Please use this identifier to cite or link to this item:
https://hdl.handle.net/2440/56231
Citations | ||
Scopus | Web of ScienceĀ® | Altmetric |
---|---|---|
?
|
?
|
Type: | Journal article |
Title: | Some statistics in bioinformatics: The fifth Armitage Lecture |
Author: | Solomon, P. |
Citation: | Statistics in Medicine, 2009; 28(23):2833-2856 |
Publisher: | John Wiley & Sons Ltd |
Issue Date: | 2009 |
ISSN: | 0277-6715 1097-0258 |
Statement of Responsibility: | Patricia J. Solomon |
Abstract: | The spirit and content of the 2007 Armitage Lecture are presented in this paper. To begin, two areas of Peter Armitage's early work are distinguished: his pioneering research on sequential methods intended for use in medical trials and the comparison of survival curves. Their influence on much later work is highlighted, and motivate the proposal of several statistical 'truths' that are presented in the paper. The illustration of these truths demonstrates biology's new morphology and its dominance over statistics in this century. An overview of a recent proteomics ovarian cancer study is given as a warning of what can happen when bioinformatics meets epidemiology badly, in particular, when the study design is poor. A statistical bioinformatics success story is outlined, in which gene profiling is helping to identify novel genes and networks involved in mouse embryonic stem cell development. Some concluding thoughts are given. |
Keywords: | sequential trial survival analysis some statistical truths AIDS epidemiology microarray data analysis bioinformatics serum proteomic test for ovarian cancer embryonic stem cells pluripotency gene expression gene profiling proteomic profiles for ovarian cancer |
DOI: | 10.1002/sim.3668 |
Grant ID: | ARC |
Published version: | http://dx.doi.org/10.1002/sim.3668 |
Appears in Collections: | Aurora harvest 5 Mathematical Sciences publications |
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.