Hidden multiple comparisons increase forensic error rates.

When wires are cut, the tool produces striations on the cut surface; as in other forms of forensic analysis, these striation marks are used to connect the evidence to the source that created them. Here, we argue that the practice of comparing two wire cut surfaces introduces complexities not present in better-investigated forensic examination of toolmarks such as those observed on bullets, as wire comparisons inherently require multiple distinct comparisons, increasing the expected false discovery rate. We call attention to the multiple comparison problem in wire examination and relate it to other situations in forensics that involve multiple comparisons, such as database searches.

Empirical Bayes small area prediction under a zero-inflated lognormal model with correlated random area effects.

Many variables of interest in agricultural or economical surveys have skewed distributions and can equal zero. Our data are measures of sheet and rill erosion called Revised Universal Soil Loss Equation - 2 (RUSLE2). Small area estimates of mean RUSLE2 erosion are of interest. We use a zero-inflated lognormal mixed effects model for small area estimation. The model combines a unit-level lognormal model for the positive RUSLE2 responses with a unit-level logistic mixed effects model for the binary indicator that the response is nonzero. In the Conservation Effects Assessment Project (CEAP) data, counties with a higher probability of nonzero responses also tend to have a higher mean among the positive RUSLE2 values. We capture this property of the data through an assumption that the pair of random effects for a county are correlated. We develop empirical Bayes (EB) small area predictors and a bootstrap estimator of the mean squared error (MSE). In simulations, the proposed predictor is superior to simpler alternatives. We then apply the method to construct EB predictors of mean RUSLE2 erosion for South Dakota counties. To obtain auxiliary variables for the population of cropland in South Dakota, we integrate a satellite-derived land cover map with a geographic database of soil properties. We provide an R Shiny application called viscover (available at https://lyux.shinyapps.io/viscover/) to visualize the overlay operations required to construct the covariates. On the basis of bootstrap estimates of the mean square error, we conclude that the EB predictors of mean RUSLE2 erosion are superior to direct estimators.

Severe fever with thrombocytopenia virus glycoproteins are targeted by neutralizing antibodies and can use DC-SIGN as a receptor for pH-dependent entry into human and animal cell lines.

Severe fever with thrombocytopenia syndrome virus (SFTSV) is a novel bunyavirus that recently emerged in China. Infection with SFTSV is associated with case-fatality rates of up to 30%, and neither antivirals nor vaccines are available at present. Development of antiviral strategies requires the elucidation of virus-host cell interactions. Here, we analyzed host cell entry of SFTSV. Employing lentiviral and rhabdoviral vectors, we found that the Gn/Gc glycoproteins (Gn/Gc) of SFTSV mediate entry into a broad range of human and animal cell lines, as well as human macrophages and dendritic cells. The Gn/Gc proteins of La Crosse virus (LACV) and Rift Valley Fever Virus (RVFV), other members of the bunyavirus family, facilitated entry into an overlapping but not identical range of cell lines, suggesting that SFTSV, LACV, and RVFV might differ in their receptor requirements. Entry driven by SFTSV Gn/Gc was dependent on low pH but did not require the activity of the pH-dependent endosomal/lysosomal cysteine proteases cathepsins B and L. Instead, the activity of a cellular serine protease was required for infection driven by SFTSV and LACV Gn/Gc. Sera from convalescent SFTS patients inhibited SFTSV Gn/Gc-driven host cell entry in a dose-dependent fashion, demonstrating that the vector system employed is suitable to detect neutralizing antibodies. Finally, the C-type lectin DC-SIGN was found to serve as a receptor for SFTSV Gn/Gc-driven entry into cell lines and dendritic cells. Our results provide initial insights into cell tropism, receptor usage, and proteolytic activation of SFTSV and will aid in the understanding of viral spread and pathogenesis.

The spike protein of the emerging betacoronavirus EMC uses a novel coronavirus receptor for entry, can be activated by TMPRSS2, and is targeted by neutralizing antibodies.

The novel human coronavirus EMC (hCoV-EMC), which recently emerged in Saudi Arabia, is highly pathogenic and could pose a significant threat to public health. The elucidation of hCoV-EMC interactions with host cells is critical to our understanding of the pathogenesis of this virus and to the identification of targets for antiviral intervention. Here we investigated the viral and cellular determinants governing hCoV-EMC entry into host cells. We found that the spike protein of hCoV-EMC (EMC-S) is incorporated into lentiviral particles and mediates transduction of human cell lines derived from different organs, including the lungs, kidneys, and colon, as well as primary human macrophages. Expression of the known coronavirus receptors ACE2, CD13, and CEACAM1 did not facilitate EMC-S-driven transduction, suggesting that hCoV-EMC uses a novel receptor for entry. Directed protease expression and inhibition analyses revealed that TMPRSS2 and endosomal cathepsins activate EMC-S for virus-cell fusion and constitute potential targets for antiviral intervention. Finally, EMC-S-driven transduction was abrogated by serum from an hCoV-EMC-infected patient, indicating that EMC-S-specific neutralizing antibodies can be generated in patients. Collectively, our results indicate that hCoV-EMC uses a novel receptor for protease-activated entry into human cells and might be capable of extrapulmonary spread. In addition, they define TMPRSS2 and cathepsins B and L as potential targets for intervention and suggest that neutralizing antibodies contribute to the control of hCoV-EMC infection.

Impact of ETIF deletion on safety and immunogenicity of equine herpesvirus type 1-vectored vaccines.

Heterologous gene transfer by viral vector systems is often limited by factors such as preexisting immunity, toxicity, low packaging capacity, or weak immunogenic potential. A novel viral vector system derived from equine herpesvirus type 1 (EHV-1) not only overcomes some of these obstacles but also promotes the robust expression of a delivered transgene and the induction of antigen-specific immune responses. Regarding an enhanced safety profile, we assessed the impact of the gene encoding the sole essential tegument protein, ETIF, on the replication and immunogenicity of recombinant EHVs. The deletion of ETIF severely attenuates replication in permissive RK13 cells and a human lung epithelial cell line but without influencing transgene expression. Whereas the intranasal administration of a recombinant luciferase EHV in BALB/c mice resulted in transgene expression in nasal cavities and lungs for 5 to 6 days, the ETIF deletion limited expression to 2 days and resulted in 30-fold-less luminescence. Attenuated replication was accompanied by a decreased capacity to induce CD8(+) T cells against a delivered HIV Gag transgene in BALB/c mice following repeated intranasal application. However, a single subcutaneous immunization with a gag DNA vaccine primed specific T cells for substantial expansion by two subsequent intranasal booster immunizations with either the gag recombinant ETIF mutant or the parental virus. In addition to inducing Gag-specific serum antibodies, this prime-boost strategy clearly outperformed three sequential immunizations with the parental or EHV-ΔETIF virus or repeated DNA vaccination by inducing substantial specific secretory IgA (sIgA) titers.

Differential downregulation of ACE2 by the spike proteins of severe acute respiratory syndrome coronavirus and human coronavirus NL63.

The human coronaviruses (CoVs) severe acute respiratory syndrome (SARS)-CoV and NL63 employ angiotensin-converting enzyme 2 (ACE2) for cell entry. It was shown that recombinant SARS-CoV spike protein (SARS-S) downregulates ACE2 expression and thereby promotes lung injury. Whether NL63-S exerts a similar activity is yet unknown. We found that recombinant SARS-S bound to ACE2 and induced ACE2 shedding with higher efficiency than NL63-S. Shedding most likely accounted for the previously observed ACE2 downregulation but was dispensable for viral replication. Finally, SARS-CoV but not NL63 replicated efficiently in ACE2-positive Vero cells and reduced ACE2 expression, indicating robust receptor interference in the context of SARS-CoV but not NL63 infection.

The soy isoflavones for reducing bone loss study: 3-yr effects on pQCT bone mineral density and strength measures in postmenopausal women.

Soy isoflavones exert inconsistent bone density-preserving effects, but the bone strength-preserving effects in humans are unknown. Our double-blind randomized controlled trial examined 2 soy isoflavone doses (80 or 120mg/d) vs placebo tablets on volumetric bone mineral density (vBMD) and strength (by means of peripheral quantitative computed tomography) in healthy postmenopausal women (46-63yr). We measured 3-yr changes in cortical BMD (CtBMD), cortical thickness (CtThk), periosteal circumference (PC), endosteal circumference (EC), and strength-strain index (SSI) at 1/3 midshaft femur (N=171), and trabecular BMD (TbBMD), PC, and SSI at 4% distal tibia (N=162). We found no treatment effect on femur CtThk, PC, or EC, or tibia TbBMD or PC. The strongest predictors (negative) of tibia TbBMD and SSI and femur CtBMD were timepoint and bone resorption; whole-body fat mass was protective of SSI. As time since last menstrual period (TLMP) increased (p=0.012), 120-mg/d dose was protective of CtBMD. The strongest predictors of femur SSI were timepoint, bone resorption, and TLMP (protective). Isoflavone tablets were negative predictors of SSI, but 80-mg/d dose became protective as bone turnover increased (p=0.011). Soy isoflavone treatment for 3yr was modestly beneficial for midshaft femur vBMD as TLMP increased and for midshaft femur SSI as bone turnover increased.

Clustering microarray data to determine normalization method.

Most of the scientific journals require published microarray experiments to meet Minimum Information About a Microarray Experiment (MIAME) standards. This ensures that other researchers have the necessary information to interpret the results or reproduce them. Required MIAME information includes raw experimental data, processed data, and data processing procedures. However, the normalization method is often reported inaccurately or not at all. It may be that the scaling factor is not even known except to experienced users of the normalization software. We propose that using a seeded clustering algorithm, researchers can identify or verify previously unknown or doubtful normalization information. For that, we generate descriptive statistics (mean, variance, quantiles, and moments) for normalized expression data from gene chip experiments available in the ArrayExpress database and cluster chips based on these statistics. To verify that clustering grouped chips by normalization method, we normalize raw data for chips chosen from experiments in ArrayExpress using multiple methods. We then generate the same descriptive statistics for the normalized data and cluster the chips using these statistics. We use this dataset of known pedigree as seeding data to identify normalization methods used in unknown or doubtful situations.

Interactive Web-based Data Visualization and Analysis Tool for Synthetizing on-farm Research Networks Data.

The on-farm research network concept enables a group of farmers to test new agricultural management practices under local conditions with support from local researchers or agronomists. Different on-farm trials based on the same experimental design are conducted over several years and sites to test the effectiveness of different innovative management practices aimed at increasing crop productivity and profitability. As a larger amount of historical trial data are being accumulated, data of all the trials require analyses and summarization. Summaries of on-farm trials are usually presented to farmers as individual field reports, which are not optimal for the dissemination of results and decision making. A more practical communication method is needed to enhance result communication and decision making. R Shiny is a new rapidly developing technology for turning R data analyses into interactive web applications. For the first time for on-farm research networks, we developed and launched an interactive web tool called ISOFAST using R Shiny. ISOFAST simultaneously reports all trial results about the same management practice to simplify interpretation of multi-site and multi-year summaries. We used a random-effects model to synthetize treatment differences at both the individual trial and network levels and generate new knowledge for farmers and agronomists. The friendly interface enables users to explore trial summaries, access model outputs, and perform economic analysis at their fingertips. This paper describes a case-study to illustrate how to use the tool and make agronomic management decisions based on the on-farm trial data. We also provided technical details and guidance for developing a similar interactive visualization tool customized for on-farm research network. ISOFAST is currently available at https://analytics.iasoybeans.com/cool-apps/ISOFAST/.

Graphical inference for Infovis.

How do we know if what we see is really there? When visualizing data, how do we avoid falling into the trap of apophenia where we see patterns in random noise? Traditionally, infovis has been concerned with discovering new relationships, and statistics with preventing spurious relationships from being reported. We pull these opposing poles closer with two new techniques for rigorous statistical inference of visual discoveries. The "Rorschach" helps the analyst calibrate their understanding of uncertainty and "line-up" provides a protocol for assessing the significance of visual discoveries, protecting against the discovery of spurious structure.

A Robust Approach to Automatically Locating Grooves in 3D Bullet Land Scans.

Land engraved areas (LEAs) provide evidence to address the same source-different source problem in forensic firearms examination. Collecting 3D images of bullet LEAs requires capturing portions of the neighboring groove engraved areas (GEAs). Analyzing LEA and GEA data separately is imperative to accuracy in automated comparison methods such as the one developed by Hare et al. (Ann Appl Stat 2017;11, 2332). Existing standard statistical modeling techniques often fail to adequately separate LEA and GEA data due to the atypical structure of 3D bullet data. We developed a method for automated removal of GEA data based on robust locally weighted regression (LOESS). This automated method was tested on high-resolution 3D scans of LEAs from two bullet test sets with a total of 622 LEA scans. Our robust LOESS method outperforms a previously proposed "rollapply" method. We conclude that our method is a major improvement upon rollapply, but that further validation needs to be conducted before the method can be applied in a fully automated fashion.

Comparison of three similarity scores for bullet LEA matching.

Recent advances in microscopy have made it possible to collect 3D topographic data, enabling more precise virtual comparisons based on the collected 3D data as a supplement to traditional comparison microscopy and 2D photography. Automatic comparison algorithms have been introduced for various scenarios, such as matching cartridge cases [1,2] or matching bullet striae [3-5]. One key aspect of validating these automatic comparison algorithms is to evaluate the performance of the algorithm on external tests, that is, using data which were not used to train the algorithm. Here, we present a discussion of the performance of the matching algorithm [6] in three studies conducted using different Ruger weapons. We consider the performance of three scoring measures: random forest score, cross correlation, and consecutive matching striae (CMS) at the land-to-land level and, using Sequential Average Maxima scores, also at the bullet-to bullet level. Cross correlation and random forest scores both result in perfect discrimination of same-source and different-source bullets. At the land-to-land level, discrimination for both cross correlation and random forest scores (based on area under the curve, AUC) is excellent (≥0.90).

Adapting the Chumbley Score to Match Striae on Land Engraved Areas (LEAs) of Bullets.

The same-source problem remains a major challenge in forensic toolmark and firearm examination. Here, we investigate the applicability of the Chumbley method (J Forensic Sci, 2018, 63, 849; J Forensic Sci, 2010, 55, 953) (10,12), developed for screwdriver markings, for same-source identification of striations on bullet LEAs. The Hamby datasets 44 and 252 measured by NIST and CSAFE (high-resolution scans) are used here. We provide methods to identify parameters that minimize error rates for matching of LEAs, and a remedial algorithm to alleviate the problem of failed tests, while increasing the power of the test and reducing error rates. For 85,491 land-to-land comparisons (84,235 known nonmatches and 1256 known matches), the adapted test does not provide a result in 176 situations (originally more than 500). The Type I and Type II error rates are 7.2% (6105 out of 84,235) and 21.4% (271 out of 1256), respectively. This puts the proposed method on similar footing as other single-feature matching approaches in the literature.

Prediction of side chain orientations in proteins by statistical machine learning methods.

We develop ways to predict the side chain orientations of residues within a protein structure by using several different statistical machine learning methods. Here side chain orientation of a given residue i is measured by an angle Omega(i) between the vector pointing from the center of the protein structure to the C(i)(alpha) atom and the vector pointing from the C(i)(alpha) atom to the center of its side chain atoms. To predict the Omega(i) angles, we construct statistical models by using several different methods such as general linear regression, a regression tree and bagging, a neural network, and a support vector machine. The root mean square errors for the different models range only from 36.67 to 37.60 degrees and the correlation coefficients are all between 30% and 34%. The performances of different models in the test set are, thus, quite similar, and show the relative predictive power of these models to be significant in comparison with random side chain orientations.

Spatial Reasoning and Data Displays.

Graphics convey numerical information very efficiently, but rely on a different set of mental processes than tabular displays. Here, we present a study relating demographic characteristics and visual skills to perception of graphical lineups. We conclude that lineups are essentially a classification test in a visual domain, and that performance on the lineup protocol is associated with general aptitude, rather than specific tasks such as card rotation and spatial manipulation. We also examine the possibility that specific graphical tasks may be associated with certain visual skills and conclude that more research is necessary to understand which visual skills are required in order to understand certain plot types.

Biomathematical description of synthetic peptide libraries.

Libraries of randomised peptides displayed on phages or viral particles are essential tools in a wide spectrum of applications. However, there is only limited understanding of a library's fundamental dynamics and the influences of encoding schemes and sizes on their quality. Numeric properties of libraries, such as the expected number of different peptides and the library's coverage, have long been in use as measures of a library's quality. Here, we present a graphical framework of these measures together with a library's relative efficiency to help to describe libraries in enough detail for researchers to plan new experiments in a more informed manner. In particular, these values allow us to answer-in a probabilistic fashion-the question of whether a specific library does indeed contain one of the "best" possible peptides. The framework is implemented in a web-interface based on two packages, discreteRV and peptider, to the statistical software environment R. We further provide a user-friendly web-interface called PeLiCa (Peptide Library Calculator, http://www.pelica.org), allowing scientists to plan and analyse their peptide libraries.

Soy Isoflavones for Reducing Bone Loss Study: effects of a 3-year trial on hormones, adverse events, and endometrial thickness in postmenopausal women.

OBJECTIVE: This study aims to assess the overall safety and potential endometrium-stimulating effects of soy isoflavone tablets consumed (3 y) by postmenopausal women and to determine endometrial thickness response to treatment among compliant women, taking into account hormone concentrations and other hypothesized modifying factors. METHODS: We randomized healthy postmenopausal women (aged 45.8-65.0 y) to placebo control or two doses (80 or 120 mg/d) of soy isoflavones at two sites. We used intent-to-treat analysis (N = 224) and compliant analysis (>95%; N = 208) to assess circulating hormone concentrations, adverse events, and endometrial thickness (via transvaginal ultrasound). RESULTS: Median values for endometrial thickness (mm) declined from baseline through 36 months. Nonparametric analysis of variance for treatment differences among groups showed no differences in absolute (or percentage of change) endometrial thickness (χ(2) P ranged from 0.12 to 0.69) or in circulating hormones at any time point. A greater number of adverse events in the genitourinary system (P = 0.005) were noted in the 80 mg/day group compared with the 120 mg/day group, whereas other systems showed no treatment effects. The model predicting endometrial thickness response (using natural logarithm) to treatment among compliant women across time points was significant (P ≤ 0.0001), indicating that estrogen exposure (P = 0.0013), plasma 17ß-estradiol (P = 0.0086), and alcohol intake (P = 0.023) contributed significantly to the response. Neither the 80 mg/day dose (P = 0.57) nor the 120 mg/day dose (P = 0.43) exerted an effect on endometrial thickness across time. CONCLUSIONS: Our randomized controlled trial verifies the long-term overall safety of soy isoflavone tablet intake by postmenopausal women who display excellent compliance. We find no evidence of treatment effects on endometrial thickness, adverse events, or circulating hormone concentrations, most notably thyroid function, across a 3-year period.

Common angle plots as perception-true visualizations of categorical associations.

Visualizations are great tools of communications-they summarize findings and quickly convey main messages to our audience. As designers of charts we have to make sure that information is shown with a minimum of distortion. We have to also consider illusions and other perceptual limitations of our audience. In this paper we discuss the effect and strength of the line width illusion, a Muller-Lyer type illusion, on designs related to displaying associations between categorical variables. Parallel sets and hammock plots are both affected by line width illusions. We introduce the common-angle plot as an alternative method for displaying categorical data in a manner that minimizes the effect from perceptual illusions. Results from user studies both highlight the need for addressing line-width illusions in displays and provide evidence that common angle charts successfully resolve this issue.

Cathepsins B and L activate Ebola but not Marburg virus glycoproteins for efficient entry into cell lines and macrophages independent of TMPRSS2 expression.

Ebola (EBOV) and Marburg virus (MARV) cause severe hemorrhagic fever. The host cell proteases cathepsin B and L activate the Zaire ebolavirus glycoprotein (GP) for cellular entry and constitute potential targets for antiviral intervention. However, it is unclear if different EBOV species and MARV equally depend on cathepsin B/L activity for infection of cell lines and macrophages, important viral target cells. Here, we show that cathepsin B/L inhibitors markedly reduce 293T cell infection driven by the GPs of all EBOV species, independent of the type II transmembrane serine protease TMPRSS2, which cleaved but failed to activate EBOV-GPs. Similarly, a cathepsin B/L inhibitor blocked macrophage infection mediated by different EBOV-GPs. In contrast, MARV-GP-driven entry exhibited little dependence on cathepsin B/L activity. Still, MARV-GP-mediated entry was efficiently blocked by leupeptin. These results suggest that cathepsins B/L promote entry of EBOV while MARV might employ so far unidentified proteases for GP activation.