A novel gene expression index (GEI) with software support for comparing microarray gene signatures

This study was aimed to examine the validity of commonly used statistical tests for comparison of expression data from simulated and real gene signatures as well as pathway-characterized gene sets. A novel algorithm based on 10 sub-gradations (5 for up- and 5 for down-regulation) of fold-changes has been designed and testified using an Excel add-in software support. Our findings showed the limitations of conventional statistics for comparing the microarray gene expression data. However, the newly introduced Gene Expression Index (GEI) appeared to be more robust and straightforward for two-group comparison of normalized data. The software automation simplifies the task and the results are displayed in a comprehensive format including a color-coded bar showing the intensity of cumulative gene expression.     

A Comparative Statistical Analysis of Genetic Distances. II–Confidence Intervals and Tests of Comparison

We present methods for constructing the confidence intervals of the estimations of absolute genetic distance and Nei's minimum and standard genetic distances. Two different estimators of the standard distance are considered: jack-knife and corrected. A Monte-Carlo simulation analysis permitted to choose the best adapted confidence interval type for each distance estimator and for each size of the samples of loci and of gametes, taking into account the magnitude of the estimation value. Simulated true confidence levels showed that the best results are obtained using abolute genetic distance. We also give rules for testing the significance of the difference of two distance estimations. Monte-Carlo simulations were performed in order to obtain the distributions of the difference of two distance estimations under the null hypothesis and consequently the true type I errors of tests. Test powers, obtained by simulation under the alternative hypothesis, also showed the superiority of absolute genetic distance.     

The unifrac significance test is sensitive to tree topology

Long et al. (BMC Bioinformatics 2014, 15(1):278) describe a “discrepancy” in using UniFrac to assess statistical significance of community differences. Specifically, they find that weighted UniFrac results differ between input trees where (a) replicate sequences each have their own tip, or (b) all replicates are assigned to one tip with an associated count. We argue that these are two distinct cases that differ in the probability distribution on which the statistical test is based, because of the differences in tree topology. Further study is needed to understand which randomization procedure best detects different aspects of community dissimilarities.     

基于极大似然准则的生物序列模体的贝叶斯假设检验

针对生物信息学中序列模体的显著性检验问题,提出了一种基于极大似然准则的贝叶斯假设检验方法.将模体的显著性检验转化为多项分布的拟合优度检验问题,选取Dirichlet分布作为多项分布的先验分布并采用Newton-Raphson算法估计Dirichlet分布的超参数,使得数据的预测分布达到最大.应用贝叶斯定理得到贝叶斯因子进行模型选择,用于评价模体检验的统计显著性,这种方法克服了传统多项分布检验中构造检验统计量并计算其在零假设下确切分布的困难.选择JASPAR数据库中107个转录因子结合位点和100组随机模拟数据进行实验,采用皮尔逊积矩相关系数作为评价检验质量的一个标准,发现实验结果好于传统的模体检验的一些方法.     

Evaluation of FORS-D Analysis: A Comparison with the Statistically Significant Stem-loop Potential

The stem-loop potential of a nucleic acid segment (expressed as a FONS value), decomposes into base composition-dependent and base order-dependent components. The latter, expressed as a FORS-D value, is derived by subtracting the value of the base composition-dependent component (FORS-M) from the FONS value. FORS-D analysis is the use of FORS-D values to estimate the potential of local base order to contribute to a stem-loop structure, and it has been used to investigate the relationship between stem-loop structure and other selective pressures on genomes. In the present study, we evaluated the reliability of FORS-D analysis by comparing it with statistically significant stem-loop potential, another robust method developed by Le and Maizel for examining stem-loop structure. We found that FORS-M values calculated using 10 randomized sequences are as reliable as those calculated using 100 randomized sequences. The resulting FORS-D values have a similar trend and distribution as statistically significant stem-loop potential, implying that FORS-D analysis is as reliable as the latter in measuring the distribution of base order-dependent stem-loop potential. Since the calculation of the FORS-M values is time consuming, the integrated program Bodslp developed by us will become a convenient tool for large-scale FORS-D analysis. The results also suggest that for some purposes the online program SigStb developed by Le and Maizel may be used as an alternative tool for FORS-D analysis. C. Y. Zhang and J. F. Wei contributed equally to this paper.     

Sampling techniques and the use of Tang's procedure in insect population dynamics studies

Some calculations were performed usingTang 's method as an aid in planning experiments for studying the population dynamics of the Jeffrey pine beetle. The population dynamics studies were aimed at detecting the importance of specific effects, e. g., tree diameter, tree height. TheTang procedure is a method of estimating the sample size required to detect effects of a given magnitude with analysis of variance tests. Using this procedure some sample calculations were performed which indicated the sample size needed, and the efficacy of different strategies of improving the results, e. g., increasing the number of trees sampled versus increasing the area of the tree sampled. The statistical parameters used in the calculations were estimated from some preliminary sampling data. Use of this procedure is recommended in insect population studies as a method of optimally planning experiments, and as a method of making precise conclusions about the significance of specific effects.     

On Fisher's Method of Combining p-Values

The problem of combining p-values from independent experiments is discussed. It is shown that Fisher's solution to the problem can be derived from a “weight-free” method that has been suggested for the purpose of ranking vector observations (Biometrics 19: 85–97, 1963). The method implies that the value p = 0.37 is a critical one: p-values below 0.37 suggest that the null hypothesis is more likely to be false, whereas p-values above 0.37 suggest that it is more likely to be true.     

循证护理对肝硬化腹水患者的疗效分析

目的:循证护理根据有价值的、可信的科学研究结果,提出问题,寻找实证,对病人采用最佳的护理模式进行护理。本文通过临床观察,旨在探讨循证护理对肝硬化腹水患者的临床疗效。方法:回顾性分析我院2009年2月到2012年6月收治的74例肝硬化腹水患者的临床资料,根据不同的护理方法将所选患者分为干预组和对照组,每组各37例。对照组患者给予常规护理,干预组患者在常规护理的基础上给予循证护理。观察两组患者的心理功能、生活能力、社交功能的改善情况,并比较两组患者的并发症发生情况及治疗的总有效率。结果:在接受不同的护理干预后,干预组患者的心理功能、生活能力、社交功能均显著优于对照组患者,差异具有统计学意义(P0.05);干预组患者的并发症发生率为5.40%,显著低于对照组患者的并发症并发率16.22%,差异显著有统计学意义(P0.01);干预组患者的总有效率为94.59%,明显高于对照组患者的总有效率83.78%,两组比较差异显著(P0.05)。结论:循证护理模式有助于肝硬化腹水患者的心理功能、生活能力、社交功能得到改善,临床效果满意,值得推广。     

Did natural selection make the Dutch taller? A cautionary note on the importance of quantification in understanding evolution

One of the main achievements of the modern synthesis is a rigorous mathematical theory for evolution by natural selection. Combining this theory with statistical models makes it possible to estimate the relevant parameters so as to quantify selection and evolution in nature. Although quantification is a sign of a mature science, statistical models are unfortunately often interpreted independently of the motivating mathematical theory. Without a link to theory, numerical results do not represent proper quantifications, because they lack the connections that designate their biological meaning. Here, we want to raise awareness and exemplify this problem by examining a recent study on natural selection in a contemporary human population. Stulp et al. (2015) concluded that natural selection may partly explain the increasing stature of the Dutch population. This conclusion was based on a qualitative assessment of the presence of selection on height. Here, we provide a quantitative interpretation of these results using standard evolutionary theory to show that natural selection has had a minuscule effect.     

Comparison of t-tests for differences in sexual dimorphism between populations

A t-test that can be used for evaluating the significance of differences in metric sexual dimorphism between populations is derived directly from mathematical considerations of the differences between distributions. It is compared with the t-test derived by Relethford and Hodges (1985), which was based upon linear regression with sex as a dummy variable. Both are determined to be mathematically equivalent, though the one derived here is more similar in form to traditional t-tests of differences and therefore may be simpler to employ. Both tests require only summary statistics for comparisons between populations and comparisons between generations within populations.