Trade-off between false positives and false negatives in the linkage analysis of complex traits

This study examines the issue of false positives in genomic scans for detecting complex trait loci using sibpair linkage methods and investigates the trade-off between the rate of false positives and the rate of false negatives. It highlights the tremendous cost in terms of power brought about by an excessive control of type I error and, at the same time, confirms that a larger number of false positives can occur otherwise in the course of a genomic scan. Finally, it compares the power and rate of false positives obtained in preplanned replicated studies conducted using a liberal significance level to those for single-step studies that use the same total sample size but stricter levels of significance. For the models considered here, replicate studies were found more attractive as long as one is willing to accept a trade-off, exchanging a much lower rate of false negatives for a slight increase in the rate of false positives. Genet. Epidemiol. 14:453–464,1997. © 1997 Wiley-Liss, Inc.     

A simple and fast two-locus quality control test to detect false positives due to batch effects in genome-wide association studies

The impact of erroneous genotypes having passed standard quality control (QC) can be severe in genome-wide association studies, genotype imputation, and estimation of heritability and prediction of genetic risk based on single nucleotide polymorphisms (SNP). To detect such genotyping errors, a simple two-locus QC method, based on the difference in test statistic of association between single SNPs and pairs of SNPs, was developed and applied. The proposed approach could detect many problematic SNPs with statistical significance even when standard single SNP QC analyses fail to detect them in real data. Depending on the data set used, the number of erroneous SNPs that were not filtered out by standard single SNP QC but detected by the proposed approach varied from a few hundred to thousands. Using simulated data, it was shown that the proposed method was powerful and performed better than other tested existing methods. The power of the proposed approach to detect erroneous genotypes was ～80% for a 3% error rate per SNP. This novel QC approach is easy to implement and computationally efficient, and can lead to a better quality of genotypes for subsequent genotype-phenotype investigations.     

"药价实高"与"药价虚高"

分析了我国“药价实高”和“药价虚高”现象，并提出解决“药价虚高”的一些看法：改变政府管制方式、规范医方用药行为、科学解决医方生存和发展问题。     

TEOAE进行新生儿听力筛查的假阳性因素分析及干预

[目的]分析瞬态声耳声发射(transient evoked otoacoustic emission,TEOAE)进行新生儿听力筛查中的假阳性因素并采取相应的干预措施,提高初筛和复筛的通过率.[方法]对TEOAE未通过者进行复查,总结TEOAE筛查中的假阳性因素.对比本院与外院初筛通过率,及本院与外院初筛阳性者均在本院复筛的通过率.[结果] 在TEOAE筛查中有很多导致假阳性的因素,认识这些因素并采取干预措施,通过率明显提高(P＜O.01).[结论]应尽量避免假阳性因素,提高新生儿听力筛查的质控水平.     

Multiple true–false items: a comparison of scoring algorithms

Multiple true–false (MTF) items are a widely used supplement to the commonly used single-best answer (Type A) multiple choice format. However, an optimal scoring algorithm for MTF items has not yet been established, as existing studies yielded conflicting results. Therefore, this study analyzes two questions: What is the optimal scoring algorithm for MTF items regarding reliability, difficulty index and item discrimination? How do the psychometric characteristics of different scoring algorithms compare to those of Type A questions used in the same exams? We used data from 37 medical exams conducted in 2015 (998 MTF and 2163 Type A items overall). Using repeated measures analyses of variance (rANOVA), we compared reliability, difficulty and item discrimination of different scoring algorithms for MTF with four answer options and Type A. Scoring algorithms for MTF were dichotomous scoring (DS) and two partial credit scoring algorithms, PS₅₀ where examinees receive half a point if more than half of true/false ratings were marked correctly and one point if all were marked correctly, and PS_1/n where examinees receive a quarter of a point for every correct true/false rating. The two partial scoring algorithms showed significantly higher reliabilities (α_PS1/n = 0.75; α_PS50 = 0.75; α_DS = 0.70, α_A = 0.72), which corresponds to fewer items needed for a reliability of 0.8 (n_PS1/n = 74; n_PS50 = 75; n_DS = 103, n_A = 87), and higher discrimination indices (r_PS1/n = 0.33; r_PS50 = 0.33; r_DS = 0.30; r_A = 0.28) than dichotomous scoring and Type A. Items scored with DS tend to be difficult (p_DS = 0.50), whereas items scored with PS_1/n become easy (p_PS1/n = 0.82). PS₅₀ and Type A cover the whole range, from easy to difficult items (p_PS50 = 0.66; p_A = 0.73). Partial credit scoring leads to better psychometric results than dichotomous scoring. PS₅₀ covers the range from easy to difficult items better than PS_1/n. Therefore, for scoring MTF, we suggest using PS₅₀.     

Misinformation,partial knowledge and guessing in true/false tests

CONTEXT: Examiners disagree on whether or not multiple choice and true/false tests should be negatively marked. Much of the debate has been clouded by neglect of the role of misinformation and by vagueness regarding both the specification of test types and "partial knowledge" in relation to guessing. Moreover, variations in risk-taking in the face of negative marking have too often been treated in absolute terms rather than in relation to the effect of guessing on test unreliability. OBJECTIVES: This paper aims to clarify these points and to compare the ill-effects on test reliability of guessing and of variable risk-taking. METHODS: Three published studies on medical students are examined. These compare responses in true/false tests obtained with both negative marking and number-right scoring. The studies yield data on misinformation and on the extent to which students may fail to benefit from distrusted partial knowledge when there is negative marking. A simple statistical model is used to compare variations in risk-taking with test unreliability due to blind guessing under number-right scoring conditions. CONCLUSIONS: Partial knowledge should be least problematic with independent true/false items. The effect on test reliability of blind guessing under number-right conditions is generally greater than that due to the over-cautiousness of some students when there is negative marking.     

Evaluating false positives in two hospital discharge data sets of the Birth Defects Monitoring Program.

The principal goal in this study was to quantify false positives in the hospital discharge data of the Birth Defects Monitoring Program conducted by the Centers of Disease Control and Prevention. The two hospital data processing agencies which contribute data to the Birth Defects Monitoring Program, the Commission on Professional and Hospital Activities and the McDonnell Douglas Health Information Systems, had respective levels of false positives of 13.2 percent and 8.5 percent, levels which were statistically different from each other. These false positive levels should be considered minimal because these data bases do not include information on sick babies who may be transferred into or out of member hospitals, and who may have their initial diagnoses significantly modified. Potential correlates of false positives were evaluated, including hospital size, diagnostic certainty, race, sex, and insurance source. Two-thirds of all false positives were due to the miscoding of correctly diagnosed anomalies, and another quarter were clearly contradicted in notes easily available before the patients were discharged. The authors hope that this study of false positives will enhance the interpretation of the Birth Defects Monitoring Program data and lead to improved understanding of data collection and processing.     

Flagging clinical adverse experiences: reducing false discoveries without materially compromising power for detecting true signals

Comparative analyses of safety/tolerability data from a typical phase III randomized clinical trial generate multiple p-values associated with adverse experiences (AEs) across several body systems. A common approach is to 'flag' any AE with a p-value less than or equal to 0.05, ignoring the multiplicity problem. Despite the fact that this approach can result in excessive false discoveries (false positives), many researchers avoid a multiplicity adjustment to curtail the risk of missing true safety signals. We propose a new flagging mechanism that significantly lowers the false discovery rate (FDR) without materially compromising the power for detecting true signals, relative to the common no-adjustment approach. Our simple two-step procedure is an enhancement of the Mehrotra-Heyse-Tukey approach that leverages the natural grouping of AEs by body systems. We use simulations to show that, on the basis of FDR and power, our procedure is an attractive alternative to the following: (i) the no-adjustment approach; (ii) a one-step FDR approach that ignores the grouping of AEs by body systems; and (iii) a recently proposed two-step FDR approach for much larger-scale settings such as genome-wide association studies. We use three clinical trial examples for illustration.     

Imputations of missing values in practice: results from imputations of serum cholesterol in 28 cohort studies

Missing values, common in epidemiologic studies, are a major issue in obtaining valid estimates. Simulation studies have suggested that multiple imputation is an attractive method for imputing missing values, but it is relatively complex and requires specialized software. For each of 28 studies in the Asia Pacific Cohort Studies Collaboration, a comparison of eight imputation procedures (unconditional and conditional mean, multiple hot deck, expectation maximization, and four different approaches to multiple imputation) and the naive, complete participant analysis are presented in this paper. Criteria used for comparison were the mean and standard deviation of total cholesterol and the estimated coronary mortality hazard ratio for a one-unit increase in cholesterol. Further sensitivity analyses allowed for systematic over- or underestimation of cholesterol. For 22 studies for which less than 10% of the values for cholesterol were missing, and for the pooled Asia Pacific Cohort Studies Collaboration, all methods gave similar results. For studies with roughly 10-60% missing values, clear differences existed between the methods, in which case past research suggests that multiple imputation is the method of choice. For two studies with over 60% missing values, no imputation method seemed to be satisfactory.     

Estimating the number of true discoveries in genome-wide association studies

Recent genome-wide association studies have reported the discoveries of genetic variants of small to moderate effects. However, most studies of complex diseases face a great challenge because the number of significant variants is less than what is required to explain the disease heritability. A new approach is needed to recognize all potential discoveries in the data. In this paper, we present a practical model-free procedure to estimate the number of true discoveries as a function of the number of top-ranking SNPs together with the confidence bounds. This approach allows a practical methodology of general utility and produces relevant statistical quantities with simple interpretation.     

Inversions in true/false and in multiple choice questions – a new form of item analysis

Total ignorance is expressed not by consistently selecting the wrong answers in a true/false or multiple choice examination, but rather by making a random choice from all the alternatives available, resulting in a predictable proportion of correct responses.
When a significant majority of students answers incorrectly in a true/false question, it must therefore be concluded that this is not the result of ignorance, but of some factor which has led to their deliberately choosing the wrong answer. This situation has been termed an 'inversion'.
Of 2057 true/false examination questions set in this department over the past 3 years, 140 (6.8%) were found to be 'inverted'. In about one third of these cases the 'inversion' was found to be due to a wrong model answer, an ambiguous or misleading question, or to incorrect teaching.
The remaining two-thirds of the 'inversions' could be ascribed to attempts to work out the answer from inadequate or inappropriate first principles.
It has been our policy to delete all 'inverted' questions from the examination (unless the model answer needed to be changed), because the score obtained when the students deliberately chose the wrong answer is less than the score they would have obtained had they been totally ignorant and had ventured a guess or had chosen not to answer the question.
This concept can also be applied to multiple choice questions.     

Multistage designs in the genomic era: providing balance in complex disease studies

In this summary paper, we describe the contributions included in the Multistage Design group (Group 14) at the Genetic Analysis Workshop 15, which was held during November 12-14, 2006. Our group contrasted and compared different approaches to reducing complexity in a genetic study through implementation of staged designs. Most groups used the simulated dataset (problem 3), which provided ample opportunities for evaluating various staged designs. A wide range of multistage designs that targeted different aspects of complexity were explored. We categorized these approaches as reducing phenotypic complexity, model complexity, analytic complexity or genetic complexity. In general we learned that: (1) when staged designs are carefully planned and implemented, the power loss compared to a single-stage analysis can be minimized and study cost is greatly reduced; (2) a joint analysis of the results from each stage is generally more powerful than treating the second stage as a replication analysis.     

Preprandial blood glucose values: influence on glycemic response studies

Through regressional analyses on data from two glycemic response studies, the influence of various preprandial blood glucose (PPBG) values upon glycemic response is highlighted. Repeated calculations on different subsamples of meals where the PPBG was below a given value showed that the influence of PPBG upon net glycemic response not was linear in the whole PPBG range. It was found that a PPBG less than 13 mmol/L had no influence upon net glycemic response whereas PPBGs greater than 13 mmol/L were significantly negatively correlated to the net glycemic response. To obtain more valid glycemic index figures we recommend that participants with PPBG greater than 13 mmol/L are excluded from glycemic response studies and that the number of participants are increased substantially.     

Issues in genomic screening: critical values, sample sizes, and the ability to detect linkage

The aims of this study were to empirically investigate the ability of affected sib pairs (ASPs) to localize a gene through screening and to explore estimation of lod score critical values through resampling. To do so, we repartitioned 25 replicates of 100 simulated nuclear families into six data sets of sizes 100, 200, 300, 400, 500, and 1,000 and chose at most one mildly ASP per family. Using all marker data, we calculated maximum lod scores across the six-chromosome genome for each set. Then, we determined the cutoff value corresponding to a 5% genome-wide false positive rate using both the method of Lander and Kruglyak [1995] and a simple resampling algorithm that allows greater scan-specific flexibility. For chromosome 1, the ability of the ASPs to detect the region between markers 9 and 10 clearly increases with the sample size, and genome-wide significance is achieved for samples of size 400 or greater. Also, as expected, the critical values based on the less conservative resampling approach are generally slightly smaller than those from theoretical calculations based on the Ornstein-Uhlenbeck diffusion process of Lander and Kruglyak.     

XX SRY-negative true hermaphrodism in two dogs: clinical, morphological, genetic and cytogenetic studies

This work aimed at giving a deeper insight into peculiar cases of intersexuality occurring in dogs and known as XX true hermaphrodism due to the existence of both testicular and ovarian tissue in one or both gonads in the presence of an XX chromosome constitution. Clinical, histological and genetic approaches were used in the study of an 8-month-old Cocker Spaniel dog and a 3-year-old mixed-breed Pitbull, both showing a female phenotype, clitoromegaly and male behavior. A normal female karyotype (2n = 78,XX) was noticed, and polymerase chain reaction failed to detect SRY in genomic DNA obtained from peripheral blood lymphocytes of both dogs. The reproductive tract was removed by standard ovariohysterectomy and processed for histology. Thereafter, a normal female phenotype was reconstructed by vaginoplasty. Histological examination revealed bilateral ovotestis in both cases: the gonads showed immature testicular parenchyma containing seminiferous tubules, Sertoli and Leydig cells, but no signs of spermatogenesis, together with differently developed ovarian follicles containing oocytes. In the ovotestes, steroidogenesis was detected by P450c17-immunoreactivity in Leydig cells as well as in theca cells, whereas no MIS-immunoreactivity was shown by the Sertoli cells. Genital tracts of Wolffian and Müllerian origin co-existed in both subjects. Both dogs belong to the very rare cases in which testicular tissue develops in the absence of the key gene, SRY. Up to date very few genetic events have been associated with this abnormal sexual differentiation: SOX9 over-expression and RSPO1 mutation. Nevertheless, neither of them has been found in these dogs.