No file available
To read the file of this research,
you can request a copy directly from the authors.
Employing Feature Selection Algorithms to Determine the Immune State of Mice with Rheumatoid Arthritis
Preprints and early-stage research may not have been peer reviewed yet.
Abstract
The immune response is a dynamic process by which the body determines whether an antigen is self or nonself. The state of this dynamic process is defined by the relative balance and population of inflammatory and regulatory actors which comprise this decision making process. The goal of immunotherapy as applied to, e.g. Rheumatoid Arthritis (RA), then, is to bias the immune state in favor of the regulatory actors - thereby shutting down autoimmune pathways in the response. While there are several known approaches to immunotherapy, the effectiveness of the therapy will depend on how this intervention alters the evolution of this state. Unfortunately, this process is determined not only by the dynamics of the process, but the state of the system at the time of intervention - a state which is difficult if not impossible to determine prior to application of the therapy.
ResearchGate has not been able to resolve any citations for this publication.
Vascularized composite allotransplantation (VCA) encompasses face and limb transplantation, but as with organ transplantation, it requires lifelong regimens of immunosuppressive drugs to prevent rejection. To achieve donor-specific immune tolerance and reduce the need for systemic immunosuppression, we developed a synthetic drug delivery system that mimics a strategy our bodies naturally use to recruit regulatory T cells (T reg ) to suppress inflammation. Specifically, a microparticle-based system engineered to release the T reg -recruiting chemokine CCL22 was used in a rodent hindlimb VCA model. These “Recruitment-MP” prolonged hindlimb allograft survival indefinitely (>200 days) and promoted donor-specific tolerance. Recruitment-MP treatment enriched T reg populations in allograft skin and draining lymph nodes and enhanced T reg function without affecting the proliferative capacity of conventional T cells. With implications for clinical translation, synthetic human CCL22 induced preferential migration of human T reg in vitro. Collectively, these results suggest that Recruitment-MP promote donor-specific immune tolerance via local enrichment of suppressive T reg .
Dry eye disease (DED) is a highly prevalent, ocular disorder characterized by an abnormal tear film and ocular surface. Recent experimental data has suggested that the underlying pathology of DED involves inflammation of the lacrimal functional unit (LFU), comprising the cornea, conjunctiva, lacrimal gland and interconnecting innervation. This inflammation of the LFU ultimately results in tissue deterioration and the symptoms of DED. Moreover, an increase of pathogenic lymphocyte infiltration and the secretion of pro-inflammatory cytokines are involved in the propagation of DED-associated inflammation. Studies have demonstrated that the adoptive transfer of regulatory T cells (Tregs) can mediate the inflammation caused by pathogenic lymphocytes. Thus, as an approach to treating the inflammation associated with DED, we hypothesized that it was possible to enrich the body's own endogenous Tregs by locally delivering a specific combination of Treg inducing factors through degradable polymer microspheres (TRI microspheres; TGF-β1, Rapamycin (Rapa), and IL-2). This local controlled release system is capable of shifting the balance of Treg/T effectors and, in turn, preventing key signs of dry eye disease such as aqueous tear secretion, conjunctival goblet cells, epithelial corneal integrity, and reduce the pro-inflammatory cytokine milieu in the tissue.
We propose a new class of universal kernel functions which admit a linear parametrization using positive semidefinite matrices. These kernels are generalizations of the Sobolev kernel and are defined by piecewise-polynomial functions. The class of kernels is termed "tessellated" as the resulting discriminant is defined piecewise with hyper-rectangular domains whose corners are determined by the training data. The kernels have scalable complexity, but each instance is universal in the sense that its hypothesis space is dense in . Using numerical testing, we show that for the soft margin SVM, this class can eliminate the need for Gaussian kernels. Furthermore, we demonstrate that when the ratio of the number of training data to features is high, this method will significantly outperform other kernel learning algorithms. Finally, to reduce the complexity associated with SDP-based kernel learning methods, we use a randomized basis for the positive matrices to integrate with existing multiple kernel learning algorithms such as SimpleMKL.
The major drawback of microarray data is the ‘curse of dimensionality problem’, this hinders the useful information of dataset and leads to computational instability. Therefore, selecting relevant genes is an imperative in microarray data analysis. Most of the existing schemes employ a two-phase processes: feature selection/extraction followed by classification. In this paper, a statistical test, ANOVA based on MapReduce is proposed to select the relevant features. After feature selection, MapReduce based K-Nearest Neighbor (K-NN) classifier is also proposed to classify the microarray data. These algorithms are successfully implemented on Hadoop framework and comparative analysis is done using various datasets.
Despite enormous progress in basic research, there are many gaps in understanding human immunity. Here we describe how new investigational tools and computational methods promise to improve the diagnosis, prognosis and therapy of the many diseases with components from the immune system.
Abstract
Background
Like microarray-based investigations, high-throughput proteomics techniques require machine learning algorithms to identify biomarkers that are informative for biological classification problems. Feature selection and classification algorithms need to be robust to noise and outliers in the data.
Results
We developed a recursive support vector machine (R-SVM) algorithm to select important genes/biomarkers for the classification of noisy data. We compared its performance to a similar, state-of-the-art method (SVM recursive feature elimination or SVM-RFE), paying special attention to the ability of recovering the true informative genes/biomarkers and the robustness to outliers in the data. Simulation experiments show that a 5 %-~20 % improvement over SVM-RFE can be achieved regard to these properties. The SVM-based methods are also compared with a conventional univariate method and their respective strengths and weaknesses are discussed. R-SVM was applied to two sets of SELDI-TOF-MS proteomics data, one from a human breast cancer study and the other from a study on rat liver cirrhosis. Important biomarkers found by the algorithm were validated by follow-up biological experiments.
Conclusion
The proposed R-SVM method is suitable for analyzing noisy high-throughput proteomics and microarray data and it outperforms SVM-RFE in the robustness to noise and in the ability to recover informative features. The multivariate SVM-based method outperforms the univariate method in the classification performance, but univariate methods can reveal more of the differentially expressed features especially when there are correlations between the features.
This paper investigates the application of the mutual information
criterion to evaluate a set of candidate features and to select an
informative subset to be used as input data for a neural network
classifier. Because the mutual information measures arbitrary
dependencies between random variables, it is suitable for assessing the
“information content” of features in complex classification
tasks, where methods bases on linear relations (like the correlation)
are prone to mistakes. The fact that the mutual information is
independent of the coordinates chosen permits a robust estimation.
Nonetheless, the use of the mutual information for tasks characterized
by high input dimensionality requires suitable approximations because of
the prohibitive demands on computation and samples. An algorithm is
proposed that is based on a “greedy” selection of the
features and that takes both the mutual information with respect to the
output class and with respect to the already-selected features into
account. Finally the results of a series of experiments are discussed
Dendritic cells (DCs) rely on glycolysis for their energy needs to induce pro-inflammatory antigen-specific immune responses. Therefore, inhibiting DC glycolysis, while presenting the self-antigen, may prevent pro-inflammatory antigen-specific immune responses. Previously we demonstrated that microparticles with alpha-ketoglutarate (aKG) in the polymer backbone (paKG MPs) were able to generate anti-inflammatory DCs by sustained delivery of the aKG metabolite, and by modulating energy metabolism of DCs. Herein, we demonstrate that paKG MPs-based delivery of a glycolytic inhibitor, PFK15, using paKG MPs induces anti-inflammatory DCs (CD86LoMHCII⁺) by down-regulating glycolysis, CD86, tnf and IL-6 genes, while upregulating oxidative phosphorylation (OXPHOS) and mitochondrial genes. Furthermore, paKG MPs delivering PFK15 and a self-antigen, collagen type II (bc2), in vivo, in a collagen-induced autoimmune arthritis (CIA) mouse model, normalized paw inflammation and arthritis score, by generating antigen-specific immune responses. Specifically, these formulations were able to reduce activation of DCs in draining lymph nodes and impressively generated proliferating bc2-specific anti-inflammatory regulatory T cells in joint-associated popliteal lymph nodes. These data strongly suggest that sustained glycolytic inhibition of DCs in the presence of an antigen can induce antigen-specific immunosuppressive responses, therefore, generating a technology that can be applicable for treating autoimmune diseases.
Metabolites control immune cell functions, and delivery of these metabolites in a sustained manner may be able to modulate function of the immune cells . In this study, alpha-ketoglutarate (aKG) and diol based polymeric-microparticles (termed paKG MPs) were synthesized to provide sustained release of aKG and promote an immunosuppressive cellular phenotype. Notably, after association with dendritic cells (DCs), paKG MPs modulated the intracellular metabolic-profile/pathways, and decreased glycolysis and mitochondrial respiration in vitro. These metabolic changes resulted in modulation of MHC-II, CD86 expression in DCs, and altered the frequency of regulatory T cells (Tregs), and T-helper type-1/2/17 cells in vitro. This unique strategy of intracellular delivery of key-metabolites in a sustaine manner provide a paradigm-shift in the immunometabolism field-based immunotherapy with applications in different diseases associated with immune disorders.
Significance
Vascularized composite allotransplantation (VCA) is an emerging field that is particularly beneficial for select amputees and patients with devastating soft tissue loss that is not amenable to conventional reconstructive surgeries. As with solid organ transplantation, VCA recipients are subjected to a lifelong regimen of antirejection drugs with a well-established sequela. Herein, we report a synthetic, controlled-release microparticle system (referred to as TRI-MP) that aims to locally enrich naturally occurring, suppressive lymphocytes to prevent allograft rejection and promote tolerance. While this study exclusively focuses on VCA, this technology has implications for other conditions characterized by aberrant inflammation.
Dry eye disease (DED) is a common ocular disorder affecting millions of individuals worldwide. The pathology of DED involves the infiltration of CD4⁺ lymphocytes, leading to tear film instability and destructive inflammation. In the healthy steady state, a population of immunosuppressive T-cells called regulatory T-cells (Treg) regulates proliferation of immune cells that would otherwise lead to a disruption of immunological homeostasis. For this reason, it has been suggested that Tregs could restore the immunological imbalance in DED. To this end, one possible approach would be to recruit the body's own, endogenous Tregs in order to enrich them at the site of inflammation and tissue destruction. Previously, we have demonstrated a reduction of inflammation and disease symptoms in models of periodontitis corresponding to recruitment of endogenous Tregs, which was accomplished by local placement of controlled release systems that sustain a gradient of the chemokine CCL22, referred to here as Treg-recruiting microspheres. Given that DED is characterized by a pro-inflammatory environment resulting in local tissue destruction, we hypothesized that the controlled release of CCL22 could also recruit Tregs to the ocular surface potentially mediating inflammation and symptoms of DED. Indeed, data suggest that Treg-recruiting microspheres are capable of overcoming the immunological imbalance of Tregs and CD4⁺ IFN-γ⁺ cells in the lacrimal gland. Administration of Treg-recruiting microspheres effectively mitigated the symptoms of DED as measured through a number of outcomes such as tear clearance, goblet cells density and corneal epithelial integrity, suggesting that recruitment of endogenous Treg can mitigate inflammation associated with DED.
As tumors employ complementary overlapping and/or independent mechanisms to evade immune surveillance, many emerging cancer immunotherapies attempt to target multiple pathways to eradicate malignant cells. Although modulation of independent pathways by simultaneous administration of multiple immune modulators (e.g., checkpoint inhibitors, cytokines, and growth factors) has shown great promise, the clinical impact remains limited due to severe toxicity associated with high systemic levels of many of these drugs. Therefore, novel platforms for efficient delivery of multi-component therapies at lower effective doses would be enabling. Here, a drug delivery platform called immunomodulatory molecule delivery system (iMods), which provides sustained extracellular delivery of a checkpoint inhibitor (anti-PD-L1) and simultaneously, targeted intracellular delivery of a tumor antigen (OVA) along with adjuvant (poly(I:C)), and the indoleamine deoxygenase inhibitor 1-MT is described. In melanoma tumor-bearing mice, combinatorial delivery of these factors with iMods leads to regression of both treated and untreated (contralateral) melanoma tumors and 100% survival. These promising therapeutic outcomes are attributed to significantly enhanced ratios of anti-tumor CD8 T-cell/tumor-protective regulatory T-cell (Treg) in tumors and tumor draining lymph nodes. Overall, the iMods delivery platform described here represents a promising advance in multi-factor cancer immunotherapy.
The human immune system is highly variable between individuals but relatively stable over time within a given person. Recent conceptual and technological advances have enabled systems immunology analyses, which reveal the composition of immune cells and proteins in populations of healthy individuals. The range of variation and some specific influences that shape an individual's immune system is now becoming clearer. Human immune systems vary as a consequence of heritable and non-heritable influences, but symbiotic and pathogenic microbes and other non-heritable influences explain most of this variation. Understanding when and how such influences shape the human immune system is key for defining metrics of immunological health and understanding the risk of immune-mediated and infectious diseases.
In recent years, a plethora of approaches have been proposed to deal with the increasingly challenging task of multi‐output regression. This study provides a survey on state‐of‐the‐art multi‐output regression methods, that are categorized as problem transformation and algorithm adaptation methods. In addition, we present the mostly used performance evaluation measures, publicly available data sets for multi‐output regression real‐world problems, as well as open‐source software frameworks. WIREs Data Mining Knowl Discov 2015, 5:216–233. doi: 10.1002/widm.1157
This article is categorized under: Technologies > Machine Learning
Plenty of feature selection methods are available in literature due to the availability of data with hundreds of variables leading to data with very high dimension. Feature selection methods provides us a way of reducing computation time, improving prediction performance, and a better understanding of the data in machine learning or pattern recognition applications. In this paper we provide an overview of some of the methods present in literature. The objective is to provide a generic introduction to variable elimination which can be applied to a wide array of machine learning problems. We focus on Filter, Wrapper and Embedded methods. We also apply some of the feature selection techniques on standard datasets to demonstrate the applicability of feature selection techniques.
DNA micro-arrays now permit scientists to screen thousands of genes simultaneously and determine whether those genes are active, hyperactive or silent in normal or cancerous tissue. Because these new micro-array devices generate bewildering amounts of raw data, new analytical methods must be developed to sort out whether cancer tissues have distinctive signatures of gene expression over normal tissues or other types of cancer tissues.
In this paper, we address the problem of selection of a small subset of genes from broad patterns of gene expression data, recorded on DNA micro-arrays. Using available training examples from cancer and normal patients, we build a classifier suitable for genetic diagnosis, as well as drug discovery. Previous attempts to address this problem select genes with correlation techniques. We propose a new method of gene selection utilizing Support Vector Machine methods based on Recursive Feature Elimination (RFE). We demonstrate experimentally that the genes selected by our techniques yield better classification performance and are biologically relevant to cancer.
In contrast with the baseline method, our method eliminates gene redundancy automatically and yields better and more compact gene subsets. In patients with leukemia our method discovered 2 genes that yield zero leave-one-out error, while 64 genes are necessary for the baseline method to get the best result (one leave-one-out error). In the colon cancer database, using only 4 genes our method is 98% accurate, while the baseline method is only 86% accurate.
Immunology as a branch of the biological sciences has advanced tremendously over the last 50 or so years. In this time, clonal selection has advanced from theory to established fact and the basic structures of antibodies and T cell receptors have been determined, together with their remarkable (and thus far, unique) mechanisms of diversification. A whole system of innate immune receptors and responses has been discovered and elucidated very rapidly, and lymphocytes and other hematopoetic cells can now be subdivided into at least 15 different distinct types. Dozens of cytokines and chemokines have been identified as mediators of cellular communication and we are the proud possessors of 350 CD antigens. A field that was once known chiefly for its impenetrable jargon, the byzantine complexity of its experiments, acrimonious disputes, and excessive theorizing is now the very model of a modern, superbly integrated, and rich biological field, one of the most successful in biology (we still have the impenetrable jargon, but, oh well). We can even claim to have saved the most lives through vaccines (albeit indirectly, as most were formulated before immunology could offer much help) and helped bring about a whole new type of pharmacology in the form of specific antibodies as drugs, not to mention the promise of direct immunomodulation made possible by knowledge of specific pathways.
In this paper we try to identify potential biomarkers for early stroke diagnosis using surface-enhanced laser desorption/ionization mass spectrometry coupled with analysis tools from machine learning and data mining. Data consist of 42 specimen samples, i.e., mass spectra divided in two big categories, stroke and control specimens. Among the stroke specimens two further categories exist that correspond to ischemic and hemorrhagic stroke; in this paper we limit our data analysis to discriminating between control and stroke specimens. We performed two suites of experiments. In the first one we simply applied a number of different machine learning algorithms; in the second one we have chosen the best performing algorithm as it was determined from the first phase and coupled it with a number of different feature selection methods. The reason for this was 2-fold, first to establish whether feature selection can indeed improve performance, which in our case it did not seem to confirm, but more importantly to acquire a small list of potentially interesting biomarkers. Of the different methods explored the most promising one was support vector machines which gave us high levels of sensitivity and specificity. Finally, by analyzing the models constructed by support vector machines we produced a small set of 13 features that could be used as potential biomarkers, and which exhibited good performance both in terms of sensitivity, specificity and model stability.
Feature selection for classification. Intelligent data analysis
- M Dash
- H Liu
Dash, M., Liu, H.: Feature selection for classification. Intelligent data analysis 1 (1997)
Feature selection using principal component analysis
- F Song
- Z Guo
- D Mei
Song, F., Guo, Z., Mei, D.: Feature selection using principal component analysis. 2010 International Conference
on System Science, Engineering Design and Manufacturing Informatization 1 (2010)
Feature selection in proteomic pattern data with support vector machines
- K Jong
- E Marchiori
- M Sebag
- A W Vaart
Jong, K., Marchiori, E., Sebag, M., Vaart, A.W.: Feature selection in proteomic pattern data with support
vector machines. 2004 Symposium on Computational Intelligence in Bioinformatics and Computational Biology
(2004)