Accurate and Ultra-Efficient p-Value Calculation for Higher Criticism Tests

In modern data science, higher criticism (HC) method is effective for detecting rare and weak signals. The computation, however, has long been an issue when the number of p-values combined ( K ) and/or the number of repeated HC tests ( N ) are large. Some computing methods have been developed, but they all have significant shortcomings, especially when a stringent significance level is required. In this paper, we propose an accurate and highly efficient computing strategy for four variations of HC. Specifically, we propose an unbiased cross-entropy-based importance sampling method (IS C E ) to benchmark all existing computing methods, and develop a modified SetTest method (MST) that resolves numerical issues of the existing SetTest approach. We further develop an ultra-fast approach (UFI) combining pre-calculated statistical tables and cubic spline interpolation. Finally, following extensive simulations, we provide a computing strategy integrating MST, UFI and other existing methods with R package "HCp" for virtually any K and small p-values ( ∼ 10 - 20 ). The method is applied to a COVID-19 disease surveillance example for spatio-temporal outbreak detection from case numbers of 804 days in 3,342 counties in the United States. Results confirm viability of the computing strategy for large-scale inferences. Supplementary materials for this article are available online.

Bayesian screening for feature selection

Biomedical applications such as genome-wide association studies screen large databases with high-dimensional features to identify rare, weakly expressed, and important continuous-valued features for subsequent detailed analysis. We describe an exact, rapid Bayesian screening approach with attractive diagnostic properties using a Gaussian random mixture model focusing on the missed discovery rate (the probability of failing to identify potentially informative features) rather than the false discovery rate ordinarily used with multiple hypothesis testing. The method provides the likelihood that a feature merits further investigation, as well as distributions of the effect magnitudes and the proportion of features with the same expected responses under alternative conditions. Important features include the dependence of the critical values on clinical and regulatory priorities and direct assessment of the diagnostic properties.

Data Fusion in Metabolomics and Proteomics for Biomarker Discovery

Proteomics and metabolomics provide key insights into status and dynamics of biological systems. These molecular studies reveal the complex mechanisms involved in disease or aging processes. Invaluable information can be obtained using various analytical techniques such as nuclear magnetic resonance, liquid chromatography, or gas chromatography coupled to mass spectrometry. Each method has inherent advantages and drawbacks, but they are complementary in terms of biological information.The fusion of different measurements is a complex topic. We describe here a framework allowing combining multiple data sets, provided by different analytical platforms. For each platform, the relevant information is extracted in the first step. The obtained latent variables are then fused and further analyzed. The influence of the original variables is then calculated back and interpreted.

Analysis of the Human Adult Urinary Metabolome Variations with Age, Body Mass Index, and Gender by Implementing a Comprehensive Workflow for Univariate and OPLS Statistical Analyses

Urine metabolomics is widely used for biomarker research in the fields of medicine and toxicology. As a consequence, characterization of the variations of the urine metabolome under basal conditions becomes critical in order to avoid confounding effects in cohort studies. Such physiological information is however very scarce in the literature and in metabolomics databases so far. Here we studied the influence of age, body mass index (BMI), and gender on metabolite concentrations in a large cohort of 183 adults by using liquid chromatography coupled with high-resolution mass spectrometry (LC-HRMS). We implemented a comprehensive statistical workflow for univariate hypothesis testing and modeling by orthogonal partial least-squares (OPLS), which we made available to the metabolomics community within the online Workflow4Metabolomics.org resource. We found 108 urine metabolites displaying concentration variations with either age, BMI, or gender, by integrating the results from univariate p-values and multivariate variable importance in projection (VIP). Several metabolite clusters were further evidenced by correlation analysis, and they allowed stratification of the cohort. In conclusion, our study highlights the impact of gender and age on the urinary metabolome, and thus it indicates that these factors should be taken into account for the design of metabolomics studies.

Discordant expression of desmoglein 2 and 3 at the mRNA and protein levels in nodular and superficial basal cell carcinoma revealed by immunohistochemistry and fluorescent in situ hybridization

BACKGROUND

Basal cell carcinoma (BCC) is the most common human cancer. It is thought that skewed expression of desmogleins (Dsgs) in BCC may promote tumourigenesis.

AIM

To comparatively examine expression of Dsg2/Dsg3, using fluorescent in situ hybridization (FISH) and immunohistochemistry (IHC) in BCC subtypes.

METHODS

In total, 84 frozen sections from patients with various clinical or histological subtypes of BCC were analyzed. Expressions of Dsg2/Dsg3 protein and Dsg2/Dsg3 mRNA were evaluated using IHC and FISH, respectively, in BCC nests and BCC-free epidermis, and then quantitatively measured.

RESULTS

There was loss of correlation between Dsg2 and Dsg3 (IHC) in nodular and superficial BCC (nBCC, sBCC), and significant correlation between Dsg2 and Dsg3 (FISH) in BCC, but not nBCC and sBCC.

CONCLUSIONS

Because more prominent aberrations of Dsg2/Dsg3 expression were seen at the protein than at the mRNA level in BCC, these comparative observations indicate greater importance of events at the proteome level than those at the genome level in tumour functional compartments. Different Dsg2/Dsg3 expression in sBCC and nBCC might corroborate the possibility that sBCC and nBCC are separate conditions. These results may contribute to better understanding of the biological behaviour of BCC.

Cutaneous expressions of interleukin-6 and neutrophil elastase as well as levels of serum IgA antibodies to gliadin nonapeptides, tissue transglutaminase and epidermal transglutaminase: implications for both autoimmunity and autoinflammation involvement in dermatitis herpetiformis

Introduction

Dermatitis herpetiformis (DH) seems to be a chronic immune-mediated inflammatory disease of partially known origin. In light of its known biological functions and its involvement in tissue pathology in other disease states, particularly in nickel-induced allergic contact dermatitis coexisting with DH, it would appear that the central and peripheral response by neutrophils and their mediators (e.g. neutrophil elastase – NE) in DH may be partially mediated by interleukin-6 (IL-6).

The aim of the study was to assess the role of IL -6 in DH lesions by examining the relationships between IL -6/NE cutaneous expression and levels of serum anti-nonapeptides of gliadin (npG) IgA, anti-tissue transglutaminase (tTG) immunoglobulin A (IgA), anti-epidermal transglutaminase (eTG) IgA in DH.

Material and methods

In total, 24 DH patients having IgA cutaneous deposition were studied. Immunohistochemistry on paraffin-embedded sections with quantitative digital morphometry was used to measure the intensity of IL -6 and NE cutaneous expressions. Levels of serum anti-npG IgA, anti-tTG IgA and anti-eTG IgA were evaluated with ELISA.

Results

We found no statistically significant correlation between the NE and IL -6 expression intensities. Our results revealed also a lack of correlations between NE/IL -6 expressions and levels of anti-npG IgA, anti-tTG IgA, anti-eTG IgA in DH. However, the IL -6 expression level was significantly lower than that of NE.

Conclusions

The lack of correlations suggested no substantial interactions between IL -6, NE, IgA/npG, IgA/tTG or IgA/eTG in DH. Presented results might indicate the heterogenetic nature of DH pathogenesis suggesting further that both autoimmune and autoinflammatory phenomena may be involved in DH cutaneous pathology.