Two-stage Testing Research Articles

Efficiently Identifying Significant Associations in Genome-wide Association Studies

Over the past several years, genome-wide association studies (GWAS) have implicated hundreds of genes in common disease. More recently, the GWAS approach has been utilized to identify regions of the genome that harbor variation affecting gene expression or expression quantitative trait loci (eQTLs). Unlike GWAS applied to clinical traits, where only a handful of phenotypes are analyzed per study, in eQTL studies, tens of thousands of gene expression levels are measured, and the GWAS approach is applied to each gene expression level. This leads to computing billions of statistical tests and requires substantial computational resources, particularly when applying novel statistical methods such as mixed models. We introduce a novel two-stage testing procedure that identifies all of the significant associations more efficiently than testing all the single nucleotide polymorphisms (SNPs). In the first stage, a small number of informative SNPs, or proxies, across the genome are tested. Based on their observed associations, our approach locates the regions that may contain significant SNPs and only tests additional SNPs from those regions. We show through simulations and analysis of real GWAS datasets that the proposed two-stage procedure increases the computational speed by a factor of 10. Additionally, efficient implementation of our software increases the computational speed relative to the state-of-the-art testing approaches by a factor of 75.

Efficient Two-Stage Group Testing Algorithms for Genetic Screening

Efficient two-stage group testing algorithms that are particularly suited for rapid and less-expensive DNA library screening and other large scale biological group testing efforts are investigated in this paper. The main focus is on novel combinatorial constructions in order to minimize the number of individual tests at the second stage of a two-stage disjunctive testing procedure. Building on recent work by Levenshtein (2003) and Tonchev (2008), several new infinite classes of such combinatorial designs are presented.

Combining the formative with the summative: the development of a twostage online test to encourage engagement and provide personal feedback in large classes

The aim of this action research project was to improve student learning by encouraging more “time on task” and to improve self-assessment and feedback through the introduction of weekly online tests in a Year 2 lecture module in biological sciences. Initially voluntary online tests were offered to students and those who participated achieved higher exam marks than those who did not, but completion rate was low. Making the tests compulsory led to high completion rates, but class performance decreased, indicating that using the same assessment for formative and for summative purposes is not always beneficial for learning. Finally, these problems were resolved by introducing a two-stage approach: the first stage of each test was formative and provided prompt feedback. However, students had to achieve 80% to progress to the second summative stage of the test. The two-stage online tests led to significantly improved class performance. This novel test design ensures that students go through at least two attempts and therefore fully benefit from the learning opportunities presented by the formative stage. Two-stage online tests present the opportunity to provide regular feedback in large classes and to improve performance not only of good but also of “weak” students.Keywords: e-learning; e-assessment; action research; higher education(Published: 29 April 2013)Citation: Research in Learning Technology 2013, 21: 19153 - http://dx.doi.org/10.3402/rlt.v21i0.19153

Proportional and non-proportional hardening behavior of dual-phase steels

The elastic–plastic response of sheet materials during non-proportional paths is seldom incorporated in constitutive equations used for routine sheet forming simulation, but can have a significant effect on formability and springback. Monotonic tension and compression, coaxial tension–compression (T–C), coaxial compression–tension (C–T), and two-stage/non-coaxial tensile tests were performed for three grades of dual phase steels: DP590, DP780, and DP980. The reverse flow curves have three characteristics: reduced yield stress (Bauschinger effect), rapid transient strain hardening over a few percent strain, and long-term or “permanent” softening. The departure of the reverse hardening curves from monotonic ones is larger than with other typical sheet forming alloys, presumably because of the effects of large second-phase martensite particles in dual-phase steels.A Modified Chaboche type model was applied in this paper to simulate the elastic–plastic behavior of DP steels. In addition to one or more standard nonlinear components of the back stress, a linear term was added to represent the “permanent” offset of hardening following a stress reversal. The parameters for the model were fit using the monotonic and reverse tensile test results only, and the model predictions were then compared with large-strain balanced biaxial bulge test, non-coaxial, two-stage tensile tests, and draw-bend springback tests. The Modified Chaboche model captured the response following a path change much better than isotropic hardening models, but there is a fundamental difference between large-strain hardening after a path change that occurs between path reversals and changes of principal strain axis. The Chaboche model can reproduce either behaviors, but not both simultaneously.

Dissolution Testing of a Controlled-Release Capsule Formulation: Challenges and Solutions Using a Semi-Automated Dissolution System

A controlled-release formulation was developed using enteric-coated beads in a hard gelatin capsule shell. This formulation requires two-stage dissolution testing at low and neutral pH. Because of high toxicity and limited stability of the active pharmaceutical ingredient in aqueous solution at room temperature, it is preferable to limit the analyst’s contact with the sample solutions during testing, and aliquots from the dissolution vessel need to be quickly filtered and refrigerated immediately after sampling. To meet these challenges, a semi-automated dissolution method was developed and validated. Dissolution samples are withdrawn and filtered into vials in the HPLC sample compartment where the temperature is controlled at 4 °C. A fast HPLC method was developed with a 3-min run per injection, enabling analysis of the samples within a very short time after collection, thereby decreasing the sample storage time and potential for degradation. An evaluation was also conducted for comparison of serial and parallel operation of two dissolution baths for the two-stage testing. Serial operation was chosen for this method and applied with a medium-exchange technique, which is safe, robust, easy to perform, compliant with USP, and gives enhanced productivity. INTRODUCTION Enteric coating is a widely used technique for controlled-release oral formulation of pharmaceutical products, specifically for delayed release (1). The formulation purposely limits drug release in the stomach, but facilitates release in the neutral environment of the intestine. It enables protection of the drug from the acidic gastric fluid to prevent possible degradation, as well as protection of the stomach from the drug to avoid possible irritation and adverse effects. The enteric materials are usually applied to substrates like tablets, pellets, or beads. The coated pellets and beads are further filled into a capsule shell (2–5). During development of an oncology drug at BristolMyers Squibb, the technology of coating on beads was applied to make a controlled-release capsule formulation. The physiochemical properties of this drug posed significant challenges for the formulation development, especially its instability in various conditions including (1) in an aqueous acidic environment,( 2) during the process of granulation, and (3) when in contact with enteric coating materials. To overcome these challenges, the enteric-coated beads were developed with multiple different coating layers. This formulation uses sugar spheres as the core substrate. The first coating layer on the sphere is a mixture of the drug and excipient. The second layer coating is an Opadry film that prevents drug from contacting the next layer, the final enteric coating. After the final enteric coating is applied, the beads are filled into a hard gelatin capsule shell (Figure 1). The various dosage strengths of the capsules are made using different fill weights. Evaluation of enteric-coated beads has been performed in both gastric and intestinal media (5, 6). A two-stage dissolution test is specified in USP General Chapter Dissolution (7), in which the integrity of the enteric coating is determined in an acidic environment and the drug release is measured in a neutral environment. The test can be performed using either medium-addition or mediumexchange methods; both start with an acid stage in 0.1 N hydrochloric acid for two hours and follow with a buffer stage in phosphate buffer at pH 6.8 for 45 min or a specific time as needed for the individual drug product. The quality of the dosage form is ensured by meeting USP accep*Corresponding author. e-mail: xujin.lu@bms.com Figure 1. Enteric-coated beads for controlled-release formulation. dx.doi.org/10.14227/DT200213P6

The complete compositional epistasis detection in genome-wide association studies

BackgroundThe detection of epistasis among genetic markers is of great interest in genome-wide association studies (GWAS). In recent years, much research has been devoted to find disease-associated epistasis in GWAS. However, due to the high computational cost involved, most methods focus on specific epistasis models, making the potential loss of power when the underlying epistasis models are not examined in these analyses.ResultsIn this work, we propose a computational efficient approach based on complete enumeration of two-locus epistasis models. This approach uses a two-stage (screening and testing) search strategy and guarantees the enumeration of all epistasis patterns. The implementation is done on graphic processing units (GPU), which can finish the analysis on a GWAS data (with around 5,000 subjects and around 350,000 markers) within two hours. Source code is available at http://bioinformatics.ust.hk/BOOST.html∖#GBOOST.ConclusionsThis work demonstrates that the complete compositional epistasis detection is computationally feasible in GWAS.

Enzyme immunoassay for Clostridium Difficile Infection: Once, Twice or Thrice? Is the Falling Incidence of Clostridium Difficile an Artefact?

Many clinical laboratories rely on stool enzyme immunoassays (EIAs) to diagnose Clostridium difficile infection (CDI). The test is limited by its modest sensitivity which has led to repeat testing. This study examines the impact of repeat stool testing in suspected CDI and describes clinicians’ patterns of ordering tests. In this retrospective study, analysis was performed on a database of toxin EIAs submitted to our laboratory for suspected CDI. For each patient, results of (up to) the first three stool samples, submitted in the first episode of diarrhoea, was analysed. Patterns of ordering tests were examined. 4,987 patients submitted 8,408 stool samples during the study period. Overall, 13.8% of EIAs were positive for C.difficile. Of these, 68% were diagnosed on the first test, 22% on the second and 10% on the third. 9.4% of first test samples were positive, rising to 14.8% for samples tested twice and 17.7% for samples tested thrice. Repeat testing increased CDI prevalence by 50%. Those testing negative for CDI on the first test were more likely to get re-tested compared to those testing positive (25% vs. 16.8%, [χ2= 15.8, p<0.0001]). The prevalence of CDI increased under the three test policy using EIA justifying the need for repeat sample testing. Clinicians were more likely to request repeat tests if the initial result was negative. The implementation of single test rules in some trust will result in a fall in the CDI rate by a third. More work is needed to assess the impact of new two-stage testing policies.

A Multistage Procedure for Decentralized Sequential Multi-Hypothesis Testing Problems

We studied the problem of sequentially testing M ≥ 2 hypotheses with a decentralized sensor network system. In such a system, the local sensors observe raw data and then send quantized observations to a fusion center, which makes a final decision regarding hypothesis is true. Motivated by the two-stage tests in Wang and Mei (2011), we propose a multistage decentralized sequential test that provides multiple opportunities for the local sensors to adjust to the optimal local quantizers. It is demonstrated that when the hypothesis testing problem is asymmetric, the multistage test is second-order asymptotically optimal. Even though this result constitutes an interesting theoretical improvement over two-stage tests that can enjoy only first-order asymptotic optimality, the corresponding practical merits seem to be only marginal. Indeed, performance gains over two-stage procedures with carefully selected thresholds are small.

Two-stage testing procedures with independent filtering for genome-wide gene-environment interaction

Several two-stage multiple testing procedures have been proposed to detect gene-environment interaction in genome-wide association studies. In this article, we elucidate general conditions that are required for validity and power of these procedures, and we propose extensions of two-stage procedures using the case-only estimator of gene-treatment interaction in randomized clinical trials. We develop a unified estimating equation approach to proving asymptotic independence between a filtering statistic and an interaction test statistic in a range of situations, including marginal association and interaction in a generalized linear model with a canonical link. We assess the performance of various two-stage procedures in simulations and in genetic studies from Women's Health Initiative clinical trials.

A procedure for dimensionality analyses of response data from various test designs.

In some popular test designs (including computerized adaptive testing and multistage testing), many item pairs are not administered to any test takers, which may result in some complications during dimensionality analyses. In this paper, a modified DETECT index is proposed in order to perform dimensionality analyses for response data from such designs. It is proven in this paper that under certain conditions, the modified DETECT can successfully find the dimensionality-based partition of items. Furthermore, the modified DETECT index is decomposed into two parts, which can serve as indices of the reliability of results from the DETECT procedure when response data are judged to be multidimensional. A simulation study shows that the modified DETECT can successfully recover the dimensional structure of response data under reasonable specifications. Finally, the modified DETECT procedure is applied to real response data from two-stage tests to demonstrate how to utilize these indices and interpret their values in dimensionality analyses.

Experimental investigation and constitutive modeling for the hardening behavior of 5754O aluminum alloy sheet under two-stage loading

The two-stage loading tests of 5754O aluminum alloy sheet were carried out. In the first loading stage, the uniaxial tensile tests of big sheets were carried out. Then small specimens were cut off from the pre-strained big sheets along different directions. And in the second loading stage, the uniaxial tensile tests of the small specimens were performed. From the experimental results, it is found that the initial yield stress of each specimen in the second loading stage decreases when the strain path changes. In addition, when the strain path changes the transient effect appeared and no obvious permanent softening was observed. In this study, in order to describe the hardening behavior of 5754O aluminum alloy sheet under two-stage loading, the Chaboche type combined isotropic–kinematic hardening models were adopted with Yld2000-2d and Hill48 as yield functions. It is proven that no permanent softening can be described with Chaboche type model in two-stage loading. Three methods for determining the parameters of the hardening models were developed in order to establish accurate isotropic–kinematic hardening model describing the hardening behavior of 5754O aluminum alloy sheet under two-stage loading. The established constitutive models were implemented into the commercial FEM code ABAQUS as a user material subroutine (UMAT) for numerical simulations. By comparing the experimental and simulated results of the two-stage loading tests, the isotropic–kinematic hardening models describing the hardening behavior of 5754O aluminum alloy sheet under two-stage loading were accurately determined. Also, the influences of the characterization method of Hill48 yield function on the accuracy of the resulting hardening models were discussed. It is also found that the established isotropic–kinematic hardening model describing the hardening behavior under two-stage loading can describe reasonably the springback profile of the three point bending of the pre-strained specimen.

Nonadditivity in loglinear models using [formula omitted] and [formula omitted] under product-multinomial sampling

Based on ϕ-divergence measures and minimum ϕ-divergence estimators (MϕEs), we present three families of test statistics for testing nonadditivity in loglinear models. The minimum ϕ-divergence estimator can be seen to be a generalization of the maximum likelihood estimator. In the process of testing nonadditivity, the two-stage tests procedure is usually used as the standard method. The unknown parameters are first estimated by some method (here MϕEs) and then these estimators which are treated as known constants are applied in the second-stage of this procedure. These three families of statistics which generalize the conclusions in Pardo and Pardo (2005) are asymptotically chi-squared. In the last section, we apply our method to a practical example and do a simulation study.

Low-cycle fatigue life of steel P92 under gradual loading

The aim of the study was to determine low-cycle fatigue life characteristics of P92 steel used in the power unit components that work under the highest effort. Steel service life was determined in the tests for total fixed strain Δεt ranges from 0.6 to 1.2 % and with application of two-stage sequence loading. Low-cycle fatigue tests were conducted at room temperature and at 550 °C on MTS-810 testing machine. The tests under sequence loading were carried out for two strain ranges: Δεt = 0.6 and 1.2 %. Sinusoidal load cycles of a coefficient R = −1 were used. Low-cycle fatigue life characteristics were developed with consideration of loading sequence impact. It was found out the fatigue life of an element under sequence loading is strictly correlated with its history of strain, which determines the degree of material microstructure damage. Higher degree of steel damage was found in two-stage tests, in which the larger range of strain Δεt = 1.2 % was used as the first one. Fractography analyses of scrap revealed the patterns of fatigue lines and bands typical for this process as well as numerous secondary cracks occurred to the boundaries of former austenite grains. Analytic method of predicting low-cycle fatigue life of an element under two-stage sequence loading is presented. For this purpose, an energy criterion for the material life developed for tests carried out in conditions of low-cycle fatigue was used.

Improved two‐stage tests for stratified phase II cancer clinical trials

In a single-arm, two-stage, phase II cancer clinical trial for efficacy screening of cytotoxic agents, a common primary endpoint is a binary (yes/no) patient response to treatment. Usually, fixed decision boundaries are used in binomial tests to determine whether the study treatment is promising enough to be studied in a large-scale, randomized phase III trial. We may know in advance that the patient response distribution for a phase II clinical trial will be heterogeneous, making it advisable to stratify patients into subgroups, each with a different prognosis. In this case, fixed decision boundaries may be inappropriate. In this article, we propose two-stage tests based on the Neyman-Pearson lemma. The proposed test statistic is a linear combination of the observed number of responders in each stratum. The test allows adjustment of the decision boundaries to the observed numbers of patients in each stratum and permits sample sizes to be increased adaptively after the originally planned number of patients is observed at each of the two stages. Our numerical results show that the proposed test is more powerful than an existing test in many cases. Finally, we present an application to a Children's Oncology Group phase II clinical trial in patients who relapsed after initial treatment for neuroblastoma.

Dressability of abagas hematite-magnetite ores

A number of magnetic separation processes are used to study preparability of two main hematite-magnetite and magnetite-muscovite ore types extacted it the Abagas iron ore deposit. The dry magnetic separation results were judged as acceptable for the rough processing stage, while the products of the wet magnetic separation were the high-grade iron ore concentrates. The two-stage beneficiation tests on complex hematite-magnetite ores demonstrated much higher cumulative iron recovery from the test ore.

Strain-induced precipitation kinetics of Nb(C,N) and precipitates evolution in austenite of Nb–Ti micro-alloyed steels

In this article, the precipitation–temperature–time (PTT) curves were obtained by plotting 1/A r-time curves (A r represents the area under the flow curves, and some details were clarified in the section “1/A method”) using two-stage interrupted compression testing. Based on the measured PTT curves, the L_J model has been applied to calculate the complete PTT curves by adjusting the modification factor, f, to be 0.022, from which it can be seen that they are both typical C-shaped with the nose temperature ranging from 900 to 925 °C. By using additivity rule, the continuous-cooling-precipitation (CCP) curves were established. And the critical cooling rates for precipitation start were also determined for different cooling start temperatures, and the effects of cooling rates and cooling start temperatures on CCP curves were systematically clarified. The evolution of precipitates was found to be in good agreement with PTT curves. Numerous fine and round precipitates were formed during holding after first compression deformation, which could have hindered static recrystallization kinetics and accounted for the plateaus on 1/A r-time curves. In addition, The cubic precipitates are mainly undissoloved TiN-rich (Ti,Nb) (C,N) carbonitrides. And based on the CCP and continuous-cooling-transformation curves of tested steel, the effect of cooling paths on precipitation and transformation were also discussed in details.

Two-Stage Model-Free Tests of Synergy in Drug Combinations

Laska et al. (1994) proposed a model-free method of detecting synergy in two drug combinations that requires no assumptions about the underlying dose-response curves of the drugs, just an estimate of the potency ratio of the two drugs. It was noted that the power of this method is highly dependent on the accuracy of the potency ratio estimated, with low power when the estimate is inaccurate. Additionally, the test used to detect synergy (the Min test) has been shown to be conservative in many practical applications, and non-monotonic alternative tests that have greater power have been proposed. We suggest a two-stage, non-monotonic alternative to the Laska et al. model-free test that is less dependent on the accuracy of potency ratio estimate and has greater power in many situations. We illustrate the method with an example of two chemotherapeutic agents.

Cost analysis in choosing group size when group testing for Potato virus Y in the presence of classification errors

In many areas of the world, Potato virus Y (PVY) is one of the most economically important disease problems in seed potatoes. In Taiwan, generation 2 (G2) class certified seed potatoes are required by law to be free of detectable levels of PVY. To meet this standard, it is necessary to perform accurate tests at a reasonable cost. We used a two-stage testing design involving group testing which was performed in Taiwan's Seed Improvement and Propagation Station to identify plants infected with PVY. At the first stage of this two-stage testing design, plants are tested in groups. The second stage involves no retesting for negative test groups and exhaustive testing of all constituent individual samples from positive test groups. In order to minimise costs while meeting government standards, it is imperative to estimate optimal group size. However, because of limited test accuracy, classification errors for diagnostic tests are inevitable; to get a more accurate estimate, it is necessary to adjust for these errors. Therefore, this paper describes an analysis of diagnostic test data in which specimens are grouped for batched testing to offset costs. The optimal batch size is determined by various cost parameters as well as test sensitivity, specificity and disease prevalence. Here, the Bayesian method is employed to deal with uncertainty in these parameters. Moreover, we developed a computer program to determine optimal group size for PVY tests such that the expected cost is minimised even when using imperfect diagnostic tests of pooled samples. Results from this research show that, compared with error free testing, when the presence of diagnostic testing errors is taken into account, the optimal group size becomes smaller. Higher diagnostic testing costs, lower costs of false negatives or smaller prevalence can all lead to a larger optimal group size. Regarding the effects of sensitivity and specificity, optimal group size increases as sensitivity increases; however, specificity has little effect on determining optimal group size. From our simulated study, it is apparent that the Bayesian method can truly update the prior information to more closely approximate the intrinsic characteristics of the parameters of interest. We believe that the results of this study will be useful in the implementation of seed potato certification programmes, particularly those which require zero tolerance for quarantine diseases in certified tubers.

Testing Statistical Hypotheses on Stochastic Ordering of Discrete Distributions

Let F 0 and F be distribution functions over a finite sample space. We propose methods to test whether F is stochastically smaller than F 0 based on a random sample from the distribution F. We propose one- and two-stage tests, investigate the properties of the power function, and provide methods for sample size determination and critical value computation. Our methods can be applied to phase II clinical trials with multinomial ordinal responses as endpoints.

Two-Stage Designs with Additional Futility Tests for Phase II Clinical Trials with Heterogeneous Patient Populations

The patient population for a phase II clinical trial sometimes consists of multiple subgroups categorized by different prognostic factors such as age, disease stage, etc. Response rates across such subgroups often are heterogeneous; that is, the therapy under study may be inefficacious in some groups but efficacious in others. In this article, we propose a two-stage testing procedure consisting of futility tests and efficacy tests. The futility test is performed first within each subgroup. If there is strong evidence that the therapy is inefficacious in some subgroups, these subgroups are excluded from further consideration. The test then is performed across only the remaining subgroups. We carefully define and compute the Type I error rate and powers under various alternative hypotheses. Our numerical studies show that with proper designs, the addition of futility tests causes little loss of power under the alternative hypothesis that the study therapy is efficacious in all subgroups. Simultaneously, the proposed testing procedure detects inefficacious subgroups with a moderate or high probability when true response rates in these subgroups are low. Finally, we present an application to a Children's Oncology Group phase II clinical trial in relapsed neuroblastoma patients.