JSM 2004 - Toronto

Abstract #300859

This is the preliminary program for the 2004 Joint Statistical Meetings in Toronto, Canada. Currently included in this program is the "technical" program, schedule of invited, topic contributed, regular contributed and poster sessions; Continuing Education courses (August 7-10, 2004); and Committee and Business Meetings. This on-line program will be updated frequently to reflect the most current revisions.

To View the Program:
You may choose to view all activities of the program or just parts of it at any one time. All activities are arranged by date and time.

The views expressed here are those of the individual authors
and not necessarily those of the ASA or its board, officers, or staff.

Back to main JSM 2004 Program page
Activity Number: 415
Type: Contributed
Date/Time: Thursday, August 12, 2004 : 8:30 AM to 10:20 AM
Sponsor: Biometrics Section
Abstract - #300859
Title: The Impact of Loess Normalization on Intraslide Correlation in Microarray Data
Author(s): Eric R. Siegel*+ and John Thaden and Pippa M. Simpson
Companies: University of Arkansas for Medical Sciences and University of Arkansas for Medical Sciences and Arkansas Children's Hospital
Address: 4301 West Markham St, #781, Little Rock, AR, 72205,
Keywords: microarray ; normalization ; correlation ; variance components ; loess ; locally weighted regression
Abstract:

Locally weighted regression (loess, lowess) is often used with demonstrated effectiveness to "normalize" or correct microarray-derived gene expression data for spot intensity differences, spatial effects, and dye bias. Less is known about its impact on systematic variation due to treatment. We used variance components analysis to study the impact of loess normalization on intraslide correlation. Our data come from SMD microarray slides gridded with DNA probing >17,000 C. elegans genes, with more than 1,000 probe species spotted at least twice per slide. After no adjustment or mixed-models adjustment by slide, we subjected the data to no normalization or to loess normalization by slide, by print-tip, and by print-tip with rescaling to a common Median Absolute Deviation. Then we estimated the intraslide correlation of each replicated probe as the ratio of its between-slide variance to the sum of its between-slide and within-slide variances. When we compared the resulting distributions, we found that all loess normalizations markedly reduced intraslide correlation. We investigate the conditions under which this would occur and how this influences treatment effect.

  • The address information is for the authors that have a + after their name.
  • Authors who are presenting talks have a * after their name.

Back to the full JSM 2004 program

JSM 2004 For information, contact jsm@amstat.org or phone (888) 231-3473. If you have questions about the Continuing Education program, please contact the Education Department.
Revised March 2004