here we leverage a systems-level modeling approach that incorporates transcriptomics data across thousands of genes from both a traditional tuberculosis mouse model and human clinical samples to implicate a previously unappreciated mechanism in pathogenesis, the unfolded protein response. We validate these findings in a mouse macrophage model and pinpoint which branch of the unfolded protein response might be activated during tuberculosis infection. These insights, originally derived from our cross-species model, may allow us to better understand human tuberculosis pathogenesis and potentially identify therapeutic targets to prevent active tuberculosis.

systems biology↗

bioRxiv2024

Markov Field network integration of multi-modal data predicts effects of immune system perturbations on intravenous BCG vaccination in macaques

Analysis of multi-modal datasets can identify multi-scale interactions underlying biological systems, but can be beset by spurious connections due to indirect impacts propagating through an unmapped biological network. For example, studies in macaques have shown that BCG vaccination by an intravenous route protects against tuberculosis, correlating with changes across various immune data modes. To eliminate spurious correlations and identify critical immune interactions in a public multi-modal dataset (systems serology, cytokines, cytometry) of vaccinated macaques, we applied Markov Fields (MF), a data-driven approach that explains vaccine efficacy and immune correlations via multivariate network paths, without requiring large numbers of samples (i.e. macaques) relative to multivariate features. Furthermore, we find that integrating multiple data modes with MFs helps to remove spurious connections. Finally, we used the MF to predict outcomes of perturbations at various immune nodes, including a B-cell depletion that induced network-wide shifts without reducing vaccine protection, which we validated experimentally.

systems biology↗

bioRxiv2024

Pooled screening for CAR function identifies novel IL13Rα2-targeted CARs for treatment of glioblastoma

Chimeric antigen receptor therapies have demonstrated potent efficacy in treating B cell malignancies, but have yet to meaningfully translate to solid tumors. Here, we utilize our pooled screening platform, CARPOOL, to expedite the discovery of CARs with anti-tumor functions necessary for solid tumor efficacy. We performed selections in primary human T cells expressing a library of 1.3x106 3rd generation CARs targeting IL13R2, a cancer testis antigen commonly expressed in glioblastoma. Selections were performed for cytotoxicity, proliferation, memory formation, and persistence upon repeated antigen challenge. Each enriched CAR robustly produced the phenotype for which it was selected, and one enriched CAR triggered potent cytotoxicity and long-term proliferation upon in vitro tumor rechallenge. It also showed significantly improved persistence and comparable antigen-specific tumor control in a microphysiological human in vitro model and a xenograft model of human glioblastoma. Taken together, this work demonstrates the utility of extending CARPOOL to diseases beyond hematological malignancies and represents the largest exploration of signaling combinations in human primary cells to date.

immunology↗

bioRxiv2023

Inference of drug off-target effects on cellular signaling using Interactome-Based Deep Learning

Many diseases emerge from dysregulated cellular signaling, and drugs are often designed to target specific nodes in cellular networks e.g. signaling proteins, or transcription factors. However, off-target effects are common and may ultimately result in failed clinical trials. Computational modeling of the cells transcriptional response to drugs could improve our understanding of their mechanisms of action. Here we develop such an approach based on ensembles of artificial neural networks, that simultaneously infer drug-target interactions and their downstream effects on intracellular signaling. Applied to gene expression data from different cell lines, it outperforms basic machine learning approaches in predicting transcription factors activity, while recovering most known drug-target interactions and inferring many new, which we validate in an independent dataset. As a case study, we explore the inferred interactions of the drug Lestaurtinib and its effects on downstream signaling. Beyond its intended target FLT3 the model predicts an inhibition of CDK2 that enhances downregulation of the cell cycle-critical transcription factor FOXM1, corroborating literature findings. Our approach can therefore enhance our understanding of drug signaling for therapeutic design.

systems biology↗

bioRxiv2023

Analysis of the Diverse Antigenic Landscape of the Malaria Invasion Protein RH5 Identifies a Potent Vaccine-Induced Human Public Antibody Clonotype

The highly conserved and essential Plasmodium falciparum reticulocyte-binding protein homolog 5 (PfRH5) has emerged as the leading target for vaccines that seek to protect against the disease-causing blood-stage of malaria. However, the features of the human vaccine-induced antibody response that confer highly potent inhibition of malaria parasite invasion into red blood cells are not well defined. Here we characterize over 200 human IgG monoclonal antibodies induced by the most advanced PfRH5 vaccine. We define the antigenic landscape of this molecule, and establish epitope specificity, antibody association rate and intra-PfRH5 antibody interactions are key determinants of functional anti-parasitic potency. In addition, we identify a germline gene combination that results in an exceptionally potent class of antibody and demonstrate its prophylactic potential to protect against P. falciparum parasite challenge in vivo. This comprehensive dataset provides a framework to guide rational design of next-generation vaccines and prophylactic antibodies to protect against blood-stage malaria.

immunology↗

bioRxiv2023

Humoral correlates of protection against Mycobacterium tuberculosis following intravenous Bacille Calmette-Guerin vaccination in rhesus macaques

Altering the route of Bacille Calmette-Guerin (BCG) immunization from low-dose intradermal vaccination to high-dose intravenous (IV) vaccination resulted in a high level of protection against Mycobacterium tuberculosis (Mtb) infection, providing an opportunity to uncover immune correlates and mechanisms of protection. In addition to strong T cell immunity, IV BCG vaccination was associated with a robust expansion of humoral immune responses that tracked with bacterial control. However, given the near complete protection afforded by high-dose IV BCG immunization, a precise correlate of immune protection was difficult to define. Here we leveraged plasma and bronchoalveolar lavage fluid (BAL) from a cohort of rhesus macaques that received decreasing doses of IV BCG and aimed to define the correlates of immunity across macaques that experienced immune protection or breakthrough infection following Mtb challenge. We show an IV BCG dose-dependent induction of mycobacterial-specific humoral immune responses, both in the plasma and in the airways. Moreover, antibody responses at peak immunogenicity significantly predicted bacterial control following challenge. Multivariate analyses revealed antibody-mediated complement and NK cell activating humoral networks as key functional signatures associated with protective immunity. Collectively, this work extends our understanding of humoral biomarkers and potential mechanisms of IV BCG mediated protection against Mtb.

immunology↗

bioRxiv2023

Autoencoder Model for Translating Omics Signatures

The development of effective therapeutics and vaccines for human diseases requires a systematic understanding of human biology. While animal and in vitro culture models have successfully elucidated the molecular mechanisms of diseases in many studies, they yet fail to adequately recapitulate human biology as evidenced by the predominant likelihood of failure in clinical trials. To address this broadly important problem, we developed AutoTransOP, a neural network autoencoder framework to map omics profiles from designated species or cellular contexts into a global latent space, from which germane information can be mapped between different contexts. This approach performs as well or better than extant machine learning methods and can identify animal/culture-specific molecular features predictive of other contexts, without requiring homology matching. For an especially challenging test case, we successfully apply our framework to a set of inter-species vaccine serology studies, where no 1-1 mapping between human and non-human primate features exists.

bioinformatics↗

bioRxiv2022

Spateo: multidimensional spatiotemporal modeling of single-cell spatial transcriptomics

Cells do not live in a vacuum, but in a milieu defined by cell-cell communication that can be measured via emerging high-resolution spatial transcriptomics approaches. However, analytical tools that fully leverage such data for kinetic modeling remain lacking. Here we present Spateo (aristoteleo/spateo-release), a general framework for quantitative spatiotemporal modeling of single-cell resolution spatial transcriptomics. Spateo delivers novel methods for digitizing spatial layers/columns to identify spatially-polar genes, and develops a comprehensive framework of cell-cell interaction to reveal spatial effects of niche factors and cell type-specific ligand-receptor interactions. Furthermore, Spateo reconstructs 3D models of whole embryos, and performs 3D morphometric analyses. Lastly, Spateo introduces the concept of "morphometric vector field" of cell migrations, and integrates spatial differential geometry to unveil regulatory programs underlying various organogenesis patterns of Drosophila. Thus, Spateo enables the study of the ecology of organs at a molecular level in 3D space, beyond isolated single cells.

systems biology↗