Inherited IL-17RC deficiency in patients with chronic mucocutaneous candidiasis

Chronic mucocutaneous candidiasis (CMC) is characterized by recurrent or persistent infections of the skin, nail, oral, and genital mucosae with Candida species, mainly C. albicans. Autosomal-recessive (AR) IL-17RA and ACT1 deficiencies and autosomal-dominant IL-17F deficiency, each reported in a single kindred, underlie CMC in otherwise healthy patients. We report three patients from unrelated kindreds, aged 8, 12, and 37 yr with isolated CMC, who display AR IL-17RC deficiency. The patients are homozygous for different nonsense alleles that prevent the expression of IL-17RC on the cell surface. The defect is complete, abolishing cellular responses to IL-17A and IL-17F homo- and heterodimers. However, in contrast to what is observed for the IL-17RA– and ACT1-deficient patients tested, the response to IL-17E (IL-25) is maintained in these IL-17RC–deficient patients. These experiments of nature indicate that human IL-17RC is essential for mucocutaneous immunity to C. albicans but is otherwise largely redundant.In humans, chronic mucocutaneous candidiasis (CMC) is characterized by infections of the skin, nail, digestive, and genital mucosae with Candida species, mainly C. albicans, a commensal of the gastrointestinal tract in healthy individuals (Puel et al., 2012). CMC is frequent in acquired or inherited disorders involving profound T cell defects (Puel et al., 2010b; Vinh, 2011; Lionakis, 2012). Human IL-17 immunity has recently been shown to be essential for mucocutaneous protection against C. albicans (Puel et al., 2010b, 2012; Cypowyj et al., 2012; Engelhardt and Grimbacher, 2012; Huppler et al., 2012; Ling and Puel, 2014). Indeed, patients with primary immunodeficiencies and syndromic CMC have been shown to display impaired IL-17 immunity (Puel et al., 2010b). Most patients with autosomal-dominant (AD) hyper-IgE syndrome (AD-HIES) and STAT3 deficiency (de Beaucoudrey et al., 2008; Ma et al., 2008; Milner et al., 2008; Renner et al., 2008; Chandesris et al., 2012) and some patients with invasive fungal infection and autosomal-recessive (AR) CARD9 deficiency (Glocker et al., 2009; Lanternier et al., 2013) or Mendelian susceptibility to mycobacterial diseases (MSMD) and AR IL-12p40 or IL-12Rβ1 deficiency (de Beaucoudrey et al., 2008, 2010; Prando et al., 2013; Ouederni et al., 2014) have low proportions of IL-17A–producing T cells and CMC (Cypowyj et al., 2012; Puel et al., 2012). Patients with AR autoimmune polyendocrine syndrome type 1 (APS-1) and AIRE deficiency display CMC and high levels of neutralizing autoantibodies against IL-17A, IL-17F, and/or IL-22 (Browne and Holland, 2010; Husebye and Anderson, 2010; Kisand et al., 2010, 2011; Puel et al., 2010a).These findings paved the way for the discovery of the first genetic etiologies of CMC disease (CMCD), an inherited condition affecting individuals with none of the aforementioned primary immunodeficiencies (Puel et al., 2011; Casanova and Abel, 2013; Casanova et al., 2013, 2014). AR IL-17RA deficiency, AR ACT1 deficiency, and AD IL-17F deficiency were described, each in a single kindred (Puel et al., 2011; Boisson et al., 2013). A fourth genetic etiology of CMCD, which currently appears to be the most frequent, has also been reported: heterozygous gain-of-function (GOF) mutations of STAT1 impairing the development of IL-17–producing T cells (Liu et al., 2011; Smeekens et al., 2011; van de Veerdonk et al., 2011; Hori et al., 2012; Takezaki et al., 2012; Tóth et al., 2012; Al Rushood et al., 2013; Aldave et al., 2013; Romberg et al., 2013; Sampaio et al., 2013; Soltész et al., 2013; Uzel et al., 2013; Wildbaum et al., 2013; Frans et al., 2014; Kilic et al., 2014; Lee et al., 2014; Mekki et al., 2014; Mizoguchi et al., 2014; Sharfe et al., 2014; Yamazaki et al., 2014). We studied three unrelated patients with CMCD without mutations of IL17F, IL17RA, ACT1, or STAT1. We used a genome-wide approach based on whole-exome sequencing (WES). We found AR complete IL-17RC deficiency in all three patients.     

Ranking and combining multiple predictors without labeled data

In a broad range of classification and decision-making problems, one is given the advice or predictions of several classifiers, of unknown reliability, over multiple questions or queries. This scenario is different from the standard supervised setting, where each classifier’s accuracy can be assessed using available labeled data, and raises two questions: Given only the predictions of several classifiers over a large set of unlabeled test data, is it possible to (i) reliably rank them and (ii) construct a metaclassifier more accurate than most classifiers in the ensemble? Here we present a spectral approach to address these questions. First, assuming conditional independence between classifiers, we show that the off-diagonal entries of their covariance matrix correspond to a rank-one matrix. Moreover, the classifiers can be ranked using the leading eigenvector of this covariance matrix, because its entries are proportional to their balanced accuracies. Second, via a linear approximation to the maximum likelihood estimator, we derive the Spectral Meta-Learner (SML), an unsupervised ensemble classifier whose weights are equal to these eigenvector entries. On both simulated and real data, SML typically achieves a higher accuracy than most classifiers in the ensemble and can provide a better starting point than majority voting for estimating the maximum likelihood solution. Furthermore, SML is robust to the presence of small malicious groups of classifiers designed to veer the ensemble prediction away from the (unknown) ground truth.Every day, multiple decisions are made based on input and suggestions from several sources, either algorithms or advisers, of unknown reliability. Investment companies handle their portfolios by combining reports from several analysts, each providing recommendations on buying, selling, or holding multiple stocks (1, 2). Central banks combine surveys of several professional forecasters to monitor rates of inflation, real gross domestic product growth, and unemployment (3–6). Biologists study the genomic binding locations of proteins by combining or ranking the predictions of several peak detection algorithms applied to large-scale genomics data (7). Physician tumor boards convene a number of experts from different disciplines to discuss patients whose diseases pose diagnostic and therapeutic challenges (8). Peer-review panels discuss multiple grant applications and make recommendations to fund or reject them (9). The examples above describe scenarios in which several human advisers or algorithms provide their predictions or answers to a list of queries or questions. A key challenge is to improve decision making by combining these multiple predictions of unknown reliability. Automating this process of combining multiple predictors is an active field of research in decision science (cci.mit.edu/research), medicine (10), business (refs. 11 and 12 and www.kaggle.com/competitions), and government (www.iarpa.gov/Programs/ia/ACE/ace.html and www.goodjudgmentproject.com), as well as in statistics and machine learning.Such scenarios, whereby advisers of unknown reliability provide potentially conflicting opinions, or propose to take opposite actions, raise several interesting questions. How should the decision maker proceed to identify who, among the advisers, is the most reliable? Moreover, is it possible for the decision maker to cleverly combine the collection of answers from all of the advisers and provide even more accurate answers?In statistical terms, the first question corresponds to the problem of estimating prediction performances of preconstructed classifiers (e.g., the advisers) in the absence of class labels. Namely, each classifier was constructed independently on a potentially different training dataset (e.g., each adviser trained on his/her own using possibly different sources of information), yet they are all being applied to the same new test data (e.g., list of queries) for which labels are not available, either because they are expensive to obtain or because they will only be available in the future, after the decision has been made. In addition, the accuracy of each classifier on its own training data is unknown. This scenario is markedly different from the standard supervised setting in machine learning and statistics. There, classifiers are typically trained on the same labeled data and can be ranked, for example, by comparing their empirical accuracy on a common labeled validation set. In this paper we show that under standard assumptions of independence between classifier errors their unknown performances can still be ranked even in the absence of labeled data.The second question raised above corresponds to the problem of combining predictions of preconstructed classifiers to form a metaclassifier with improved prediction performance. This problem arises in many fields, including combination of forecasts in decision science and crowdsourcing in machine learning, which have each derived different approaches to address it. If we had external knowledge or historical data to assess the reliability of the available classifiers we could use well-established solutions relying on panels of experts or forecast combinations (11–14). In our problem such knowledge is not always available and thus these solutions are in general not applicable. The oldest solution that does not require additional information is majority voting, whereby the predicted class label is determined by a rule of majority, with all advisers assigned the same weight. More recently, iterative likelihood maximization procedures, pioneered by Dawid and Skene (15), have been proposed, in particular in crowdsourcing applications (16–23). Owing to the nonconvexity of the likelihood function, these techniques often converge only to a local, rather than global, maximum and require careful initialization. Furthermore, there are typically no guarantees on the quality of the resulting solution.In this paper we address these questions via a spectral analysis that yields four major insights:

1. Under standard assumptions of independence between classifier errors, in the limit of an infinite test set, the off-diagonal entries of the population covariance matrix of the classifiers correspond to a rank-one matrix.
2. The entries of the leading eigenvector of this rank-one matrix are proportional to the balanced accuracies of the classifiers. Thus, a spectral decomposition of this rank-one matrix provides a computationally efficient approach to rank the performances of an ensemble of classifiers.
3. A linear approximation of the maximum likelihood estimator yields an ensemble learner whose weights are proportional to the entries of this eigenvector. This represents an efficient, easily constructed, unsupervised ensemble learner, which we term Spectral Meta-Learner (SML).
4. An interest group of conspiring classifiers (a cartel) that maliciously attempts to veer the overall ensemble solution away from the (unknown) ground truth leads to a rank-two covariance matrix. Furthermore, in contrast to majority voting, SML is robust to the presence of a small-enough cartel whose members are unknown.

In addition, we demonstrate the advantages of spectral approaches based on these insights, using both simulated and real-world datasets. When the independence assumptions hold approximately, SML is typically better than most classifiers in the ensemble and their majority vote, achieving results comparable to the maximum likelihood estimator (MLE). Empirically, we find SML to be a better starting point for computing the MLE that consistently leads to improved performance. Finally, spectral approaches are also robust to cartels and therefore helpful in analyzing surveys where a biased subgroup of advisers (a cartel) may have corrupted the data.     

GluN2B subunit deletion reveals key role in acute and chronic ethanol sensitivity of glutamate synapses in bed nucleus of the stria terminalis

The bed nucleus of the stria terminalis (BNST) is a critical region for alcohol/drug-induced negative affect and stress-induced reinstatement. NMDA receptor (NMDAR)-dependent plasticity, such as long-term potentiation (LTP), has been postulated to play key roles in alcohol and drug addiction; yet, to date, little is understood regarding the mechanisms underlying LTP of the BNST, or its regulation by ethanol. Acute and chronic exposure to ethanol modulates glutamate transmission via actions on NMDARs. Despite intense investigation, tests of subunit specificity of ethanol actions on NMDARs using pharmacological approaches have produced mixed results. Thus, we use a conditional GluN2B KO mouse line to assess both basal and ethanol-dependent function of this subunit at glutamate synapses in the BNST. Deletion of GluN2B eliminated LTP, as well as actions of ethanol on NMDAR function. Further, we show that chronic ethanol exposure enhances LTP formation in the BNST. Using KO-validated pharmacological approaches with Ro25-6981 and memantine, we provide evidence suggesting that chronic ethanol exposure enhances LTP in the BNST via paradoxical extrasynaptic NMDAR involvement. These findings demonstrate that GluN2B is a key point of regulation for ethanol's actions and suggest a unique role of extrasynaptic GluN2B-containing receptors in facilitating LTP.     

On relaxations and aging of various glasses

Slow relaxation occurs in many physical and biological systems. "Creep" is an example from everyday life. When stretching a rubber band, for example, the recovery to its equilibrium length is not, as one might think, exponential: The relaxation is slow, in many cases logarithmic, and can still be observed after many hours. The form of the relaxation also depends on the duration of the stretching, the "waiting time." This ubiquitous phenomenon is called aging, and is abundant both in natural and technological applications. Here, we suggest a general mechanism for slow relaxations and aging, which predicts logarithmic relaxations, and a particular aging dependence on the waiting time. We demonstrate the generality of the approach by comparing our predictions to experimental data on a diverse range of physical phenomena, from conductance in granular metals to disordered insulators and dirty semiconductors, to the low temperature dielectric properties of glasses.     

Mental and Physical Health Consequences of the September 11, 2001 (9/11) Attacks in Primary Care: A Longitudinal Study

The magnitude of the September 11, 2001 (9/11) attacks was without precedent in the United States, but long‐term longitudinal research on its health consequences for primary care patients is limited. We assessed the prevalence and exposure‐related determinants of mental disorders, functioning, general medical conditions, and service utilization, 1 and 4 years after the 9/11 attacks, in an urban primary care cohort (N = 444) in Manhattan. Although the prevalence of posttraumatic stress disorder (PTSD) and levels of functional impairment declined over time, a substantial increase in suicidal ideation and missed work was observed. Most medical outcomes and service utilization indicators demonstrated a short‐term increase after the 9/11 attacks (mean change of +20.3%), followed by a minor decrease in the subsequent year (mean change of ?3.2%). Loss of a close person was associated with the highest risk for poor mental health and functional status over time. These findings highlight the importance of longitudinal assessments of mental, functional, and medical outcomes in urban populations exposed to mass trauma and terrorism.     

Educational lectures enhance knowledge about the human immunodeficiency virus (HIV) and reduce risky behavior and fear among methadone maintenance treatment patients

Abstract

Background: The aim of this study was to characterize human immunodeficiency virus/acquired immunodeficiency syndrome (HIV/AIDS)-related knowledge and stigma among methadone maintenance treatment (MMT) patients and evaluate the contribution of an educational lecture in reducing risky behavior and unjustified overprotective behavior due to fear and stigma among MMT patients. Methods: Patients from an MMT clinic within a tertiary medical center were invited to an educational lecture on HIV/AIDS. Seventy participants (of current 330) were chosen by a random sample (December 2015), plus at-risk patients and HIV patients. Attendee compliance and change in scores of questionnaires on knowledge (modified HIV-K-Q-22) and on sexual and injection behaviors were studied. Results: Forty-six patients (65.7% compliance) attended the lecture, and their knowledge and behavior scores improved 2?weeks post-lecture (knowledge: from 14.2?±?3 to 19.0?±?2.2 [P?<?.0005], sexual behavior: from 12.1?±?2.9 to 8.8?±?3.0 [P?<?.0005], and injection behavior: from 7.3?±?6.2 to 0.2?±?1.3 [P?<?.0005]). The unjustified fear of proximity to HIV carriers reported by 50% attendees fell to 35% post-lecture. Eight months post-lecture, the scores on knowledge and risky behavior of 21 randomly chosen attendees were still better than pre-lecture scores (knowledge: 15.4?±?2.3 vs. 17.2?±?1.8 [paired t test, P?=?.001], sexual behavior: 13.2?±?2.3 vs. 9.7?±?2.9 [P?<?.0005], and injection behavior: 9.3?±?5.6 vs. 2.8?±?3.1 [P?<?.0005]). Drug abuse and treatment adherence were not related to intervention and to risky behavior. Conclusions: More knowledge, less fear, and less risky behavior immediately and at 8?months post-lecture reflect the success and importance of the educational intervention. Future efforts are needed in order to reduce ignorance and fear associated with HIV/AIDS.