Chapter 2. Review of Fundamental Statistical Concepts

  1. Dr Francisco Azuaje Senior Member Associate Editor

Published Online: 14 JAN 2010

DOI: 10.1002/9780470686423.ch2

Bioinformatics and Biomarker Discovery: “Omic” Data Analysis for Personalized Medicine

Bioinformatics and Biomarker Discovery: “Omic” Data Analysis for Personalized Medicine

How to Cite

Azuaje, F. (2010) Review of Fundamental Statistical Concepts, in Bioinformatics and Biomarker Discovery: “Omic” Data Analysis for Personalized Medicine, John Wiley & Sons, Ltd, Chichester, UK. doi: 10.1002/9780470686423.ch2

Author Information

  1. Public Research Centre for Health (CRP-Santé), Luxembourg

Publication History

  1. Published Online: 14 JAN 2010
  2. Published Print: 19 FEB 2010

ISBN Information

Print ISBN: 9780470744604

Online ISBN: 9780470686423

SEARCH

Keywords:

  • biomarkers;
  • diagnosis;
  • prognosis;
  • statistical analysis

Summary

In this chapter, basic statistical concepts and problems relevant to biomarker discovery research are introduced. This includes brief reviews of parameter estimation, error types and hypothesis testing, as well as guidelines for selecting and interpreting statistical tests. The problem of multiple-hypotheses testing is explained, including methods based on family-wise error and false discovery rates. Fundamental statistical analysis techniques are introduced: correlation, regression and classification, and survival analysis. The problem of assessing the predictive quality of biomarkers and models using different types of statistical measures is discussed. An introduction to data sampling size estimation is presented. The chapter concludes with a discussion on common pitfalls and misinterpretations in statistical analysis for biomarker discovery