Search

The Escherichia coli whole-cell modeling project seeks to create the most detailed computational model of an E. coli cell in order to better understand and predict the behavior of this model organism. Details about the approach, framework, and current version of the model are discussed. Currently, the model includes the functions of 43% of characterized genes, with ongoing efforts to include additional data and mechanisms. As additional information is incorporated in the model, its utility and predictive power will continue to increase, which means that discovery efforts can be accelerated by community involvement in the generation and inclusion of data. This project will be an invaluable resource to the E. coli community that could be used to verify expected physiological behavior, to predict new outcomes and testable hypotheses for more efficient experimental design iterations, and to evaluate heterogeneous data sets in the context of each other through deep curation.

Summary Bacterial and viral pathogens affect the homeostasis of their hosts in complex and oftentimes unintuitive ways. Mechanistic insights into these host–pathogen interactions may one day allow us to discover more effective antimicrobials and antiviral therapies. The mechanisms that affect overall infection outcomes in the tissue and organ scales are often dictated by behaviors at the cellular and subcellular scales, such as cell metabolism, reproduction, aggregation, and survival. Computational modeling allows us to determine these large-scale outcomes by modeling smaller-scale phenomena, and this field of “multiscale computational modeling” has recently seen tremendous growth and innovation. In this article, we review recent studies of multiscale computational modeling of infection and their impact on the mechanistic understanding of host–pathogen interactions toward developing better therapies and multiscale modeling frameworks.

Summary Quantitative systems biology, in which predictive mathematical models are constructed to guide the design of experiments and predict experimental outcomes, is at an exciting transition point, where the foundational scientific principles are becoming established, but the impact is not yet global. The next steps necessary for mathematical modeling to transform biological research and applications, in the same way it has already transformed other fields, is not completely clear. The purpose of this perspective is to forecast possible answers to this question—what needs to happen next—by drawing on the experience gained in another field, specifically meteorology. We review here a number of lessons learned in weather prediction that are directly relevant to biological systems modeling, and that we believe can enable the same kinds of global impact in our field as atmospheric modeling makes today.

Bacteriophages (phages) are the most abundant biological entity in the human body, but until recently the role that phages play in human health was not well characterized. Although phages do not cause infections in human cells, phages can alter the severity of bacterial infections by the dissemination of virulence factors amongst bacterial hosts. Recent studies, made possible with advances in genome engineering and microscopy, have uncovered a novel role for phages in the human body – the ability to modulate the physiology of the mammalian cells that can harbor intracellular bacteria. In this review, we synthesize key results on how phages traverse through mammalian cells – including uptake, distribution, and interaction with intracellular receptors – highlighting how these steps in turn influence host cell killing of bacteria. We discuss the implications of the growing field of phage–mammalian cell interactions for phage therapy.

Summary Half of the bacteria in the human gut microbiome are lysogens containing integrated prophages, which may activate in stressful immune environments. Although lysogens are likely to be phagocytosed by macrophages, whether prophage activation occurs or influences the outcome of bacterial infection remains unexplored. To study the dynamics of bacteria-phage interactions in living cells—in particular, the macrophage-triggered induction and lysis of dormant prophages in the phagosome—we adopted a tripartite system where murine macrophages engulf E. coli, which are lysogenic with an engineered bacteriophage λ, containing a fluorescent lysis reporter. Pre-induced prophages are capable of lysing the host bacterium and propagating infection to neighboring bacteria in the same phagosome. A non-canonical pathway, mediated by PhoP, is involved with the native λ phage induction inside phagocytosed E. coli. These findings suggest two possible mechanisms by which induced prophages may function to aid the bactericidal activity of macrophages.

Recent advances in computer vision and machine learning underpin a collection of algorithms with an impressive ability to decipher the content of images. These deep learning algorithms are being applied to biological images and are transforming the analysis and interpretation of imaging data. These advances are positioned to render difficult analyses routine and to enable researchers to carry out new, previously impossible experiments. Here we review the intersection between deep learning and cellular image analysis and provide an overview of both the mathematical mechanics and the programming frameworks of deep learning that are pertinent to life scientists. We survey the field's progress in four key applications: image classification, image segmentation, object tracking, and augmented microscopy. Last, we relay our labs' experience with three key aspects of implementing deep learning in the laboratory: annotating training data, selecting and training a range of neural network architectures, and deploying solutions. We also highlight existing datasets and implementations for each surveyed application.

Over the last decade, multiple studies have shown that signaling proteins activated in different temporal patterns—such as oscillatory, transient, and sustained—can result in distinct gene expression patterns or cell fates. However, the molecular events that ensure appropriate stimulus- and dose-dependent dynamics are not often understood and are difficult to investigate. Here, we used single-cell analysis to dissect the mechanisms underlying the stimulus- and dose-encoding patterns in the innate immune signaling network. We found that Toll-like receptor (TLR) and interleukin-1 receptor (IL-1R) signaling dynamics relied on a dose-dependent, auto-inhibitory loop that rendered cells refractory to further stimulation. Using inducible gene expression and optogenetics to perturb the network at different levels, we identified the IL-1R–associated kinase1 (IRAK1) as the dose-sensing node responsible for limiting signal flow during the innate immune response. Although the kinase activity of IRAK1 was not required for signal propagation, it played a critical role in inhibiting the nucleocytoplasmic oscillations of the transcription factor NF-κB. Thus, protein activities that may be “dispensable” from a topological perspective can nevertheless be essential in shaping the dynamic response to the external environment.

During the course of a bacterial infection, cells are exposed simultaneously to a range of bacterial and host factors, which converge on the central transcription factor nuclear factor (NF)-κB. How do single cells integrate and process these converging stimuli? Here we tackle the question of how cells process combinatorial signals by making quantitative single-cell measurements of the NF-κB response to combinations of bacterial lipopolysaccharide and the stress cytokine tumor necrosis factor. We found that cells encode the presence of both stimuli via the dynamics of NF-κB nuclear translocation in individual cells, suggesting the integration of NF-κB activity for these stimuli occurs at the molecular and pathway level. However, the gene expression and cytokine secretion response to combinatorial stimuli were more complex, suggesting that other factors in addition to NF-κB contribute to signal integration at downstream layers of the response. Taken together, our results support the theory that during innate immune threat assessment, a pathogen recognized as both foreign and harmful will recruit an enhanced immune response. Our work highlights the remarkable capacity of individual cells to process multiple input signals and suggests that a deeper understanding of signal integration mechanisms will facilitate efforts to control dysregulated immune responses.

During an infection, immune cells must identify the specific level of threat posed by a given bacterial input in order to generate an appropriate response. Given that they use a general non-self-recognition system, known as Toll-like receptors (TLRs), to detect bacteria, it remains unclear how they transmit information about a particular threat. To determine whether host cells can use signaling dynamics to transmit contextual information about a bacterial stimulus, we use live-cell imaging to make simultaneous quantitative measurements of host MAPK and NF-κB signaling, two key pathways downstream of TLRs, and bacterial infection and load. This combined, single-cell approach reveals that NF-κB and MAPK signaling dynamics are sufficient to discriminate between (1) pathogen-associated molecular patterns (PAMPs) versus bacteria, (2) extracellular versus intracellular bacteria, (3) pathogenic versus non-pathogenic bacteria, and (4) the presence or absence of features indicating an active intracellular bacterial infection, such as replication and effector secretion.

Most features of NF-κB activation dynamics vary significantly with respect to ligand type and concentration. The distribution of the time between two nuclear entries is an invariant feature in populations but not individual cells, suggesting an additional level of control, which regulates the overall distribution of translocation timing., The activation dynamics of nuclear factor (NF)-κB have been shown to affect downstream gene expression. On activation, NF-κB shuttles back and forth across the nuclear envelope. Many dynamic features of this shuttling have been characterized, and most features vary significantly with respect to ligand type and concentration. Here, we report an invariant feature with regard to NF-κB dynamics in cellular populations: the distribution—the average, as well as the variance—of the time between two nuclear entries (the period). We find that this period is conserved, regardless of concentration and across several different ligands. Intriguingly, the distributions observed at the population level are not observed in individual cells over 20-h time courses. Instead, the average period of NF-κB nuclear translocation varies considerably among single cells, and the variance is much smaller within a cell than that of the population. Finally, analysis of daughter-cell pairs and isogenic populations indicates that the dynamics of the NF-κB response is heritable but diverges over multiple divisions, on the time scale of weeks to months. These observations are contrary to the existing theory of NF-κB dynamics and suggest an additional level of control that regulates the overall distribution of translocation timing at the population level.

SummaryIncreasing evidence has shown that population dynamics are qualitatively different from single-cell behaviors. Reporters to probe dynamic, single-cell behaviors are desirable yet relatively scarce. Here, we describe an easy-to-implement and generalizable technology to generate reporters of kinase activity for individual cells. Our technology converts phosphorylation into a nucleocytoplasmic shuttling event that can be measured by epifluorescence microscopy. Our reporters reproduce kinase activity for multiple types of kinases and allow for calculation of active kinase concentrations via a mathematical model. Using this technology, we made several experimental observations that had previously been technicallyunfeasible, including stimulus-dependent patterns of c-Jun N-terminal kinase (JNK) and nuclear factor kappa B (NF-κB) activation. We also measured JNK, p38, and ERK activities simultaneously, finding that p38 regulates the peak number, but not the intensity, of ERK fluctuations. Our approach opens the possibility of analyzing a wide range of kinase-mediated processes in individual cells.

Whole-cell modeling promises to facilitate scientific inquiry by prioritizing future experiments based on existing datasets. To test this promise, we compared simulated growth rates with new measurements for all viable single-gene disruption strains in Mycoplasma genitalium. The discrepancies between simulations and experiments led to novel model predictions about specific kinetic parameters that we subsequently validated. These findings represent the first application of whole-cell modeling to accelerate biological discovery.

Mechanistic ‘whole-cell’ models are needed to develop a complete understanding of cell physiology. However, extracting biological insights from whole-cell models requires running and analyzing large numbers of simulations. We developed WholeCellSimDB, a database for organizing whole-cell simulations. WholeCellSimDB was designed to enable researchers to search simulation metadata to identify simulations for further analysis, and quickly slice and aggregate simulation results data. In addition, WholeCellSimDB enables users to share simulations with the broader research community. The database uses a hybrid relational/hierarchical data format architecture to efficiently store and retrieve both simulation setup metadata and results data. WholeCellSimDB provides a graphical Web-based interface to search, browse, plot and export simulations; a JavaScript Object Notation (JSON) Web service to retrieve data for Web-based visualizations; a command-line interface to deposit simulations; and a Python API to retrieve data for advanced analysis. Overall, we believe WholeCellSimDB will help researchers use whole-cell models to advance basic biological science and bioengineering. Database URL: http://www.wholecellsimdb.org Source code repository URL: http://github.com/CovertLab/WholeCellSimDB

Background Lipopolysaccharide (LPS), found in the outer membrane of gram-negative bacteria, elicits a strong response from the transcription factor family Nuclear factor (NF)-κB via Toll-like receptor (TLR) 4. The cellular response to lipopolysaccharide varies depending on the source and preparation of the ligand, however. Our goal was to compare single-cell NF-κB dynamics across multiple sources and concentrations of LPS. Methodology/Principal Findings Using live-cell fluorescence microscopy, we determined the NF-κB activation dynamics of hundreds of single cells expressing a p65-dsRed fusion protein. We used computational image analysis to measure the nuclear localization of the fusion protein in the cells over time. The concentration range spanned up to nine orders of magnitude for three E. coli LPS preparations. We find that the LPS preparations induce markedly different responses, even accounting for potency differences. We also find that the ability of soluble TNF receptor to affect NF-κB dynamics varies strikingly across the three preparations. Conclusions/Significance Our work strongly suggests that the cellular response to LPS is highly sensitive to the source and preparation of the ligand. We therefore caution that conclusions drawn from experiments using one preparation may not be applicable to LPS in general.

Objective: The lesions of Parkinson disease spread through the brain in a characteristic pattern that corresponds to axonal projections. Previous observations suggest that misfolded a-synuclein could behave as a prion, moving from neuron to neuron and causing endogenous a-synuclein to misfold. Here, we characterized and quantified the axonal transport of a-synuclein fibrils and showed that fibrils could be transferred from axons to second-order neurons following anterograde transport. Methods: We grew primary cortical mouse neurons in microfluidic devices to separate somata from axonal projections in fluidically isolated microenvironments. We used live-cell imaging and immunofluorescence to characterize the transport of fluorescent a-synuclein fibrils and their transfer to second-order neurons. Results: Fibrillar a-synuclein was internalized by primary neurons and transported in axons with kinetics consistent with slow component-b of axonal transport (fast axonal transport with saltatory movement). Fibrillar a-synuclein was readily observed in the cell bodies of second-order neurons following anterograde axonal transport. Axon-to-soma transfer appeared not to require synaptic contacts. Interpretation: These results support the hypothesis that the progression of Parkinson disease can be caused by neuron-to-neuron spread of a-synuclein aggregates and that the anatomical pattern of progression of lesions between axonally connected areas results from the axonal transport of such aggregates. That the transfer did not appear to be trans-synaptic gives hope that a-synuclein fibrils could be intercepted by drugs during the extracellular phase of their journey. ANN NEUROL 2012;72:517–524

SummaryUnderstanding how complex phenotypes arise from individual molecules and their interactions is a primary challenge in biology that computational approaches are poised to tackle. We report a whole-cell computational model of the life cycle of the human pathogen Mycoplasma genitalium that includes all of its molecular components and their interactions. An integrative approach to modeling that combines diverse mathematics enabled the simultaneous inclusion of fundamentally different cellular processes and experimental measurements. Our whole-cell model accounts for all annotated gene functions and was validated against a broad range of data. The model provides insights into many previously unobserved cellular behaviors, including in vivo rates of protein-DNA association and an inverse relationship between the durations of DNA replication initiation and replication. In addition, experimental analysis directed by model predictions identified previously undetected kinetic parameters and biological functions. We conclude that comprehensive whole-cell models can be used to facilitate biological discovery.

Viral infection depends on a complex interplay between host and viral factors. Here, the authors link host susceptibility to viral infection to a network encompassing sulfur metabolism, tRNA modification, competitive binding, and programmed ribosomal frameshifting., The iron-sulfur cluster biosynthesis pathway in Escherichia coli exerts a protective effect during lambda phage infection, while a tRNA thiolation pathway enhances viral infection.tRNALys uridine 34 modification inhibits programmed ribosomal frameshifting to influence the ratio of lambda phage proteins gpG and gpGT.The role of the iron-sulfur cluster biosynthesis pathway in infection is indirect, via competitive binding of the shared sulfur donor IscS.Based on the universality of many key components of this network, in both the host and the virus, these findings may have broad relevance to understanding other infections, including viral infection of humans., Viral infection depends on a complex interplay between host and viral factors. Here, we link host susceptibility to viral infection to a network encompassing sulfur metabolism, tRNA modification, competitive binding, and programmed ribosomal frameshifting (PRF). We first demonstrate that the iron-sulfur cluster biosynthesis pathway in Escherichia coli exerts a protective effect during lambda phage infection, while a tRNA thiolation pathway enhances viral infection. We show that tRNALys uridine 34 modification inhibits PRF to influence the ratio of lambda phage proteins gpG and gpGT. Computational modeling and experiments suggest that the role of the iron-sulfur cluster biosynthesis pathway in infection is indirect, via competitive binding of the shared sulfur donor IscS. Based on the universality of many key components of this network, in both the host and the virus, we anticipate that these findings may have broad relevance to understanding other infections, including viral infection of humans.

Latently infecting viruses are an important class of virus that plays a key role in viral evolution and human health. Here we report a genome-scale forward-genetics screen for host-dependencies of the latently-infecting bacteriophage lambda. This screen identified 57 Escherichia coli (E. coli) genes—over half of which have not been previously associated with infection—that when knocked out inhibited lambda phage's ability to replicate. Our results demonstrate a highly integrated network between lambda and its host, in striking contrast to the results from a similar screen using the lytic-only infecting T7 virus. We then measured the growth of E. coli under normal and infected conditions, using wild-type and knockout strains deficient in one of the identified host genes, and found that genes from the same pathway often exhibited similar growth dynamics. This observation, combined with further computational and experimental analysis, led us to identify a previously unannotated gene, yneJ, as a novel regulator of lamB gene expression. A surprising result of this work was the identification of two highly conserved pathways involved in tRNA thiolation—one pathway is required for efficient lambda replication, while the other has anti-viral properties inhibiting lambda replication. Based on our data, it appears that 2-thiouridine modification of tRNAGlu, tRNAGln, and tRNALys is particularly important for the efficient production of infectious lambda phage particles., Author Summary In this study, we took advantage of a new genetic resource for E. coli mutants to screen for previously undiscovered lambda phage host-dependencies. We then assessed the dynamics of infection in these different E. coli mutants and applied a mathematical model of infection in an attempt to further classify the role of these novel interactions. This model-driven approach to biological discovery led us to identify the previously uncharacterized gene yneJ as a regulator of lamB gene expression. In addition, we identified two highly conserved pathways involved in post-transcriptional modification of tRNA—one pathway was required for efficient lambda replication, while the other has anti-viral properties inhibiting lambda replication. This finding is important as it illustrates a new potential anti-viral strategy that could be applied broadly to other viruses.

BackgroundUnderstanding the transcriptional regulatory networks that map out the coordinated dynamic responses of signaling proteins, transcription factors and target genes over time would represent a significant advance in the application of genome wide expression analysis. The primary challenge is monitoring transcription factor activities over time, which is not yet available at the large scale. Instead, there have been several developments to estimate activities computationally. For example, Network Component Analysis (NCA) is an approach that can predict transcription factor activities over time as well as the relative regulatory influence of factors on each target gene.ResultsIn this study, we analyzed a gene expression data set in blood leukocytes from human subjects administered with lipopolysaccharide (LPS), a prototypical inflammatory challenge, in the context of a reconstructed regulatory network including 10 transcription factors, 99 target genes and 149 regulatory interactions. We found that the computationally estimated activities were well correlated to their coordinated action. Furthermore, we found that clustering the genes in the context of regulatory influences greatly facilitated interpretation of the expression data, as clusters of gene expression corresponded to the activity of specific factors or more interestingly, factor combinations which suggest coordinated regulation of gene expression. The resulting clusters were therefore more biologically meaningful, and also led to identification of additional genes under the same regulation.ConclusionUsing NCA, we were able to build a network that accounted for between 8–11% genes in the known transcriptional response to LPS in humans. The dynamic network illustrated changes of transcription factor activities and gene expressions as well as interactions of signaling proteins, transcription factors and target genes.