Scholars Central

Conference abstractPublished 10 Jun 2013

The contaminated chi-square model and its application to microarray data analysis

Authors: Richard Charnigo

Ask about this article

Abstract

Select any words in the title or abstract, or press this and choose a sentence.

A Contaminated Beta (CB) model may be used to describe the distribution of P values corresponding to a large number of hypothesis tests, such as those assessing differential expression at numerous genes on a microarray. If those P values originate from Z or T statistics, then one may also describe the Z or T statistics themselves using a Contaminated Normal (CN) model. An advantage of the CN model over the CB model is that, under some circumstances, there may be more power to reject an omnibus null hypothesis of no differential expression and more accurate estimation of the proportion of genes differentially expressed. However, comparisons involving three or more groups may give rise to chi-square or F statistics rather than Z or T statistics, so that the CN model is inapplicable in these instances. Therefore, we propose using a Contaminated Chi-Square (CCS) model to describe the distribution of chi-square or F statistics corresponding to a large number of hypothesis tests. Using the CCS model, we present methodology for testing an omnibus null hypothesis of no differential expression and for estimating the proportion of genes differentially expressed. Through simulation studies and real data analyses, we ascertain the circumstances under which using the CCS model for chi-square or F statistics is preferable to using the CB model for P values.

Ask about this article

Answers come only from this article's title, abstract and keywords, each sentence pointing to where it comes from. Your question is sent to OpenAI to write the answer. Scholars Central does not store it.

Try:

Similar by topic

More from this journal

The newest in Journal of Biometrics & Biostatistics.

All its articles

Follow

New articles from what you follow go into your email digest, once you agree to alert emails.

  • JournalJournal of Biometrics & Biostatistics
  • SubjectMathematics & Statistics
  • BylineArticles whose byline prints “Richard Charnigo”

For authors and organisers

Publish, present and keep up with your field

Find the right journal for your paper and submit your manuscript.

Photo: Pexels