Title: Optimized LOWESS normalization parameter selection for DNA microarray data

By: John A. Berger, Sampsa Hautaniemi, Anna-Kaarina Järvinen, Henrik Edgren, Sanjit K. Mitra and Jaakko Astola

BMC Bioinformatics, vol. 5, no. 194, Dec. 2004

This webpage contains supplementary information that is meant to accompany the BMC Bioinformatics manuscript.

Self versus self microarray experiments

Please consult original manuscript for complete details. Briefly, for the self versus self hybridizations, custom cDNA microarray experiments proceed as follows. Altogether, three microarray hybridizations were performed using custom printed cDNA microarray slides from the same print batch. Labelling, hybridization and washing were done as described previously by Monni et al. [1] and Järvinen et al. [2]. Briefly, total RNA was extracted from cell lines BT-474, HBL-100, and MCF-7 and labelled with Cy3-dUTP and Cy5-dUTP (Amersham Biosciences, Piscataway, NJ). The custom printed cDNA microarrays comprised of 11,520 clones from Incyte Genomics IRAL cDNA library and 1,136 clones from Research Genetics library. Microarrays were printed on poly-l-lysine coated slides using an Omnigrid arrayer (GeneMachines) as described previously [2]. Microarrays were scanned with an Agilent laser confocal scanner (Agilent Technologies, Palo Alto, CA) and gridded using the DEARRAY software developed by Chen et al. [3, 4].

BT474 vs. BT474: TIFF Data and Table Data
MCF7 vs. MCF7: TIFF Data and Table Data
HBL100 vs. HBL100: TIFF Data and Table Data

Table Data is the output of the DEARRAY software [3, 4].

The pen locations are in the second column, which is titled "PenAt," while the red and green foreground intensity values are located in the columns titled "SR_Mean" and "SG_Mean". Background intensities are located in the columns titled "SR_bkMean" and "SG_bkMean". Ratio quality has been assessed and the values for each spot are reported in the column titled "rQuality." For our analysis, the spots with rQuality less than 0.5 have been removed from the preprocessing stages. The quality cutoff value has been shown to represent less reliable cDNA microarray measurements due to either low signal intensity, high local background level, uneven distribution of the target intensity, and/or small target size. For an unbiased assessment, spot quality assessment was also performed using the method by Hautaniemi et al. [5]. The MIAME information for these experiments are located here.

Breast cancer microarray experiments

Please click here for the replicated breast cancer experiments, which were originally published in [2].


[1] Monni O, Bärlund M, Mousses S, Kononen J, Sauter G, Heiskanen M, Paavola P, Avela K, Chen Y, Bittner M, Kallioniemi A: Comprehensive copy number and gene expression profiling of the 17q23 amplicon in human breast cancer. Proc. Natl. Acad. Sci. USA 2001, 98:57115716.

[2] Järvinen AK, Hautaniemi S, Edgren H, Auvinen P, Saarela J, Kallioniemi OP, Monni O: Are data from different gene expression microarray platforms comparable? Genomics 2004, 84:11641168.

[3] Chen Y, Kamat V, Dougherty ER, Bittner ML, Meltzer PS, Trent JM: Ratio statistics of gene expression levels and applications to microarray data analysis. Bioinformatics 2002, 18:12071215.

[4] Chen Y, Dougherty ER, Bittner ML: Ratio-based decisions and the quantitative analysis of cDNA microarray images. Journal of Biomedical Optics 1997, 2:364-374.

[5] Hautaniemi S, Edgren H, Vesanen P, Wolf M, Järvinen AK, Yli-Harja O, Astola J, Kallioniemi O, Monni O: A Novel Strategy for Microarray Quality Control Using Bayesian Networks. Bioinformatics 2003, 19:20312038.


John A. Berger
Dept. of Electrical and Computer Engineering
University of California
Santa Barbara, CA 93106-9560
Ph. (805) 893-8312
E-mail. berger (at) ece.ucsb.edu

Last Updated: April 18, 2005