Vipin Narang

Systems immunology: a survey of modeling formalisms, applications and simulation tools

Immunological studies frequently analyze individual components (e.g., signaling pathways) of immune systems in a reductionist manner. In contrast, systems immunology aims to give a synthetic understanding of how these components function together as a whole. While immunological research involves in vivo and in vitro experiments, systems immunology research can also be conducted in silico. With an increasing interest in systems-level studies spawned by high-throughput technologies, many immunologists are looking forward to insights provided by computational modeling and simulation. However, modeling and simulation research has mainly been conducted in computational fields, and therefore, little material is available or accessible to immunologists today. This survey is an attempt at bridging the gap between immunologists and systems immunology modeling and simulation. Modeling and simulation refer to building and executing an in silico replica of an immune system. Models are specified within a mathematical or algorithmic framework called formalism and then implemented using software tools. A plethora of modeling formalisms and software tools are reported in the literature for systems immunology. However, it is difficult for a new entrant to the field to know which of these would be suitable for modeling an immunological application at hand. This paper covers three aspects. First, it introduces the field of system immunology emphasizing on the modeling and simulation components. Second, it gives an overview of the principal modeling formalisms, each of which is illustrated with salient applications in immunological research. This overview of formalisms and applications is conducted not only to illustrate their power but also to serve as a reference to assist immunologists in choosing the best formalism for the problem at hand. Third, it lists major software tools, which can be used to practically implement models in these formalisms. Combined, these aspects can help immunologists to start experimenting with in silico models. Finally, future research directions are discussed. Particularly, we identify integrative frameworks to facilitate the coupling of different modeling formalisms and modeling the adaptation properties through evolution of immune systems as the next key research efforts necessary to further develop the multidisciplinary field of systems immunology.

Functionally diverse human T cells recognize non-microbial antigens presented by MR1

MHC class I-related molecule MR1 presents riboflavin- and folate-related metabolites to mucosal-associated invariant T cells, but it is unknown whether MR1 can present alternative antigens to other T cell lineages. In healthy individuals we identified MR1-restricted T cells (named MR1T cells) displaying diverse TCRs and reacting to MR1-expressing cells in the absence of microbial ligands. Analysis of MR1T cell clones revealed specificity for distinct cell-derived antigens and alternative transcriptional strategies for metabolic programming, cell cycle control and functional polarization following antigen stimulation. Phenotypic and functional characterization of MR1T cell clones showed multiple chemokine receptor expression profiles and secretion of diverse effector molecules, suggesting functional heterogeneity. Accordingly, MR1T cells exhibited distinct T helper-like capacities upon MR1-dependent recognition of target cells expressing physiological levels of surface MR1. These data extend the role of MR1 beyond microbial antigen presentation and indicate MR1T cells are a normal part of the human T cell repertoire. DOI: http://dx.doi.org/10.7554/eLife.24476.001

IgG1 memory B cells keep the memory of IgE responses

The unique differentiation of IgE cells suggests unconventional mechanisms of IgE memory. IgE germinal centre cells are transient, most IgE cells are plasma cells, and high affinity IgE is produced by the switching of IgG1 cells to IgE. Here we investigate the function of subsets of IgG1 memory B cells in IgE production and find that two subsets of IgG1 memory B cells, CD80+CD73+ and CD80−CD73−, contribute distinctively to the repertoires of high affinity pathogenic IgE and low affinity non-pathogenic IgE. Furthermore, repertoire analysis indicates that high affinity IgE and IgG1 plasma cells differentiate from rare CD80+CD73+ high affinity memory clones without undergoing further mutagenesis. By identifying the cellular origin of high affinity IgE and the clonal selection of high affinity memory B cells into the plasma cell fate, our findings provide fundamental insights into the pathogenesis of allergies, and on the mechanisms of antibody production in memory B cell responses.IgE is an important mediator of protective immunity as well as allergic reaction, but how high affinity IgE antibodies are produced in memory responses is not clear. Here the authors show that IgE can be generated via class-switch recombination in IgG1 memory B cells without additional somatic hypermutation.

Gene network inference using continuous time Bayesian networks: a comparative study and application to Th17 cell differentiation

BackgroundDynamic aspects of gene regulatory networks are typically investigated by measuring system variables at multiple time points. Current state-of-the-art computational approaches for reconstructing gene networks directly build on such data, making a strong assumption that the system evolves in a synchronous fashion at fixed points in time. However, nowadays omics data are being generated with increasing time course granularity. Thus, modellers now have the possibility to represent the system as evolving in continuous time and to improve the models’ expressiveness.ResultsContinuous time Bayesian networks are proposed as a new approach for gene network reconstruction from time course expression data. Their performance was compared to two state-of-the-art methods: dynamic Bayesian networks and Granger causality analysis. On simulated data, the methods comparison was carried out for networks of increasing size, for measurements taken at different time granularity densities and for measurements unevenly spaced over time. Continuous time Bayesian networks outperformed the other methods in terms of the accuracy of regulatory interactions learnt from data for all network sizes. Furthermore, their performance degraded smoothly as the size of the network increased. Continuous time Bayesian networks were significantly better than dynamic Bayesian networks for all time granularities tested and better than Granger causality for dense time series. Both continuous time Bayesian networks and Granger causality performed robustly for unevenly spaced time series, with no significant loss of performance compared to the evenly spaced case, while the same did not hold true for dynamic Bayesian networks. The comparison included the IRMA experimental datasets which confirmed the effectiveness of the proposed method. Continuous time Bayesian networks were then applied to elucidate the regulatory mechanisms controlling murine T helper 17 (Th17) cell differentiation and were found to be effective in discovering well-known regulatory mechanisms, as well as new plausible biological insights.ConclusionsContinuous time Bayesian networks were effective on networks of both small and large size and were particularly feasible when the measurements were not evenly distributed over time. Reconstruction of the murine Th17 cell differentiation network using continuous time Bayesian networks revealed several autocrine loops, suggesting that Th17 cells may be auto regulating their own differentiation process.

Automated Identification of Core Regulatory Genes in Human Gene Regulatory Networks.

Human gene regulatory networks (GRN) can be difficult to interpret due to a tangle of edges interconnecting thousands of genes. We constructed a general human GRN from extensive transcription factor and microRNA target data obtained from public databases. In a subnetwork of this GRN that is active during estrogen stimulation of MCF-7 breast cancer cells, we benchmarked automated algorithms for identifying core regulatory genes (transcription factors and microRNAs). Among these algorithms, we identified K-core decomposition, pagerank and betweenness centrality algorithms as the most effective for discovering core regulatory genes in the network evaluated based on previously known roles of these genes in MCF-7 biology as well as in their ability to explain the up or down expression status of up to 70% of the remaining genes. Finally, we validated the use of K-core algorithm for organizing the GRN in an easier to interpret layered hierarchy where more influential regulatory genes percolate towards the inner layers. The integrated human gene and miRNA network and software used in this study are provided as supplementary materials (S1 Data) accompanying this manuscript.

An approach to divide pre-detected Devanagari words from the scene images into characters

A methodology to segment the Devanagari words, extracted from the scene images, into characters is presented. Scene images include street signs, shop names, product advertisements, posters on streets, etc. Such words are prone to multiple sources of noise and these make the segmentation very challenging. The problem gets more complicated while developing the text recognition methodologies for different scripts because there is no general solution to this problem and recognizing text in some scripts can be tougher than in others. An indigenous database is created for this purpose. It consists of 130 samples, manually extracted from 200 natural scene images. The results obtained by applying the proposed techniques are encouraging. The average performance is found to be 55.77 %. The execution time for a typical word of size 1169 × 353 is found to be 4.76 s. The database and the results can serve as baseline for the future researchers.

Comparing mathematical models of cell adhesion in tumors

Cancer progression accompanies changes in cell adhesion characteristics, which is crucial for cancer cell invasion and metastasis. Drugs altering cell adhesion have been suggested as a possible therapeutic treatment for cancer. Tumor cell adhesion is thus an important topic of current investigation. Recently individual cell (agent)-based in silico tumor models have incorporated mathematical models of cell adhesion. However, cell adhesion has been modeled in various ways in different studies. The first, Lennard-Jones potential model, is extrapolated from attractive/repulsive interactions between inert molecules. The second, JKR model, is derived from van-der Waals contact forces between non-spontaneously adhering solid elastic bodies. The third, cellular Potts model, is an adaptation of the Ising model of ferromagnetism to a spontaneously adhering population of cells. We compare these three mathematical models of cell adhesion and show how they give different perspectives to tumor growth, morphology and effectiveness of therapy. We also discuss how these models predict multi-nodular tumor morphology which is hitherto unaddressed in the literature.

Devanagari Character Recognition in Scene Images

Character recognition in scene images is an extremely challenging task. Although several techniques are reported performing well, they pertain to English only. This paper focuses on Devanagari character recognition from scene images. Devanagari script is very popular language and has very typical characteristics different from other scripts, particularly English. Combination of basic Devanagari consonants and vowels in multi-variegated ways can yield as many as 100s of characters. Building a classifier to recognize all these classes will be a difficult task. To alleviate this problem, a novel part-based model technique is proposed. 40 basic classes were identified from the Devanagari script for the same purpose. The technique was proposed so as to classify an instance of one these classes in any given test sample. Procuring a large dataset for training is not feasible in the case of scene images. To simultaneously solve this problem, we developed our technique that can use either the machine printed or the handwritten dataset for training. We present our results on the publicly available dataset (DSIW2K) containing images of street scenes taken in New Delhi, India.

Publisher Correction: IgG1 memory B cells keep the memory of IgE responses

The originally published version of this Article contained errors in Fig. 4 that were introduced during the production process. In panel c, the two uppermost labels ‘IgE spleen’ and ‘IgE BM’ incorrectly read ‘IgG1 spleen’ and ‘IgE1 BM’, respectively. These errors have now been corrected in both the PDF and HTML versions of the Article.

Influenza Vaccine-Induced Antibody Responses Are Not Impaired by Frailty in the Community-Dwelling Elderly With Natural Influenza Exposure

Background: Elderly adults over 65 years of age are recommended to receive seasonal influenza vaccination as they are at a higher risk of infection and its complications than the younger community. The elderly are often stratified according to frailty status where frail individuals are more susceptible to adverse health outcomes than their non-frail counterparts, however, it is not known whether immunity induced by influenza vaccination is impaired in the frail elderly. Study Design: Two hundred and five elderly subjects of Chinese ethnicity in Singapore (mean age 73.3 ± 5.3 years, 128 females and 77 males) were administered the recommended trivalent inactivated 2013–14 seasonal influenza vaccine (Vaxigrip™) containing A/H1N1, A/H3N2, and B strains. The elderly subjects were stratified into three groups according to Frieds frailty criteria (59 frail, 85 pre-frail, 61 robust) and were also ranked by Rockwoods frailty index (RFI). Statistical associations were evaluated between frailty status and pre- and post-vaccination antibody titres in sera measured by Hemagglutination inhibition (HAI) and microneutralization (MN) assays. Immunological responses across frailty strata were also studied in terms of leukocyte cellular distribution, cytokine levels and gene expression. Results: Post-vaccination, 83.4% of the subjects seroconverted for A/H1N1, 80.5% for A/H3N2, and 81% for the B strain. The seroconversion rates were comparable across frailty groups (A/H1N1, ANOVA, p = 0.7910; A/H3N2, ANOVA, p = 0.8356, B, ANOVA, p = 0.9741). Geometric mean titres of HAI and MN as well as seroprotection rates were also similar in all three frailty groups and uncorrelated with RFI (Spearman, r = 0.023, p = 0.738). No statistically significant differences were observed between the frailty groups in vaccine-induced modulation of leukocyte populations, cytokine responses, and gene expression profiles of peripheral blood mononuclear cells (PBMCs). Whereas, post- and pre-vaccination HAI titres were positively correlated after adjusting for age and gender (A/H1N1, R2 = 0.216, p = 9.1e−11; A/H3N2, R2 = 0.166, p = 3.4e−8; B, R2 = 0.104, p = 3.1e−5). With most subjects lacking previous history of influenza vaccination, the pre-vaccination titres were likely due to natural exposure and seen to match the pattern of influenza subtype prevalence in the time period of vaccination. Conclusion: The majority of the elderly subjects seroconverted for seasonal influenza upon vaccination, and importantly, influenza vaccination-induced humoral immune responses and seroprotection were similar across the frailty strata, indicating that frail individuals may also benefit from influenza vaccination. Pre-existing antibodies due to natural exposure appeared to positively influence vaccine-induced antibody responses.