Discrimination ability of prediction models for ordinal outcomes: Relationships between existing measures and a new measure

In this paper, we focus on measures to evaluate discrimination of prediction models for ordinal outcomes. We review existing extensions of the dichotomous c‐index—which is equivalent to the area under the receiver operating characteristic (ROC) curve—suggest a new measure, and study their relationships. The volume under the ROC surface (VUS) scores sets of cases including one case from each outcome category. VUS considers sets as either correctly or incorrectly ordered by the model. All other existing measures assess pairs of cases. We propose an ordinal c‐index (ORC) that is set‐based but, contrary to VUS, scores sets more gradually by indicating the closeness of the model‐based ordering to the perfect ordering. As a result, the ORC does not decrease rapidly as the number of outcome categories increases. It turns out that the ORC can be rewritten as the average of pairwise c‐indexes. Hence, the ORC has both a set‐ and pair‐based interpretation. There are several relationships between the existing measures, leading to only two types of existing measures: a prevalence‐weighted average of pairwise c‐indexes and the VUS. Our suggested measure ORC positions itself in between as it is set‐based but turns out to equal an unweighted average of pairwise c‐indexes. The measures are demonstrated through a case study on the prediction of six‐month outcome after traumatic brain injury. In conclusion, the set‐based nature and graded scoring system make the ORC an attractive measure with a simple interpretation, together with its prevalence‐independence that is a natural property of a discrimination measure.     

Numbats and aardwolves—how low is low? A re-affirmation of the need for statistical rigour in evaluating regression predictions

Many comparative physiological studies aim to determine if a particular species differs from a prediction based on a linear allometric regression for other species. However, the judgment as to whether the species in question conforms to this allometric relationship is often not based on any formal statistical analysis. An appropriate statistical method is to compare the new species’ value with the 95% confidence limits for predicting an additional datum from the relationship for the other species. We examine the basal metabolic rate (BMR) of the termitivorous numbat (Myrmecobius fasciatus) and aardwolf (Proteles cristatus) to demonstrate the use of the 95% prediction limits to determine statistically if they have a lower-than-expected BMR compared to related species. The numbat’s BMR was 83.6% of expected from mass, but fell inside the 95% prediction limits for a further datum; a BMR < 72.5% of predicted was required to fall below the one-tail 95% prediction limits. The aardwolf had a BMR that was only 74.2% of predicted from the allometric equation, but it also fell well within the 95% prediction limits; a BMR of only 41.8% of predicted was necessary to fall below the one-tail 95% prediction limits. We conclude that a formal statistical approach is essential, although it is difficult to demonstrate that a single species statistically differs from a regression relationship for other species.     

Characterization,antimicrobial, antioxidant,and anticoagulant activities of silver nanoparticles synthesized from Petiveria alliacea L. leaf extract

Abstract

Phytosynthesis of silver nanoparticles (AgNPs) using leaf extract of Petiveria alliacea (PA) was the focus of this research work. The PA-AgNPs were characterized by UV–Vis spectroscopy, Fourier transform infrared spectroscopy (FTIR), transmission electron microscopy (TEM), and selected area electron diffraction (SAED) study. Studies were made on the AgNPs for antibacterial, antifungal, anticoagulant, free-radical scavenging, and hydrogen peroxide scavenging activities. The crystalline PA-AgNPs were monodispersed, with a size range of 16.70–33.74?nm and maximum absorption at 410?nm. FTIR analysis displayed prominent peaks at 3430.6, 1711.8, and 1165.9/cm, which showed the existence of phenolic compounds and proteins in the synthesis of AgNPs. PA-AgNPs was active against Escherichia coli, Klebsiella pneumoniae, and Staphylococcus aureus, with 100% inhibition. The PA-AgNPs also displayed good antifungal properties, as the concentrations of 100 and 150?µg/mL had 100% inhibition toward Aspergillus fumigatus and Aspergillus flavus. However, there was 66.67% inhibition of Aspergillus niger. It scavenged both DPPH and H₂O₂ by 70.69 and 89.02%, respectively. PA-AgNPs also prevented the coagulation of human blood. This study, being the first of its kind to use the leaf extract of PA for the synthesis of AgNPs has shown that PA-AgNPs can find biomedical applications.     

Increased chemical resistance explains low herbivore colonization of introduced seaweed

The success of introduced species is often attributed to release from co-evolved enemies in the new range and a subsequent decreased allocation to defense (EICA), but these hypotheses have rarely been evaluated for systems with low host-specificity of enemies. Here, we compare herbivore utilization of the brown seaweed, Fucus evanescens, and its coexisting competitors both in its native and new ranges, to test certain predictions derived from these hypotheses in a system dominated by generalist herbivores. While F. evanescens was shown to be a preferred host in its native range, invading populations supported a less diverse herbivore fauna and it was less preferred in laboratory choice experiments with important herbivores, when compared to co-occurring seaweeds. These results are consistent with the enemy release hypothesis, despite the fact that the herbivore communities in both regions were mainly composed of generalist species. However, in contrast to the prediction of EICA, analysis of anti-grazing compounds indicated a higher allocation to defense in introduced compared to native F. evanescens. The results suggest that the invader is subjected to less intense enemy control in the new range, but that this is due to an increased allocation to defense rather than release from specialized herbivores. This indicates that increased resistance to herbivory might be an important strategy for invasion success in systems dominated by generalist herbivores.     

Monte Carlo simulation studies on the prediction of protein folding types from amino acid composition

A number of methods to predicting the folding type of a protein based on its amino acid composition have been developed during the past few years. In order to perform an objective and fair comparison of different prediction methods, a Monte Carlo simulation method was proposed to calculate the asymptotic limit of the prediction accuracy [Zhang and Chou (1992),Biophys. J. 63, 1523–1529, referred to as simulation method I]. However, simulation method I was based on an oversimplified assumption, i.e., there are no correlations between the compositions of different amino acids. By taking into account such correlations, a new method, referred to as simulation method II, has been proposed to recalculate the objective accuracy of prediction for the least Euclidean distance method [Nakashimaet al. (1986),J. Biochem. 99, 152–162] and the least Minkowski distance method [Chou (1989),Prediction in Protein Structure and the Principles of Protein Conformation, Plenum Press, New York, pp. 549–586], respectively. The results show that the prediction accuracy of the former is still better than that of the latter, as found by simulation method I; however, after incorporating the correlative effect, the objective prediction accuracies become lower for both methods. The reason for this phenomenon is discussed in detail. The simulation method and the idea developed in this paper can be applied to examine any other statistical prediction method, including the computersimulated neural network method.     

The replication map of the chromosome ofMycobacterium phlei

It was the aim of the present work to construct the replication map of the chromosome ofMycobacterium phlei. The method of mutagenesis of the replication point by N-methyl-N-nitroso-N’-nitroguanidine in synchronously dividing populations and the method of analysis of gene frequency were applied. The order of replication of 19 genes on the chromosome was determined by means of induction of back mutations and forward mutations in auxotrophic mutants PAleu and PAmet and in double auxotrophic mutants with methionine as a reference marker.     

Predator-recognition Behaviour in Two Species of Gobiid Fishes,Asterropteryx semipunctatus and Gnatholepis anjerensis

When Asterropteryx semipunctatus and Gnatholepis anjerensis detect a predator they bob rhythmically, raising and lowering the body. Bobbing and reduced movement was transmitted visually between fish in adjacent tanks, both within and between species. Fish isolated from other fish for at least 5 days still bobbed when they encountered a predator model. G. anjerensis remained resident in one location for an average of 61 days from marking to disappearance or the end of the study. They could, therefore, benefit if bobbing reduced predator hunting success. Three general categories of hypotheses have been proposed to account for predator-recognition behaviour: assessment of the predator, informing the predator that it is detected, and warning the sender's conspecifics. Two of these, informing the predator and warning, seem most likely to account for goby bobbing. These hypotheses are not mutually exclusive.     

Inheritance of porcine receptors for enterotoxigenic Escherichia coli with fimbriae F4ad and their relation to other F4 receptors

Enteric Escherichia coli infections are a highly relevant cause of disease and death in young pigs. Breeding genetically resistant pigs is an economical and sustainable method of prevention. Resistant pigs are protected against colonization of the intestine through the absence of receptors for the bacterial fimbriae, which mediate adhesion to the intestinal surface. The present work aimed at elucidation of the mode of inheritance of the F4ad receptor which according to former investigations appeared quite confusing. Intestines of 489 pigs of an experimental herd were examined by a microscopic adhesion test modified in such a manner that four small intestinal sites instead of one were tested for adhesion of the fimbrial variant F4ad. Segregation analysis revealed that the mixed inheritance model explained our data best. The heritability of the F4ad phenotype was estimated to be 0.7±0.1. There are no relations to the strong receptors for variants F4ab and F4ac. Targeted matings allowed the discrimination between two F4ad receptors, that is, a fully adhesive receptor (F4adR^FA) expressed on all enterocytes and at all small intestinal sites, and a partially adhesive receptor (F4adR^PA) variably expressed at different sites and often leading to partial bacterial adhesion. In pigs with both F4ad receptors, the F4adR^PA receptor is masked by the F4adR^FA. The hypothesis that F4adR^FA must be encoded by at least two complementary or epistatic dominant genes is supported by the Hardy–Weinberg equilibrium statistics. The F4adR^PA receptor is inherited as a monogenetic dominant trait. A comparable partially adhesive receptor for variant F4ab (F4abR^PA) was also observed but the limited data did not allow a prediction of the mode of inheritance. Pigs were therefore classified into one of eight receptor phenotypes: A1 (F4abR^FA/F4acR⁺/F4adR^FA); A2 (F4abR^FA/F4acR⁺/F4adR^PA); B (F4abR^FA/F4acR⁺/F4adR⁻); C1 (F4abR^PA/F4acR⁻/F4adR^FA); C2 (F4abR^PA/F4acR⁻/F4adR^PA); D1 (F4abR⁻/F4acR⁻/F4adR^FA); D2 (F4abR⁻/F4acR⁻/F4adR^PA); E (F4abR⁻/F4acR⁻/F4adR⁻).     

Histochemical procedures for the simultaneous visualization of neutral sugars and either sialic acid and its side chain O-acyl variants or O-sulphate ester. I. Methods based upon the selective periodate oxidation of sialic acids

Summary Five new methods, based upon the selective oxidation of sialic acid residues with 0.4mm periodic acid in approximately 1m hydrochloric acid at 4°C for 1 h (PA^*), have been devised for the simultaneous visualization of neutral sugars and either sialic acid and its side chainO-acyl variants orO-sulphate ester. In the first of these, the selective periodate oxidation—borohydride reduction—saponification—selective periodate oxidation—Thionin Schiff—saponification—borohydride reduction—periodic acid—Schiff (PA^*—Bh—KOH—PA^*—T—KOH—Bh—PAS) technique, sialic acids withO-acyl substituents at C7, C8 or C9 (or which have two of three side chainO-acyl substituents) stain blue while neutral sugars with periodate-sensitivevicinal diols (hexose, 6-deoxyhexose, andN-acetylhexosamine) stain magenta. The second method, the saponification—selective periodate oxidation—Thionin Schiff—saponification—borohydride reduction—periodic acid—Schiff (KOH—PA^*—T—KOH—Bh—PAS), stains all sialic acids blue and neutral sugars magenta. In the third procedure, the selective periodate oxidation—Thionin Schiff—borohydride reduction—periodic acid—Schiff—saponification (PA^*—T—Bh—PAS—KOH) method, sialic acids without side chain substituents (or which have anO-acyl substituent at C7) stain blue and neutral sugars stain magenta. In the fourth method, the saponification-selective periodate oxidation—borohydride reduction—Alcian Blue pH 1.0—periodic acid—Schiff (KOH—PA^*—Bh—AB1.0—PAS) technique,O-sulphate esters stain aquamarine blue and neutral sugars stain magenta. In all of these techniques mixtures of the components stain in various shades of purple. Performance of the KOH—PA^*—Bh—AB1.0—PAS technique without the Alcian Blue pH 1.0 step provides a method for the selective identification of neutral sugars in macromolecules that also contain sialic acids.     

Mapping of the chromosome ofMycobacterium phlei by means of mutagenesis of the replication point

The aim of the present work was to construct a replication map of the chromosome ofMycobacterium phlei. The method of mutagenesis of the replication point by means of nitrosoguanidine was applied to synchronously multiplying populations. Back mutations and forward mutations were induced m auxotrophic mutants PAmet and PAleu as well as in double auxotrophic mutants with methlomne as the reference marker and the following order of replication of eleven genes on the chromosome was thus established:leu-Eth, Res-Stm, Oyk-pur-met, arg, Cyk-Bac-inl     

Prediction of protein domain boundaries from inverse covariances

It has been known even since relatively few structures had been solved that longer protein chains often contain multiple domains, which may fold separately and play the role of reusable functional modules found in many contexts. In many structural biology tasks, in particular structure prediction, it is of great use to be able to identify domains within the structure and analyze these regions separately. However, when using sequence data alone this task has proven exceptionally difficult, with relatively little improvement over the naive method of choosing boundaries based on size distributions of observed domains. The recent significant improvement in contact prediction provides a new source of information for domain prediction. We test several methods for using this information including a kernel smoothing‐based approach and methods based on building alpha‐carbon models and compare performance with a length‐based predictor, a homology search method and four published sequence‐based predictors: DOMCUT, DomPRO, DLP‐SVM, and SCOOBY‐DOmain. We show that the kernel‐smoothing method is significantly better than the other ab initio predictors when both single‐domain and multidomain targets are considered and is not significantly different to the homology‐based method. Considering only multidomain targets the kernel‐smoothing method outperforms all of the published methods except DLP‐SVM. The kernel smoothing method therefore represents a potentially useful improvement to ab initio domain prediction. Proteins 2013. © 2012 Wiley Periodicals, Inc.     

Notes on Estimation of the General Odds Ratio and the General Risk Difference for Paired‐Sample Data

Under the matched‐pair design, this paper discusses estimation of the general odds ratio OR_G for ordinal exposure in case‐control studies and the general risk difference RD_G for ordinal outcomes in cross‐sectional or cohort studies. To illustrate the practical usefulness of interval estimators of OR_G and RD_G developed here, this paper uses the data from a case‐control study investigating the effect of the number of beverages drunk at “burning hot” temperature on the risk of possessing esophageal cancer, and the data from a cross‐sectional study comparing the grade distributions of unaided distance vision between two eyes. Finally, this paper notes that using the commonly‐used statistics related to odds ratio for dichotomous data by collapsing the ordinal exposure into two categories: the exposure versus the non‐exposure, tends to be less efficient than using the statistics related to OR_G proposed herein.     

Timing and location of reproduction in African waterfowl: an overview of >100 years of nest records

The timing and location of reproduction are fundamental elements of reproductive success for all organisms. Understanding why animals choose to reproduce at particular times and in particular places is also important for our understanding of other aspects of organismal ecology, such as their habitat requirements, movement strategies, and biogeography. Although breeding patterns in waterfowl are relatively well documented, most studies are from northern temperate regions and the influences of location and time of year on breeding in Afrotropical ducks (Anatidae) are poorly understood. We outline six alternative (but not mutually exclusive) hypotheses that might explain where and when Afrotropical ducks choose to breed. To explore these hypotheses, we assembled and analyzed a new database of c. 22,000 breeding records for 16 Afrotropical ducks and one introduced Palearctic species (the Mallard Anas platyrhynchos). The full database is available on line as an appendix to this article. We identified five distinct breeding strategies as well as two outliers. Peak breeding for 9 of 16 indigenous duck species occurs during the dry season. We found no evidence for spatial synchrony or spatial autocorrelation in breeding, suggesting a high level of flexibility in waterfowl responses to prevailing conditions in any given year. More intensive analyses of alternative hypotheses are needed, but our initial analysis suggests that the timing of breeding for the majority of Afrotropical ducks is driven by a combination of resource availability and predation risk.     

Pseudomonas aeruginosa cytochrome c551 denaturation by five systematic urea derivatives that differ in the alkyl chain length

Reversible denaturation of Pseudomonas aeruginosa cytochrome c₅₅₁ (PAc₅₅₁) could be followed using five systematic urea derivatives that differ in the alkyl chain length, i.e. urea, N-methylurea (MU), N-ethylurea (EU), N-propylurea (PU), and N-butylurea (BU). The BU concentration was the lowest required for the PAc₅₅₁ denaturation, those of PU, EU, MU, and urea being gradually higher. Furthermore, the accessible surface area difference upon PAc₅₅₁ denaturation caused by BU was found to be the highest, those by PU, EU, MU, and urea being gradually lower. These findings indicate that urea derivatives with longer alkyl chains are stronger denaturants. In this study, as many as five systematic urea derivatives could be applied for the reversible denaturation of a single protein, PAc₅₅₁, for the first time, and the effects of the alkyl chain length on protein denaturation were systematically verified by means of thermodynamic parameters.     

NetPhosBac – A predictor for Ser/Thr phosphorylation sites in bacterial proteins

There is ample evidence for the involvement of protein phosphorylation on serine/threonine/tyrosine in bacterial signaling and regulation, but very few exact phosphorylation sites have been experimentally determined. Recently, gel‐free high accuracy MS studies reported over 150 phosphorylation sites in two bacterial model organisms Bacillus subtilis and Escherichia coli. Interestingly, the analysis of these phosphorylation sites revealed that most of them are not characteristic for eukaryotic‐type protein kinases, which explains the poor performance of eukaryotic data‐trained phosphorylation predictors on bacterial systems. We used these large bacterial datasets and neural network algorithms to create the first bacteria‐specific protein phosphorylation predictor: NetPhosBac. With respect to predicting bacterial phosphorylation sites, NetPhosBac significantly outperformed all benchmark predictors. Moreover, NetPhosBac predictions of phosphorylation sites in E. coli proteins were experimentally verified on protein and site‐specific levels. In conclusion, NetPhosBac clearly illustrates the advantage of taxa‐specific predictors and we hope it will provide a useful asset to the microbiological community.     

Applying Canonical Analysis to a Kind of Generalized Regression

This paper is concerned with the generalized model E(φ(Y)! X) =φ(X) involving the transformations on both the predictor vector X and the response variable Y. For this purpose, Taylor expansions and canonical analysis are applied. For optimizing the expansions, it is shown by a simulation study that not only prediction error, the combination of model error and noise error, is an important index, but the distribution of the residuals and the t-values of the coefficients also must be considered. Furthermore, the results of penicillin titrition show that the practical situations often need to be considered in selecting an appropriate model for a real-life problem.     

A Hybrid Deep Learning Model for Predicting Molecular Subtypes of Human Breast Cancer Using Multimodal Data

BackgroundThe prediction of breast cancer subtypes plays a key role in the diagnosis and prognosis of breast cancer. In recent years, deep learning (DL) has shown good performance in the intelligent prediction of breast cancer subtypes. However, most of the traditional DL models use single modality data, which can just extract a few features, so it cannot establish a stable relationship between patient characteristics and breast cancer subtypes.DatasetWe used the TCGA-BRCA dataset as a sample set for molecular subtype prediction of breast cancer. It is a public dataset that can be obtained through the following link: https://portal.gdc.cancer.gov/projects/TCGA-BRCAMethodsIn this paper, a Hybrid DL model based on the multimodal data is proposed. We combine the patient's gene modality data with image modality data to construct a multimodal fusion framework. According to the different forms and states, we set up feature extraction networks respectively, and then we fuse the output of the two feature networks based on the idea of weighted linear aggregation. Finally, the fused features are used to predict breast cancer subtypes. In particular, we use the principal component analysis to reduce the dimensionality of high-dimensional data of gene modality and filter the data of image modality. Besides, we also improve the traditional feature extraction network to make it show better performance.ResultsThe results show that compared with the traditional DL model, the Hybrid DL model proposed in this paper is more accurate and efficient in predicting breast cancer subtypes. Our model achieved a prediction accuracy of 88.07% in 10 times of 10-fold cross-validation. We did a separate AUC test for each subtype, and the average AUC value obtained was 0.9427. In terms of subtype prediction accuracy, our model is about 7.45% higher than the previous average.