Multiple hypothesis testing in radiobiological omics study / 中华放射医学与防护杂志

With the development of high-throughput sequencing technology, the high-dimensional massive data obtained in omics study puts forward new requirements for statistical analysis. In this case, the traditional theory of single hypothesis testing is no longer applicable, and the issue of multiple hypothesis testing has received increasing attention. This paper introduced three commonly used error measures in multiple testing-family-wise error rate (FWER), false discovery rate (FDR), and positive false discovery rate (pFDR), and the control process in radiobiological omics data analysis, in order to provide a reference for statistical analysis of radiobiological data.

How to use χ2 test correctly——the analysis of a common risk difference for the data of a multiway table and the implementation of SAS software / 四川精神卫生

The purpose of this article was to introduce the risk rate difference analysis method of the g×2×2 table data and the calculation method based on the SAS software. The tasks that needed to be completed included the following two terms： first， the point estimation of the common risk rate difference and its confidence interval estimation； second， the test "whether the common risk rate difference was equal to 0". Among them， there were a total of 6 approaches for achieving the aforementioned first task； and three approaches for achieving the aforementioned second task. Based on the SAS software and an example， the article realized the point estimation， the confidence interval estimation and the hypothesis test of the common risk rate difference， respectively. Next， the SAS output results were explained and the statistical and professional conclusions were made.

Regularized Hypothesis Testing in Random Fields with Applications to Neuroimaging / Pruebas de Hipótesis Regularizadas en Campos Aleatorios con Aplicaciones a Neuroimágenes

Abstract The task of determining for which elements of a random field (e.g., pixels in an image) a certain null hypothesis may be rejected is a relevant problem in several scientific areas. In the current contribution, we introduce a new method for performing this task, the regularized hypothesis testing (RHT) method, focusing on its use in neuroimaging research. RHT is based on the formulation of the hypothesis testing task as a Bayesian estimation problem, with the previous application of a Markovian random field. The latter allows for the incorporation of local spatial information and considers different noise models, including spatially correlated noise. In tests on synthetic data showing regular activation levels on uncorrelated noise fields, RHT furnished a true positive rate (TPR) of 0.97, overcoming the state-of-the-art morphology-based hypothesis testing (MBHT) method and the traditional family-wise error rate (FWER) method, which afforded 0.93 and 0.58, respectively. For fields with highly correlated noise, the TPR provided by RHT was 0.65, and by MBHT and FWER was 0.35 and 0.29, respectively. For tests utilizing real functional magnetic resonance imaging (fMRI) data, RHT managed to locate the activation regions when 60% of the original signal were removed, while MBHT located only one region and FWER located none.

Resumen En varias áreas científicas aparece el problema de determinar los elementos de un campo aleatorio (por ejemplo, píxeles en una imagen) en los que se puede rechazar una cierta hipótesis nula. En este artículo presentamos un nuevo método para realizar esta tarea, centrado en aplicaciones para investigación de neuroimagen. Nuestra propuesta se basa en la formulación de pruebas de hipótesis como un problema de estimación Bayesiana, usando como a priori un campo aleatorio Markoviano, que permite incorporar información espacial local y considera diferentes modelos de ruido, incluido el ruido correlacionado espacialmente. Para pruebas en datos sintéticos con niveles de activación regulares sobre campos de ruido no correlacionado, nuestro método obtiene una tasa de verdaderos positivos (TPR) de 0.97, superando al método del estado del arte MBHT y al método de control FWER que obtienen 0.93 y 0.58 respectivamente; para campos con ruido altamente correlacionado, nuestro método obtiene un TPR de 0.65, mientras que MBHT y FWER obtienen 0.35 y 0.29 respectivamente. Para pruebas con datos reales de fMRI, nuestro método localiza las regiones de activación cuando removemos 60% de la señal original, mientras que MBHT no localiza región alguna y FWER localiza una de las dos regiones.

Control de falsos descubrimientos en mapeo asociativo con poblaciones estructuradas / False discovery rate control in association mapping with genetically structured populations

Las pruebas de asociación entre marcadores moleculares y variables fenotípicas son cruciales para la identificación de QTL (Quantitative Trait Loci). Los avances biotecnológicos incrementaron la disponibilidad de marcadores genéticos y consecuentemente el número de pruebas de la asociación fenotipo-genotipo. El incremento de pruebas de significancia estadística a realizar en simultaneo (multiplicidad) demanda correcciones de los valores-p obtenidos para cada prueba de hipótesis de manera de mantener acotada las tasas de error para la familia de pruebas de asociación. Las correcciones estadísticas clásicas para el problema de multiplicidad, como Bonferroni, el método de control de la tasa de falsos descubrimientos (FDR) y el número efectivo de pruebas (Meff), son ampliamente usadas, pero fueron desarrolladas para datos independientes. Sin embargo, cuando las poblaciones de mapeo están genéticamente estructuradas los datos dejan de ser independientes. En este trabajo, proponemos un método de corrección por multiplicidad basado en estimación del número efectivo de pruebas desde un modelo que ajusta por la estructura de correlación subyacente. Se evalúa el desempeño del procedimiento propuesto a través del análisis de los valores-p obtenidos para un conjunto de QTL simulados. Los resultados sugieren que el método propuesto provee control de la tasa de falsos positivos y presenta mayor potencia que otros métodos de corrección por multiplicidad usados en mapeo asociativo.

The association tests between molecular markers and phenotypic traits are crucial for the Quantitative Trait Loci (QTL) identification. Biotechnological advances increased the molecular marker information; consequently, the number of genotype-phenotype association tests required incremented too. The multiple statistical inferences (multiplicity) demand corrections of the p-values obtained for each comparison in order to keep limited the error rates for the family of association tests. However, classic statistical correction methods such as Bonferroni, False Discovery Rate (FDR) and the Effective Number of Independent Test (Meff) were developed in the context of independent data. Wherever, when the population genetic structure is present, the data are no longer independent. In this paper, we propose a method of correction for multiplicity based on estimation of the effective number of tests from a model that adjust for the underlying correlation structure. We evaluate the performance of the proposed procedure in the estimation of p-values for a set of simulated QTL. The results suggest that the proposed method provides control of FDR and has more power than other methods for multiplicity correction used in association mapping.

Problems needing special attention when hypothesis test is applied / 中华医学图书情报杂志

Whether to select one-tailed test or two-tailed test,how to establish null hypothesis and alternative hypothesis,how to improve the test efficacy were elaborated,and the common problems encountered when hypothesis was applied in scientific papers were pointed out with examples.

Uso de pruebas estadísticas inferenciales en la Revista Cubana de Medicina Militar / Use of Inferential Statistical Tests in the Cuban Journal of Military Medicine

Introducción: desde hace años, existe un debate sobre el uso de las pruebas estadísticas inferenciales en los reportes de resultados de investigación, se destaca la crítica al empleo de las pruebas de significación estadística y sus limitaciones. Objetivos: determinar la frecuencia de empleo de las pruebas de significación estadística (PSE) e intervalos de confianza (IC) por tipos de estudio publicado, cómo se reflejan los resultados de estas, la influencia del tamaño de la muestra, así comosu vinculación con las conclusiones. Resultados: en el periodo 2010 - 2015 de 150 artículos originales, 98 por ciento fueron descriptivos o explicativos y de ellos, el 95 por ciento emplea las PSE, solas o con IC. Predomina el uso de las PSE solas (69 por ciento de los trabajos). En el 25 por ciento se explica la selección del nivel de significación utilizado y el 53 por ciento de los estudios reflejan las cifras exactas de las pruebas realizadas. Solo el 15 por ciento menciona la influencia del tamaño de la muestra en relación con los resultados de las pruebas estadísticas. En las conclusiones, el 86 por ciento de los artículos se refieren adecuadamente a los objetivos del estudio. Conclusiones: predomina el uso de las PSE e IC, fundamentalmente de las PSE, más de la mitad de los trabajos mencionan los resultados precisos de las pruebas, la mayoría no argumenta la relación de estos resultados con el tamaño de la muestra y los autores elaboran las conclusiones de acuerdo con los objetivos planteados en el estudio(AU)

Introduction: For years there has been a debate about the use of inferential statistical tests in the reports of research results, highlighting the criticism to the use of tests of statistical significance and its limitations. Objectives: To determine the frequency of use of statistical significance tests (SST) and confidence intervals (CI) by published study types, how the results are reported, and the influence of sample size, as well as their relationship with the conclusions. Results: In the period 2010-2015 of 150 original articles, 98 percent were descriptive or explanatory and of them, 95 percent used SST alone or with CI. The use of SST alone (69 percent of the articles) predominates. In 25 percent the significance level selection is explained and 53 percent of the studies reflect the exact figures of the tests performed. Only 15 percent mentions the influence of sample size on the results of statistical tests. In the conclusions, 86 percent of the articles refer adequately to the objectives of the study. Conclusions: SST and CI use predominate, mainly SST, more than half of the studies mention the precise results of the tests, most do not argue the relation of these results to the sample size and the authors elaborate the conclusions in accordance with the objectives set out in the study(AU)

Revelation of shrunken or stretched binomial dispersion and public perception of situations which might spread AIDS or HIV.

Background: In 1985, the center for disease control coined the name: “Acquired Immune Deficiency Syndrome (AIDS)” to refer a deadly illness. The World Health Organization (WHO) estimated that about 33.4 million people were suffering with AIDS and two million people (including 330,000 children) died in 2009 alone in many parts of the world. A scary fact is that the public worry about situations which might spread AIDS according to reported survey result in Meulders et al. (2001). This article develops and illustrates an appropriate statistical methodology to understand the meanings of the data. Methods: While the binomial model is a suitable underlying model for their responses, the data mean and dispersion violates the model’s required functional balance between them. This violation is called over-under dispersion. This article creates an innovative approach to assess whether the functional imbalance is too strong to reject the binomial model for the data. In a case of rejecting the model, what is a correct way of warning the public about the spreads of AIDS in a specified situation? This question is answered. Results: In the survey data about how AIDS/HIV might spread according to fifty respondents in thirteen nations, the functional balance exists only in three cases: “needle”, “blood” and “sex” justifying using the usual binomial model (1). In all other seven cases: “glass”, “eating”, “object”, “toilet”, “hands”, “kissing”, and “care” of an AIDS or HIV patient, there is a significant imbalance between the dispersion and its functional equivalence in terms of the mean suggesting that the new binomial called imbalanced binomial distribution (6) of this article should be used. The statistical power of this methodology is indeed excellent and hence the practitioners should make use of it. Conclusion: The new model called imbalanced binomial distribution (6) of this article is versatile enough to be useful in other research topics in the disciplines such as medicine, drug assessment, clinical trial outcomes, business, marketing, finance, economics, engineering and public health.

An assessment of nurses' sufficient immunity when treating infectious patients using bumped-up binomial model.

Background: In times of an outbreak of a contagious deadly epidemic1-4 such as severe acute respiratory syndrome (SARS), the patients are quarantined and rushed to an emergency department of a hospital for treatment. Paradoxically, the nurses who treat them to become healthy get infected in spite of the nurses’ precautionary defensive alertness. This is so unfortunate because the nurses might not have been in close contact with the virus otherwise in their life. The nurses’ sufficient immunity level is a key factor to avert hospital site infection. Is it possible to quantify informatics about the nurses’ immunity from the virus? Methods: The above question is answered with a development of an appropriate new model and methodology. This new frequency trend is named Bumped-up Binomial Distribution (BBD). Several useful properties of the BBD are derived, applied, and explained using SARS data5 in the literature. Though SARS data are considered in the illustration, the contents of the article are versatile enough to analyze and interpret data from other contagious diseases. Results: With the help of BBD (3) and the Toronto data in Table 1, we have identified the informatics about the attending nurses’ sufficient immunity level. There were 32 nurses providing 16 patient care services. Though the nurses were precautionary to avoid infection, not all of them were immune to infection in those activities. Using the new methodology of this article, their sufficient immunity level is estimated to be only 0.25 in a scale of zero to one with a p-value of 0.001. It suggests that the nurses’ sufficient immunity level is statistically significant. The power of accepting the true alternative hypothesis of 0.50 immunity level, if it occurs, is calculated to be 0.948 in a scale of zero to one. It suggests that the methodology is powerful. Conclusions: The estimate of nurse’s sufficient immunity level is a helpful factor for the hospital administrators in the time of making work schedules and assignments of the nurses to highly contagious patients who come in to the emergency or regular wings of the hospital for treatment. When the approach and methodology of this article are applied, it would reduce if not a total elimination of the hospital site infections among the nurses and physicians.

Does smoking delay pregnancy? Data analysis by a tweaked geometric distribution answers.

Background: Smoking is generally known to be carcinogenic and health hazardous. What is not clear is whether the smoking impacts on the woman’s reproductive process. There have been medical debates on whether a woman in the child bearing age may delay her pregnancy due to smoking. A definitive conclusion on this issue has not been reached perhaps due to a lack of appropriate data evidence. The missing link to answer the question might be exercising a suitable model to extract the pertinent data information on the number of missed menstrual cycles by smoking women versus non-smoking women. This article develops and demonstrates a statistical methodology to answer the question. Methods: To construct such a needed methodology, a new statistical distribution is introduced as an underlying model for the data on the number of missed menstrual cycles by women who smoke. This new distribution is named Tweaked Geometric Distribution (TGD). Several useful properties of the TGD are derived and explained using a historical data in the literature. Results: In the data of 100 smokers and 486 non-smokers, on the average, smoking women missed 3.22 menstrual cycles and non-smoking women missed only 1.96 menstrual cycles before becoming pregnant. The smoking women exhibited more variation than the non-smoking women and it suggests that the non-smoking women are more homogeneous while the smoking women are more heterogeneous. Furthermore, the impairment level to pregnancy due to smoking among the 486 women is estimated to be 5% in a possible scale of zero to one. The 5% impairment level appears like a small amount, but its impact can be felt once it is cast in terms of fecundity. What is fecundity? The terminology fecundity refers the chance for a woman to become pregnant. The fecundity is 0.24 for smoking woman while it is 0.34 for non-smoking woman. The fecundity of a non-smoking woman is more than twice the fecundity of a smoking woman. Conclusion: The smoking is really disadvantageous to any one in general and particularly to a woman who wants to become pregnant.

Interpretando correctamente en salud pública estimaciones puntuales, intervalos de confianza y contrastes de hipótesis / Accurate interpretation of point estimates, confidence intervals, and hypothesis tests in public health

El presente ensayo trata de aclarar algunos conceptos utilizados habitualmente en el campo de investigación de la salud pública, que en numerosas situaciones son interpretados de manera incorrecta. Entre ellos encontramos la estimación puntual, los intervalos de confianza, y los contrastes de hipótesis. Estableciendo un paralelismo entre estos tres conceptos, podemos observar cuáles son sus diferencias más importantes a la hora de ser interpretados, tanto desde el punto de vista del enfoque clásico como desde la óptica bayesiana.

This essay reviews some statistical concepts frequently used in public health research that are commonly misinterpreted. These include point estimates, confidence intervals, and hypothesis tests. By comparing them using the classical and the Bayesian perspectives, their interpretation becomes clearer.