Search Results

You are looking at 1 - 10 of 25 items :

Clear All

Abstract

High-throughput sequencing techniques are increasingly affordable and produce massive amounts of data. Together with other high-throughput technologies, such as microarrays, there are an enormous amount of resources in databases. The collection of these valuable data has been routine for more than a decade. Despite different technologies, many experiments share the same goal. For instance, the aims of RNA-seq studies often coincide with those of differential gene expression experiments based on microarrays. As such, it would be logical to utilize all available data. However, there is a lack of biostatistical tools for the integration of results obtained from different technologies. Although diverse technological platforms produce different raw data, one commonality for experiments with the same goal is that all the outcomes can be transformed into a platform-independent data format – rankings – for the same set of items. Here we present the R package TopKLists, which allows for statistical inference on the lengths of informative (top-k) partial lists, for stochastic aggregation of full or partial lists, and for graphical exploration of the input and consolidated output. A graphical user interface has also been implemented for providing access to the underlying algorithms. To illustrate the applicability and usefulness of the package, we integrated microRNA data of non-small cell lung cancer across different measurement techniques and draw conclusions. The package can be obtained from CRAN under a LGPL-3 license.

. 1. INTRODUCTION There is an extensive literature on the construction of bivariate exponential models, for example, Gumbel [9], Freund [7], Downton [5] and so on. Marshall and Olkin [11] proposed a multivariate extension of exponential distributions which is much of interest in both theoretical development and applications. Block and Basu [4] proposed an absolutely continuous modification of the Marshall and Olkin's [11] model. Hyakutake [10] gave the Marshall and Olkin's [11] model having location parameters. AMS 1980 aubjeot Classification: 62 F 07 Key

: 62 F 07, 62 G 99 Key words and phrases: Subset selection, locally optimal, locally strongly monotone, joint type II censoring 2 S. S. Gupta, TaChen Liang In practice, it sometimes happens that the actual values of the random variables can only be observed under great cost, or not at all, while their ordering is readily observable. This occurs for instance in life-testing when one only observes the order in which the parts under investigation fail without being able to record the actual time of fail- ure. In problems of this type, one may desire to

cat ions : 62 F 07, 61 D 05 Key words and phrases: Comparison with a control; Dir ichlet integral; inverse sampling procedure; least favorable configuration; multinomial d istr ibut ion 34 P. Chen of simultaneous comparison of k given categories among themselves and with reference to a standard or control is of practical interest. This problem has been investigated by several authors under different types of formulations with different criteria to be satisfied by an acceptable procedure. In this paper, we study the corresponding problem for multi- nomial

different drugs for a specific disease. In such situations, a natural goal is to identify the best population, where quality of the ith population is measured in terms of i = 1 , . . . , k. A population π1 is said to be bet ter than irj if i?, > . Let denote the i th smallest of . . . , ΰ^, and let π^) denote the population associ- ated with i = 1 , . . . , k. Assume tha t there is no prior knowledge about which of . . . ,-TTfc is Call a population t t ^ j as the "best" population. In case of ties, we AMS 1980 subject classification: 62 F 07 Key words and phrases

Washington, D. C., August 13-16. AMS 1980 subject classification: 62F07, 62K99. Keywords and phrases: Subset selection, repeated measurements designs, multivariate normal and multivariate t, expected subset size. 64 H. J. Chen so on. Because of large differences between the individuals, the re- sponse of them to the same treatment may show relatively large vari- ability, part of which is due to differences between individuals exist- ing prior to the experiment. Since the treatments' effects are fixed and eliminating the variability between individuals will produce

propose sequential tests which are shown to have a substantial (asymptotic) saving in the average sample sizes compared to the known procedure now used in practice [see Gibbons et al. (1977), Section 2.3]. Our procedures are developed through comparing likelihoods, rather than just deciding through the largest sample mean (which is the usual practice). AMS 1970 subject classifications. 62 F 07, 62 L 10, 62 C 25 Key words and phrases. Ranking and selection, multiple hypoth- eses testing, sequential analysis, sample-size economy. 1 7 2 Ν. Mukhopadhyay 1

theory of rank tests. It is assumed that the number (or percentage) of students falling in each of the ordered groups is speci- fied in advance and the goal is to find a strategy for which the risk func- tions (or expected loss) is bounded above by a preassigned small e > 0 (also called 1-P below). The number of tests given to the class is also assumed to be fixed in advance. Comparisons are made between the different risks of the proposed strategies when the parameters are in the least favorable con- figuration. AMS 1980 subject classifications: 62 F 07, 62 H 30

respect to every reasonable loss function, procedures of the type (ψ, ψ*, &*) form an essentially complete class within fij. AMS 1970 subject classifications. 62 F 07, 62 F 05, 62 F 15, 62 L 99 Key words and phrases. Multiple decision procedures, two-stage selec- tion procedures, screening procedures, sequen- tial procedures, Bayes i an procedures, essential- ly complete class. 428 Shanti S. Gupta-Klaus J. Miescke 1. Introduction Let π^,.,.,π^ be k given populations associated with unknown parame- ters θ^,.,.,θ^ e Ω, where n e IR is ari unbounded or bounded

, are observed sufficient statistics and π is the sign of permutation. In papers [4] and [1] the ranking of populations is performed according to the initial Bechhofer’s treatment of ranking problem [2], namely: λ1 contains k1 largest components of θ , λ2 contains k2 next largest components of θ , . . . , and λs contains ks smallest components of θ . However, our result cannot be obtained on the basis of these papers due to an additional condition εi = 1. AMS 1991 Subject Classifications: Primary 62F07; Secondary 62N05 Key words and phrases: ranking of populations