A conservative test for multiple comparison based on highly correlated test statistics

In genetics, we often encounter a large number of highly correlated test statistics. The most famous conservative bound for multiple comparison is Bonferroni's bound, which is suitable when the test statistics are independent but not when the test statistics are highly correlated. This article proposes a new conservative bound that is easily calculated without multiple integration and is a good approximation when the test statistics are highly correlated. The performance of the proposed method is evaluated by simulation and real data analysis.     

A comparison of methods for assessing toxicant effects on algal growth

Summary Ankistrodesmus braunii was used as the test organism to assess the effects of mercuric chloride on the growth of this species. Optical density measurements correlated highly with dry weight determinations, chlorophyll content, total cell counts, and respiring cell counts. Any one of the above methods may be used alone or in combination with other methods to assess the effects of toxicants on algal growth and survival.     

Simple test statistics for major gene detection: a numerical comparison

Summary We compare 22 simple tests for the detection of major gene segregation in livestock populations. These tests belong to two groups: methods based on the comparison of within-family distribution and methods based on the comparison of parents' and offspring performances. The power of the 22 tests and the robustness of the two more powerful of these 22 are evaluated by simulation. Thirteen types of major loci, differing in the within-genotype means, variances or alleles frequencies, are studied. Thirty hierarchically balanced populations defined by the number of sire families (5–20), dams per sire (1–20) and progenies per dam (1–20) are simulated. The quantiles are estimated from 2000 samples, the power from 1000 samples and the robustness from 100 samples. The more powerful tests are the within family-variance heterogenity test (Bartlett test) and the within-family mean-variance regression (Fain 1978). Their robustness may be very low, in particular when the trait distribution is skewed.     

A comparison of methods for assessing yeast viability

Summary Eight different methods were used to assess the cell viability of four strains of Saccharomyces. Staining with Mg-ANS, primuline yellow, FITC and methylene blue gave a good index of yeast cell viability. The standard plate count technique and microcolony formation also gave a good measure of cell viability. Fluorescent staining with acridine orange was the least useful of the methods tested. INT dye reduction gave a good index of respiring cells depending upon the yeast strain tested.     

A comparison of techniques for assessing dispersion patterns

Summary A series of artificial populations with different types of spatial distribution is used to test the value of a number of published indices in identifying non-randomness. Among tests based on distance measurements, that of Hopkins is generally the most powerful, through there are reasons for preferring that of Pielou where the vegetation has not been completely mapped. Among tests involving quadrat measurements, the variance/mean ratio appears to be best. Where contiguous series of quadrats are available hierarchical analysis of variance is a more powerful test than Monte-Carlo pairing of quadrats, though the latter is a more flexible technique.The value of different methods for describing pattern, once non-randomness has been established, is also discussed.This research was supported by Grants Nos. GB-12058, 15886 and 19843 from the National Science Foundation. Ronald Yorgason, Ken Rea and Bob Bayn gave assistance in computer programming and data analysis. Interpretation of results remains our responsibility.     

Haplotype-sharing analysis using Mantel statistics for combined genetic effects

We applied a new approach based on Mantel statistics to analyze the Genetic Analysis Workshop 14 simulated data with prior knowledge of the answers. The method was developed in order to improve the power of a haplotype sharing analysis for gene mapping in complex disease. The new statistic correlates genetic similarity and phenotypic similarity across pairs of haplotypes from case-control studies. The genetic similarity is measured as the shared length between haplotype pairs around a genetic marker. The phenotypic similarity is measured as the mean corrected cross-product based on the respective phenotypes. Cases with phenotype P1 and unrelated controls were drawn from the population of Danacaa. Power to detect main effects was compared to the X2-test for association based on 3-marker haplotypes and a global permutation test for haplotype association to test for main effects. Power to detect gene x gene interaction was compared to unconditional logistic regression. The results suggest that the Mantel statistics might be more powerful than alternative tests.     

A sequential test for assessing observed agreement between raters

Assessing the agreement between two or more raters is an important topic in medical practice. Existing techniques, which deal with categorical data, are based on contingency tables. This is often an obstacle in practice as we have to wait for a long time to collect the appropriate sample size of subjects to construct the contingency table. In this paper, we introduce a nonparametric sequential test for assessing agreement, which can be applied as data accrues, does not require a contingency table, facilitating a rapid assessment of the agreement. The proposed test is based on the cumulative sum of the number of disagreements between the two raters and a suitable statistic representing the waiting time until the cumulative sum exceeds a predefined threshold. We treat the cases of testing two raters' agreement with respect to one or more characteristics and using two or more classification categories, the case where the two raters extremely disagree, and finally the case of testing more than two raters' agreement. The numerical investigation shows that the proposed test has excellent performance. Compared to the existing methods, the proposed method appears to require significantly smaller sample size with equivalent power. Moreover, the proposed method is easily generalizable and brings the problem of assessing the agreement between two or more raters and one or more characteristics under a unified framework, thus providing an easy to use tool to medical practitioners.     

A friendly statistics package for microarray analysis

SUMMARY: The friendly statistics package for microarray analysis (FSPMA) is a tool that aims to fill the gap between simple to use and powerful analysis. FSPMA is a platform-independent R-package that allows efficient exploration of microarray data without the need for computer programming. Analysis is based on a mixed model ANOVA library (YASMA) that was extended to allow more flexible comparisons and other useful operations like k nearest neighbour imputing and spike-based normalization. Processing is controlled by a definition file that specifies all the steps necessary to derive analysis results from quantified microarray data. In addition to providing analysis without programming, the definition file also serves as exact documentation of all the analysis steps. AVAILABILITY: The library is available under GPL 2 license and, together with additional information, provided at http://www.ccbi.cam.ac.uk/software/psyk/software.html#fspma     

A step-up procedure for selecting variables associated with survival.

Multivariate concomitant information on a subject's condition usually accompanies survival time data. Using a model in which each subject's lifetime is exponentially distributed, this paper suggests a method which utilizes a step-up procedure for choosing the most important variables associated with survival. Maximum likelihood (ML) estimates are utilized, and the likelihood ratio is employed as the criterion for adding significant concomitant variables. An example using multiple myeloma survival data and sixteen concomitant variables is discussed in which three variables are chosen to predict survival.     

A shared parameter model for the estimation of longitudinal concomitant intervention effects

We investigate a change-point approach for modeling and estimating the regression effects caused by a concomitant intervention in a longitudinal study. Since a concomitant intervention is often introduced when a patient's health status exhibits undesirable trends, statistical models without properly incorporating the intervention and its starting time may lead to biased estimates of the intervention effects. We propose a shared parameter change-point model to evaluate the pre- and postintervention time trends of the response and develop a likelihood-based method for estimating the intervention effects and other parameters. Application and statistical properties of our method are demonstrated through a longitudinal clinical trial in depression and heart disease and a simulation study.     

Multilevel models for survival analysis with random effects

A method for modeling survival data with multilevel clustering is described. The Cox partial likelihood is incorporated into the generalized linear mixed model (GLMM) methodology. Parameter estimation is achieved by maximizing a log likelihood analogous to the likelihood associated with the best linear unbiased prediction (BLUP) at the initial step of estimation and is extended to obtain residual maximum likelihood (REML) estimators of the variance component. Estimating equations for a three-level hierarchical survival model are developed in detail, and such a model is applied to analyze a set of chronic granulomatous disease (CGD) data on recurrent infections as an illustration with both hospital and patient effects being considered as random. Only the latter gives a significant contribution. A simulation study is carried out to evaluate the performance of the REML estimators. Further extension of the estimation procedure to models with an arbitrary number of levels is also discussed.     

A case-cohort design for assessing covariate effects in longitudinal studies

The case-cohort design for longitudinal data consists of a subcohort sampled at the beginning of the study that is followed repeatedly over time, and a case sample that is ascertained through the course of the study. Although some members in the subcohort may experience events over the study period, we refer to it as the "control-cohort." The case sample is a random sample of subjects not in the control-cohort, who have experienced at least one event during the study period. Different correlations among repeated observations on the same individual are accommodated by a two-level random-effects model. This design allows consistent estimation of all parameters estimable in a cohort design and is a cost-effective way to study the effects of covariates on repeated observations of relatively rare binary outcomes when exposure assessment is expensive. It is an extension of the case-cohort design (Prentice, 1986, Biometrika73, 1-11) and the bidirectional case-crossover design (Navidi, 1998, Biometrics54, 596-605). A simulation study compares the efficiency of the longitudinal case-cohort design to a full cohort analysis, and we find that in certain situations up to 90% efficiency can be obtained with half the sample size required for a full cohort analysis. A bootstrap method is presented that permits testing for intra-subject homogeneity in the presence of unidentifiable nuisance parameters in the two-level random-effects model. As an illustration we apply the design to data from an ongoing study of childhood asthma.     

A comparison of response and production protocols for assessing perceived exertion

Two cycle ergometer protocols for assessing perceived exertion were compared before and after a fatiguing run. In the response (R) protocol, the subject rated the perceived exertion (RPE) of a series of power outputs assigned by the investigator. In the production (P) protocol, the investigator selected the RPE values and the subject adjusted his power output using a hand-held control. The relationship between RPE and power output (the regression coefficient and the slope and intercept of the regression line) was the same for both protocols. Fatigue due to the run caused a small increase in RPE (average 1.5 units) at a given exercise intensity and a commensurate decrease in power output (average 19 W) for a given RPE. The P protocol is safer than the R protocol because it makes no assumptions with regard to the physical condition of the subject. It is superior to the R protocol because it is an interval scale. These advantages suggest that the P protocol should be used instead of, or at least in addition to, the more traditional R protocol.