Search

The 6th edition of the ACBN newsletter includes information about recent developments, highlights from the research performed in ACBN, such as: We introduce Nikolaus Fortelny and his group, who joined ACBN at the start of the year. They bring in exciting competence in computational systems biology. A feature on nanomaterials in the context of immunology and allergy by Martin Himly - is there a connection between these topics? An especially interesting paper on extracellular vesicles from Melanie Schürz and others, group of Nicole Meisner-Kober.

Gene regulatory networks. Interactive diagrams for the web-browser-based exploration of LCH-subset-specific gene regulatory networks.

Supplementary Figures

Marker genes and genomic regions. Lists of differentially expressed genes between LCH cells and non-LCH immune cells, marker genes for each LCH cell subset, and differentially accessible genomic regions between LCH cell subsets.

Langerhans cell histiocytosis (LCH) is a rare neoplasm predominantly affecting children. It occupies a hybrid position between cancers and inflammatory diseases, which makes it an attractive model for studying cancer development. To explore the molecular mechanisms underlying the pathophysiology of LCH and its characteristic clinical heterogeneity, we investigated the transcriptomic and epigenomic diversity in primary LCH lesions. Using single-cell RNA sequencing, we identified multiple recurrent types of LCH cells within these biopsies, including putative LCH progenitor cells and several subsets of differentiated LCH cells. We confirmed the presence of proliferative LCH cells in all analyzed biopsies using IHC, and we defined an epigenomic and gene-regulatory basis of the different LCH-cell subsets by chromatin-accessibility profiling. In summary, our single-cell analysis of LCH uncovered an unexpected degree of cellular, transcriptomic, and epigenomic heterogeneity among LCH cells, indicative of complex developmental hierarchies in LCH lesions.Significance:This study sketches a molecular portrait of LCH lesions by combining single-cell transcriptomics with epigenome profiling. We uncovered extensive cellular heterogeneity, explained in part by an intrinsic developmental hierarchy of LCH cells. Our findings provide new insights and hypotheses for advancing LCH research and a starting point for personalizing therapy.See related commentary by Gruber et al., p. 1343.This article is highlighted in the In This Issue feature, p. 1325

Background: Cardiac fibroblasts (CFs) have a central role in the ventricular remodeling process associated with different types of fibrosis. Recent studies have shown that fibroblasts do not respond homogeneously to heart injury. Because of the limited set of bona fide fibroblast markers, a proper characterization of fibroblast population heterogeneity in response to cardiac damage is lacking. The purpose of this study was to define CF heterogeneity during ventricular remodeling and the underlying mechanisms that regulate CF function. Methods: Collagen1α1–GFP (green fluorescent protein)–positive CFs were characterized after myocardial infarction (MI) by single-cell and bulk RNA sequencing, assay for transposase-accessible chromatin sequencing, and functional assays. Swine and patient samples were studied using bulk RNA sequencing. Results: We identified and characterized a unique CF subpopulation that emerges after MI in mice. These activated fibroblasts exhibit a clear profibrotic signature, express high levels of Cthrc1 (collagen triple helix repeat containing 1), and localize into the scar. Noncanonical transforming growth factor–β signaling and different transcription factors including SOX9 are important regulators mediating their response to cardiac injury. Absence of CTHRC1 results in pronounced lethality attributable to ventricular rupture. A population of CFs with a similar transcriptome was identified in a swine model of MI and in heart tissue from patients with MI and dilated cardiomyopathy. Conclusions: We report CF heterogeneity and their dynamics during the course of MI and redefine the CFs that respond to cardiac injury and participate in myocardial remodeling. Our study identifies CTHRC1 as a novel regulator of the healing scar process and a target for future translational studies.

The mammalian immune system implements a remarkably effective set of mechanisms for fighting pathogens1. Its main components are haematopoietic immune cells, including myeloid cells that control innate immunity, and lymphoid cells that constitute adaptive immunity2. However, immune functions are not unique to haematopoietic cells, and many other cell types display basic mechanisms of pathogen defence3–5. To advance our understanding of immunology outside the haematopoietic system, here we systematically investigate the regulation of immune genes in the three major types of structural cells: epithelium, endothelium and fibroblasts. We characterize these cell types across twelve organs in mice, using cellular phenotyping, transcriptome sequencing, chromatin accessibility profiling and epigenome mapping. This comprehensive dataset revealed complex immune gene activity and regulation in structural cells. The observed patterns were highly organ-specific and seem to modulate the extensive interactions between structural cells and haematopoietic immune cells. Moreover, we identified an epigenetically encoded immune potential in structural cells under tissue homeostasis, which was triggered in response to systemic viral infection. This study highlights the prevalence and organ-specific complexity of immune gene activity in non-haematopoietic structural cells, and it provides a high-resolution, multi-omics atlas of the epigenetic and transcriptional networks that regulate structural cells in the mouse. Structural cells implement a broad range of immune-regulatory functions beyond their roles as barriers and connective tissues, and they utilize an epigenetically encoded potential for immune gene activation in their rapid response to viral infection.

The adult human skin contains a vast number of T cells that are essential for skin homeostasis and pathogen defense. T cells are first observed in the skin at the early stages of gestation; however, our understanding of their contribution to early immunity has been limited by their low abundance and lack of comprehensive methodologies for their assessment. Here, we describe a new workflow for isolating and expanding significant amounts of T cells from fetal human skin. Using multiparametric flow cytometry and in situ immunofluorescence, we found a large population with a naive phenotype and small populations with a memory and regulatory phenotype. Their molecular state was characterized using single-cell transcriptomics and TCR repertoire profiling. Importantly, culture of total fetal skin biopsies facilitated T cell expansion without a substantial impact on their phenotype, a major prerequisite for subsequent functional assays. Collectively, our experimental approaches and data advance the understanding of fetal skin immunity and potential use in future therapeutic interventions., Summary: This study provides versatile tools for the isolation, phenotyping and expansion of human fetal skin T cells, enabling the study of their complexity and heterogeneity allowing new research in skin development and immunity.

The Bruton tyrosine kinase (BTK) inhibitor ibrutinib provides effective treatment for patients with chronic lymphocytic leukemia (CLL), despite extensive heterogeneity in this disease. To define the underlining regulatory dynamics, we analyze high-resolution time courses of ibrutinib treatment in patients with CLL, combining immune-phenotyping, single-cell transcriptome profiling, and chromatin mapping. We identify a consistent regulatory program starting with a sharp decrease of NF-κB binding in CLL cells, which is followed by reduced activity of lineage-defining transcription factors, erosion of CLL cell identity, and acquisition of a quiescence-like gene signature. We observe patient-to-patient variation in the speed of execution of this program, which we exploit to predict patient-specific dynamics in the response to ibrutinib based on the pre-treatment patient samples. In aggregate, our study describes time-dependent cellular, molecular, and regulatory effects for therapeutic inhibition of B cell receptor signaling in CLL, and it establishes a broadly applicable method for epigenome/transcriptome-based treatment monitoring., Ibrutinib, a Bruton tyrosine kinase inhibitor, provides effective treatment for chronic lymphocytic leukemia (CLL). Here, the authors describe time-dependent molecular changes to malignant cells and to the immune system in patients undergoing ibrutinib therapy, with can be used for therapy monitoring.

By characterizing the poorly defined T cell composition in developing human fetal skin, Reitermaier et al. identify a unique population of TCR αβγδ+ T cells. These cells may contribute to early skin development and immune defense against invading pathogens in utero., T cells in human skin play an important role in the immune defense against pathogens and tumors. T cells are present already in fetal skin, where little is known about their cellular phenotype and biological function. Using single-cell analyses, we identified a naive T cell population expressing αβ and γδ T cell receptors (TCRs) that was enriched in fetal skin and intestine but not detected in other fetal organs and peripheral blood. TCR sequencing data revealed that double-positive (DP) αβγδ T cells displayed little overlap of CDR3 sequences with single-positive αβ T cells. Gene signatures, cytokine profiles and in silico receptor–ligand interaction studies indicate their contribution to early skin development. DP αβγδ T cells were phosphoantigen responsive, suggesting their participation in the protection of the fetus against pathogens in intrauterine infections. Together, our analyses unveil a unique cutaneous T cell type within the native skin microenvironment and point to fundamental differences in the immune surveillance between fetal and adult human skin., Graphical Abstract

Glioblastoma might have widespread effects on the neural organization and cognitive function, and even focal lesions may be associated with distributed functional alterations. However, functional changes do not necessarily follow obvious anatomical patterns and the current understanding of this interrelation is limited. In this study, we used resting-state functional magnetic resonance imaging to evaluate changes in global functional connectivity patterns in 15 patients with glioblastoma. For six patients we followed longitudinal trajectories of their functional connectome and structural tumour evolution using bi-monthly follow-up scans throughout treatment and disease progression. In all patients, unilateral tumour lesions were associated with inter-hemispherically symmetric network alterations, and functional proximity of tumour location was stronger linked to distributed network deterioration than anatomical distance. In the longitudinal subcohort of six patients, we observed patterns of network alterations with initial transient deterioration followed by recovery at first follow-up, and local network deterioration to precede structural tumour recurrence by two months. In summary, the impact of focal glioblastoma lesions on the functional connectome is global and linked to functional proximity rather than anatomical distance to tumour regions. Our findings further suggest a relevance for functional network trajectories as a possible means supporting early detection of tumour recurrence.

Controlled macrophage differentiation and activation in the initiation and resolution of inflammation is crucial for averting progression to chronic inflammatory and autoimmune diseases. Here we show a negative feedback mechanism for proinflammatory IFN-γ activation of macrophages driven by macrophage-associated matrix metalloproteinase 12 (MMP12). Through C-terminal truncation of IFN-γ at 135Glu↓Leu136 the IFN-γ receptor-binding site was efficiently removed thereby reducing JAK-STAT1 signaling and IFN-γ activation of proinflammatory macrophages. In acute peritonitis this signature was absent in Mmp12–/– mice and recapitulated in Mmp12+/+ mice treated with a MMP12-specific inhibitor. Similarly, loss-of-MMP12 increases IFN-γ–dependent proinflammatory markers and iNOS+/MHC class II+ macrophage accumulation with worse lymphadenopathy, arthritic synovitis and lupus glomerulonephritis. In active human systemic lupus erythematosus, MMP12 levels were lower and IFN-γ higher compared to treated patients or healthy individuals. Hence, macrophage proteolytic truncation of IFN-γ attenuates classical activation of macrophages as a prelude for resolving inflammation., IFN-γ is central in inflammatory pathogenesis, response to infection and autoimmune diseases. Here the authors show that MMP12 expression is reduced in patients with SLE and that MMP12 post-translationally truncates IFN-y, inhibiting its function and affecting pathogenesis of mouse models of peritonitis, SLE and rheumatoid arthritis.

Protein interactions shape proteome function and thus biology. Identification of protein interactions is a major goal in molecular biology, but biochemical methods, although improving, remain limited in coverage and accuracy. Whereas computational predictions can guide biochemical experiments, low validation rates of predictions remain a major limitation. Here, we investigated computational methods in the prediction of a specific type of interaction, the inhibitory interactions between proteases and their inhibitors. Proteases generate thousands of proteoforms that dynamically shape the functional state of proteomes. Despite the important regulatory role of proteases, knowledge of their inhibitors remains largely incomplete with the vast majority of proteases lacking an annotated inhibitor. To link inhibitors to their target proteases on a large scale, we applied computational methods to predict inhibitory interactions between proteases and their inhibitors based on complementary data, including coexpression, phylogenetic similarity, structural information, co-annotation, and colocalization, and also surveyed general protein interaction networks for potential inhibitory interactions. In testing nine predicted interactions biochemically, we validated the inhibition of kallikrein 5 by serpin B12. Despite the use of a wide array of complementary data, we found a high false positive rate of computational predictions in biochemical follow-up. Based on a protease-specific definition of true negatives derived from the biochemical classification of proteases and inhibitors, we analyzed prediction accuracy of individual features, thereby we identified feature-specific limitations, which also affected general protein interaction prediction methods. Interestingly, proteases were often not coexpressed with most of their functional inhibitors, contrary to what is commonly assumed and extrapolated predominantly from cell culture experiments. Predictions of inhibitory interactions were indeed more challenging than predictions of nonproteolytic and noninhibitory interactions. In summary, we describe a novel and well-defined but difficult protein interaction prediction task and thereby highlight limitations of computational interaction prediction methods.

Langerhans cell histiocytosis (LCH) is a rare neoplasm predominantly affecting children. It occupies a hybrid position between cancers and inflammatory diseases, which makes it an attractive model for studying cancer development. To explore the molecular mechanisms underlying the pathophysiology of LCH and its characteristic clinical heterogeneity, we investigated the transcriptomic and epigenomic diversity in primary LCH lesions. Using single-cell RNA sequencing, we identified multiple recurrent types of LCH cells within these biopsies, including putative LCH progenitor cells and several subsets of differentiated LCH cells. We confirmed the presence of proliferative LCH cells in all analyzed biopsies using IHC, and we defined an epigenomic and gene-regulatory basis of the different LCH-cell subsets by chromatin-accessibility profiling. In summary, our single-cell analysis of LCH uncovered an unexpected degree of cellular, transcriptomic, and epigenomic heterogeneity among LCH cells, indicative of complex developmental hierarchies in LCH lesions. Significance: This study sketches a molecular portrait of LCH lesions by combining single-cell transcriptomics with epigenome profiling. We uncovered extensive cellular heterogeneity, explained in part by an intrinsic developmental hierarchy of LCH cells. Our findings provide new insights and hypotheses for advancing LCH research and a starting point for personalizing therapy. See related commentary by Gruber et al., p. 1343. This article is highlighted in the In This Issue feature, p. 1325

Cardiac fibroblasts have a central role during the ventricular remodeling process associated with different types of cardiac injury. Recent studies have shown that fibroblasts do not respond homogeneously to heart damage, suggesting that the adult myocardium may contain specialized fibroblast subgroups with specific functions. Due to the limited set of bona fide fibroblast markers, a proper characterization of fibroblast population dynamics in response to cardiac damage is still missing. Using single-cell RNA-seq, we identified and characterized a fibroblast subpopulation that emerges in response to myocardial infarction (MI) in a murine model. These activated fibroblasts exhibit a clear pro-fibrotic signature, express high levels of the hormone CTHRC1 and of the immunomodulatory co-receptor CD200 and localize to the injured myocardium. Combining epigenomic profiling with functional assays, we show Sox9 and the non-canonical TGF-β signaling as important regulators mediating their response to cardiac damage. We show that the absence of CTHRC1, in this activated fibroblast subpopulation, results in pronounced lethality due to ventricular rupture in a mouse model of myocardial infarction. Finally, we find evidence for the existence of similar mechanisms in a pig pre-clinical model of MI and establish a correlation between CTHRC1 levels and cardiac function after MI.

Chronic lymphocytic leukemia (CLL) is a genetically, epigenetically, and clinically heterogeneous disease. Despite this heterogeneity, the Bruton tyrosine kinase (BTK) inhibitor ibrutinib provides effective treatment for the vast majority of CLL patients. To define the underlining regulatory program, we analyzed high-resolution time courses of ibrutinib treatment in closely monitored patients, combining cellular phenotyping (flow cytometry), single-cell transcriptome profiling (scRNA-seq), and chromatin mapping (ATAC-seq). We identified a consistent regulatory program shared across all patients, which was further validated by an independent CLL cohort. In CLL cells, this program starts with a sharp decrease of NF-κB binding, followed by reduced regulatory activity of lineage-defining transcription factors (including PAX5 and IRF4) and erosion of CLL cell identity, finally leading to the acquisition of a quiescence-like gene signature which was shared across several immune cell types. Nevertheless, we observed patient-to-patient variation in the speed of its execution, which we exploited to predict patient-specific dynamics in the response to ibrutinib based on pre-treatment samples. In aggregate, our study describes the cellular, molecular, and regulatory effects of therapeutic B cell receptor inhibition in CLL at high temporal resolution, and it establishes a broadly applicable method for epigenome/transcriptome-based treatment monitoring.

RationaleLung transplantation is the ultimate treatment option for patients with end-stage respiratory diseases but bears the highest mortality rate among all solid organ transplantations due to chronic lung allograft dysfunction (CLAD). The mechanisms leading to CLAD remain elusive due to an insufficient understanding of the complex post-transplant adaptation processes.ObjectivesTo better understand these lung adaptation processes after transplantation and to investigate their association with future changes in allograft function.MethodsWe performed an exploratory cohort study of bronchoalveolar lavage samples from 78 lung recipients and donors. We analysed the alveolar microbiome using 16S rRNA sequencing, the cellular composition using flow cytometry, as well as metabolome and lipidome profiling.Measurements and main resultsWe established distinct temporal dynamics for each of the analysed data sets. Comparing matched donor and recipient samples, we revealed that recipient-specific as well as environmental factors, rather than the donor microbiome, shape the long-term lung microbiome. We further discovered that the abundance of certain bacterial strains correlated with underlying lung diseases even after transplantation. A decline in forced expiratory volume during the first second (FEV1) is a major characteristic of lung allograft dysfunction in transplant recipients. By using a machine learning approach, we could accurately predict future changes in FEV1 from our multi-omics data, whereby microbial profiles showed a particularly high predictive power.ConclusionBronchoalveolar microbiome, cellular composition, metabolome and lipidome show specific temporal dynamics after lung transplantation. The lung microbiome can predict future changes in lung function with high precision.

Deregulated cathepsin proteolysis occurs across numerous cancers, but in vivo substrates mediating tumorigenesis remain ill-defined. Applying 8-plex iTRAQ terminal amine isotopic labeling of substrates (TAILS), a systems-level N-terminome degradomics approach, we identified cathepsin B, H, L, S, and Z in vivo substrates and cleavage sites with the use of six different cathepsin knockout genotypes in the Rip1-Tag2 mouse model of pancreatic neuroendocrine tumorigenesis. Among 1,935 proteins and 1,114 N termini identified by TAILS, stable proteolytic products were identified in wild-type tumors compared with one or more different cathepsin knockouts (17%–44% of 139 cleavages). This suggests a lack of compensation at the substrate level by other cathepsins. The majority of neo-N termini (56%–83%) for all cathepsins was consistent with protein degradation. We validated substrates, including the glycolytic enzyme pyruvate kinase M2 associated with the Warburg effect, the ER chaperone GRP78, and the oncoprotein prothymosin-alpha. Thus, the identification of cathepsin substrates in tumorigenesis improves the understanding of cathepsin functions in normal physiology and cancer. ISSN:2666-3864 ISSN:2211-1247

Glioblastoma is characterized by widespread genetic and transcriptional heterogeneity, yet little is known about the role of the epigenome in glioblastoma disease progression. Here, we present genome-scale maps of DNA methylation in matched primary and recurring glioblastoma tumors, using data from a highly annotated clinical cohort that was selected through a national patient registry. We demonstrate the feasibility of DNA methylation mapping in a large set of routinely collected FFPE samples, and we validate bisulfite sequencing as a multipurpose assay that allowed us to infer a range of different genetic, epigenetic, and transcriptional characteristics of the profiled tumor samples. On the basis of these data, we identified subtle differences between primary and recurring tumors, links between DNA methylation and the tumor microenvironment, and an association of epigenetic tumor heterogeneity with patient survival. In summary, this study establishes an open resource for dissecting DNA methylation heterogeneity in a genetically diverse and heterogeneous cancer, and it demonstrates the feasibility of integrating epigenomics, radiology, and digital pathology for a national cohort, thereby leveraging existing samples and data collected as part of routine clinical practice. In-depth methylation analysis of formalin-fixed paraffin-embedded glioblastoma samples demonstrates heterogeneity between primary and recurring tumors and enables prediction of composition of the tumor microenvironment and insights into progression.

Glioblastoma is characterized by widespread genetic and transcriptional heterogeneity, yet little is known about the role of the epigenome in glioblastoma disease progression. Here, we present genome-scale maps of the DNA methylation dynamics in matched primary and recurring glioblastoma tumors, based on a national population registry and a comprehensively annotated clinical cohort. We demonstrate the feasibility of DNA methylation mapping in a large set of routinely collected formalin-fixed paraffin-embedded (FFPE) samples, and we validate bisulfite sequencing as a multi-purpose assay that allowed us to infer a range of different genetic, epigenetic, and transcriptional tumor characteristics. Based on these data, we identified characteristic differences between primary and recurring tumors, links between DNA methylation and the tumor microenvironment, and an association of epigenetic tumor heterogeneity with patient survival. In summary, this study provides a resource for dissecting DNA methylation heterogeneity in genetically diverse and heterogeneous tumors, and it demonstrates the feasibility of integrating epigenomics, radiology, and digital pathology in a representative national cohort, leveraging samples and data collected as part of routine clinical practice.

The knowledgebase TopFIND is an analysis platform focussed on protein termini, their origin, modification and hence their role on protein structure and function. Here, we present a major update to TopFIND, version 3, which includes a 70% increase in the underlying data to now cover a 90,696 proteins, 165,044 N-termini, 130,182 C-termini, 14,382 cleavage sites and 33,209 substrate cleavages in H. sapiens, M. musculus, A. thaliana, S. cerevisiae and E. coli. New features include the mapping of protein termini and cleavage entries across protein isoforms and significantly, the mapping of protein termini originating from alternative transcription and alternative translation start sites. Furthermore, two analysis tools for complex data analysis based on the TopFIND resource are now available online: TopFINDer, the TopFIND ExploRer, characterizes and annotates proteomics-derived N- or C-termini sets for their origin, sequence context and implications for protein structure and function. Neo-termini are also linked to associated proteases. PathFINDer identifies indirect connections between a protease and list of substrates or termini thus supporting the evaluation of complex proteolytic processes in vivo. To demonstrate the utility of the tools, a recent N-terminomics data set of inflamed murine skin has been re-analyzed. In re-capitulating the major findings originally performed manually, this validates the utility of these new resources. The point of entry for the resource is http://clipserve.clip.ubc.ca/topfind from where the graphical interface, all application programming interfaces (API) and the analysis tools are freely accessible.

Protein–protein interaction networks (interactomes) define the functionality of all biological systems. In apoptosis, proteolysis by caspases is thought to initiate disassembly of protein complexes and cell death. Here we used a quantitative proteomics approach, protein correlation profiling (PCP), to explore changes in cytoplasmic and mitochondrial interactomes in response to apoptosis initiation as a function of caspase activity. We measured the response to initiation of Fas‐mediated apoptosis in 17,991 interactions among 2,779 proteins, comprising the largest dynamic interactome to date. The majority of interactions were unaffected early in apoptosis, but multiple complexes containing known caspase targets were disassembled. Nonetheless, proteome‐wide analysis of proteolytic processing by terminal amine isotopic labeling of substrates (TAILS) revealed little correlation between proteolytic and interactome changes. Our findings show that, in apoptosis, significant interactome alterations occur before and independently of caspase activity. Thus, apoptosis initiation includes a tight program of interactome rearrangement, leading to disassembly of relatively few, select complexes. These early interactome alterations occur independently of cleavage of these protein by caspases.

Almost all regulatory processes in biology ultimately lead to or originate from modifications of protein function. However, it is unclear to which extent each mechanism of regulation actually affects proteins and thus phenotypes. We assessed the extent of N-terminal protein truncation in a global analysis of N-terminomics data and find that most proteins have N-terminally truncated proteoforms. Because N-terminomics analyses do not identify the process generating the identified N-termini, we compared identified termini to the three N-termini generating events: protein cleavage, alternative translation, and alternative splicing. Of these, we sought to identify the most likely cause of N-terminal protein truncations in the human proteome. We found that protease cleavage and alternative protein translation are the likely cause for most shortened proteoforms. However, the vast majority (about 90%) of N-termini remain unexplained by any of these processes identified to date, so revealing large gaps in our knowledge of protein termini and their genesis. Further analysis and annotation of terminomics data is required, to which end we have created the TopFIND database, a major systematic annotation effort for protein termini. We outline the new features in version 3.0 of the updated database and the new bioinformatics tools available and encourage submission of generated data to fill current knowledge gaps.

Network modeling of interactions between proteases and their inhibitors reveals a network of new protein connections and cascades in the protease web., Proteolytic processing is an irreversible posttranslational modification affecting a large portion of the proteome. Protease-cleaved mediators frequently exhibit altered activity, and biological pathways are often regulated by proteolytic processing. Many of these mechanisms have not been appreciated as being protease-dependent, and the potential in unraveling a complex new dimension of biological control is increasingly recognized. Proteases are currently believed to act individually or in isolated cascades. However, conclusive but scattered biochemical evidence indicates broader regulation of proteases by protease and inhibitor interactions. Therefore, to systematically study such interactions, we assembled curated protease cleavage and inhibition data into a global, computational representation, termed the protease web. This revealed that proteases pervasively influence the activity of other proteases directly or by cleaving intermediate proteases or protease inhibitors. The protease web spans four classes of proteases and inhibitors and so links both recently and classically described protease groups and cascades, which can no longer be viewed as operating in isolation in vivo. We demonstrated that this observation, termed reachability, is robust to alterations in the data and will only increase in the future as additional data are added. We further show how subnetworks of the web are operational in 23 different tissues reflecting different phenotypes. We applied our network to develop novel insights into biologically relevant protease interactions using cell-specific proteases of the polymorphonuclear leukocyte as a system. Predictions from the protease web on the activity of matrix metalloproteinase 8 (MMP8) and neutrophil elastase being linked by an inactivating cleavage of serpinA1 by MMP8 were validated and explain perplexing Mmp8 −/− versus wild-type polymorphonuclear chemokine cleavages in vivo. Our findings supply systematically derived and validated evidence for the existence of the protease web, a network that affects the activity of most proteases and thereby influences the functional state of the proteome and cell activity., Author Summary Proteases modify the structure and activity of all proteins by peptide bond hydrolysis and are increasingly recognized as integral regulatory components of numerous biological mechanisms. Deregulated protease activity is a common characteristic of many diseases. However, protease drug development is complicated by an incomplete understanding of protease biology. One missing piece in this puzzle is the interplay between proteases: Some proteases activate other proteases, whereas some proteases inactivate inhibitors, leading to currently unpredictable cleavage of additional proteins. Using database annotations we mathematically modeled protease interactions. Our model includes 1,230 proteins and shows connections between 141,523 pairs of proteases, substrates, and inhibitors. Thus, proteases interact on a large scale to form the protease web, which links most studied groups of proteases and their inhibitors, indicating that the potential of regulation through this network is very large. We found that this interplay is robust to targeted or untargeted pruning of the protease web and that protease inhibitors are central to network connectivity. Our model was used to decipher proteolytic pathways that drive inflammatory processes in vivo. Consequently, protease regulatory interactions should be recognized and explored further to understand in vivo roles and to select better drug targets that avoid side effects arising from inhibition of unexpected activities.

Background Single-cell RNA-seq (scRNA-seq) is emerging as a powerful tool to dissect cell-specific effects of drug treatment in complex tissues. This application requires high levels of precision, robustness, and quantitative accuracy—beyond those achievable with existing methods for mainly qualitative single-cell analysis. Here, we establish the use of standardized reference cells as spike-in controls for accurate and robust dissection of single-cell drug responses. Results We find that contamination by cell-free RNA can constitute up to 20% of reads in human primary tissue samples, and we show that the ensuing biases can be removed effectively using a novel bioinformatics algorithm. Applying our method to both human and mouse pancreatic islets treated ex vivo, we obtain an accurate and quantitative assessment of cell-specific drug effects on the transcriptome. We observe that FOXO inhibition induces dedifferentiation of both alpha and beta cells, while artemether treatment upregulates insulin and other beta cell marker genes in a subset of alpha cells. In beta cells, dedifferentiation and insulin repression upon artemether treatment occurs predominantly in mouse but not in human samples. Conclusions This new method for quantitative, error-correcting, scRNA-seq data normalization using spike-in reference cells helps clarify complex cell-specific effects of pharmacological perturbations with single-cell resolution and high quantitative accuracy.

Deep learning has emerged as a powerful methodology for predicting a variety of complex biological phenomena. However, its utility for biological discovery has so far been limited, given that generic deep neural networks provide little insight into the biological mechanisms that underlie a successful prediction. Here we demonstrate deep learning on biological networks, where every node has a molecular equivalent (such as a protein or gene) and every edge has a mechanistic interpretation (e.g., a regulatory interaction along a signaling pathway).With knowledge-primed neural networks (KPNNs), we exploit the ability of deep learning algorithms to assign meaningful weights to multi-layered networks for interpretable deep learning. We introduce three methodological advances in the learning algorithm that enhance interpretability of the learnt KPNNs: Stabilizing node weights in the presence of redundancy, enhancing the quantitative interpretability of node weights, and controlling for the uneven connectivity inherent to biological networks. We demonstrate the power of our approach on two single-cell RNA-seq datasets, predicting T cell receptor stimulation in a standardizedin vitromodel and inferring cell type in Human Cell Atlas reference data comprising 483,084 immune cells.In summary, we introduce KPNNs as a method that combines the predictive power of deep learning with the interpretability of biological networks. While demonstrated here on single-cell sequencing data, this method is broadly relevant to other research areas where prior domain knowledge can be represented as networks.