Supplementary Microarray Data
Wang J, Coombes KR, Highsmith WE, Keating MJ, Abruzzo LV.
Differences in gene expression between B-cell chronic lymphocytic
leukemia and normal B cells: a meta-analysis of three microarray
studies. Bioinformatics. 2004 Nov 22;20(17):3166-78.
This paper introduces a novel method for combining data from multiple
microarray platforms. We applied the method to data from three studies
of CLL. Two of the studies were carried out elsewhere:
- Lymphochip glass array data
from the publication by Rosenwald et al. is available
Microarray Database. To find the data, it's probably best to limit
the organism to Homo sapiens and sort by author.
- Affymetrix array data from the publication by Klein
et al. was obtained by contacting the senior author, Riccardo Dalla-Favera.
The third data set was produced a few years ago in the laboratory
of Dr. L.V. Abruzzo. This data set consists of microarray data on 12
samples (6 from CLL patients, 6 from peripheral blood B cells from
normal controls). The experiments were performed on Research Genetics
nylon microarrays, using radioactively labeled target
samples. Following standard practixe at the time, each membrane was
stripped and re-used in multiple experiments. Our paper describing sources
of variation on nylon membrane arrays found that membranes could
be safely reused up to four times. Each Research Genetics nylon
microarray contained 5760 spots; a complete set of 6 different
microarrays provided genome-wide coverage, using a total of more than
30000 cDNA clones.
Processed nylon microarray data
- Normalized array data in a single
tab-separated-values text file, with 12 columns and 34,560
rows. Includes a header row with sample names and an ID column
containing numbers that are indices into the annotation file.
- Gene annotations in a single
tab-separated-values text file with 34,560 rows. Row indices in this
annotation file are matched with the normalize array data. Blank
spots have an empty annotation list. Most annotations are old; only
the GenBank accession numbers (acc) should be viewed as reliable.
Raw nylon microarray quantification data