Additional File 2:

Table of the of 3,471 common target set. Shown in the table for each of the 3,471 common targets RefSeq identifiers (CommonREFSEQ_4platform) are the HUGO gene symbol (Gene Symbol) and Entrez Gene public data bank identifiers. Also given are the intra-platform intensity quantile attribution calculated for each platform respectively (PX Q, AFFYMETRIX.Q, AGILENT.Q, AMERSHAM.Q) as well as the consensus quantile (cQ) calculated based on the mean quantile attribution obtained for each platform. ANOVA targets (p < 0.001) are indicated for each platform ("1") where "0" indicates a target not filtered and not an ANOVA target. AMER_MCF_BATT, AFFY_MCF_BATT, AGILENT_MCF_BATT indicates the targets that passed the p < 0.001 filter for the respective t tests comparing the MCF7 and Batt P12 samples (see Figure 4 and Tables 1 and 2). Filtered log₂-intensities are given for each slide hybridized by each platform (PX, Affymetrix (AFFY), Amersham (Amer) or the filtered log₂-ratios obtained for Agilent (Agilent44K). "NA" is shown for all empty fields corresponding to filtered data points.

Format: XLS Size: 1.8MB Download file

This file can be viewed with: Microsoft Excel Viewer

Additional File 5:

Cluster analysis using the common ANOVA targets. Sample cluster dendrograms of all 36 platforms samples alone (A) or with other data including 66 tumor samples (B) and gene cluster dendrogram with associated heat map based on the expression profile of 1,096 (A) common ANOVA targets (or 925 U133A targets mapped to the 1,096 targets for B) found between Affymetrix, Agilent and Amersham. Mapping to the Affymetrix U133A array was based on RefSeq identifiers. Clustered sample groups corresponding to the cell line and similar tumor type are indicated by colored bars. Log₂-ratios were centered and color coded (below heat map B). MCF-7 samples are represented with blue bars, BattP12 and mesothelioma samples are represented with red bars while Jurkat and TALL samples are represented by green bars. BattP12 is a mesothelioma cell line and clusters with the MESO samples (red bar) which as a group clusters with the MCF7 samples (blue bar). Jurkat is a T-cell acute lymphoblastic leukemia and clusters with T-ALL samples (green bar). As a part of the CIT program we obtained raw Affymetrix HG-U133A data for 66 samples corresponding to either pooled tumors of a particular cancer type and/or derived cell lines (unpublished data). These samples include the universal reference RNA from Stratagene (see above). Tissues represented are as follows: B-cell acute lymphoblastic leukemia (BALL, n = 4); colon cell line (COLO, n = 8); follicular lymphoma (FOLL, n = 4); hepatocarcinoma (FOIE, n = 3); myeloblastic acute leukemia (LAM, n = 3); epidermotrophal lymphoma (LEPI, n = 2); chronic lymphoblastic leukemia (LLC, n = 1); myeloblastic chronic leukemia (LMC, n = 2); mesothelioma cell line (MESO, n = 8); epithelial cells (PEAU, n = 4); lung (POUM n = 3); T-cell acute lymphoblastic leukemia T (TALL, n = 20); thyroid (THYR n = 4). Labeled cRNA reactions, hybridizations and image analyses for all 66 samples were carried out at the IGBMC, Strasbourg, France (March, 2002). Raw data was normalized and summarized expression values were generated using RMA (described above). For all cluster analyses, normalized intensity values were transformed into log₂-ratios using the reference sample as the denominator. Cluster analysis was performed using log₂ ratios (ratios calculated as described above) and DNA-Chip Analyzer (dChip) Version 1.3, with the following parameters: distant metric = 1-pearson correlation; linkage = centroid; row standardization.

Format: PDF Size: 401KB Download file

This file can be viewed with: Adobe Acrobat Reader

Additional File 4:

Array and QRT-PCR data for the 15 genes analyzed by QRT-PCR. This table is divided into three sections. Section I shows the filtered log₂ intensities (Affymetrix and Amersham) or log₂-ratios (Agilent) for the 15 genes analyzed by QRT-PCR. RNA batches are indicated above each table. Multiple targets were available for ERBB2 and RPS6KB1. Section II shows Ct values for each RNA batch obtained for each of the 2 batches (RNA batches are indicated above each table where Batch 1: green; Batch 2: pink). For sections I and II A red X is given for data points that were either filtered out by a given platform (microarray data) or for genes that were below the level of detection in a given cell line determined by QRT-PCR. Section III shows the number of possible versus available gene pair combinations QRT-PCR fold change bins used in Figure 5. The number of possible gene pair combinations per fold change bin were based on the distribution of relative expression levels calculated using the -ΔΔCt values to compare with log₂-ratio array data (cell line/reference for all 3 platforms, shown in yellow) or using -ΔCt values to compare with log₂ intensities (from Amersham and Affymetrix, shown in blue). Nine different bins were created with the following code (in bold) corresponding the following log₂-ratios (x) intervals: <-5: x < -5; -3: -5 = x < -1.25; -1: -1.25 = x < -0.75; -0.5: -0.75 = x < -0.25; 0: -0.25 = x < 0.25; 0.5: 0.25 = x > 1.25; 1: 0.25 = x > 1.25; 3: 1.25 = x > 5.0; > = 5: x = 5.0.

Format: XLS Size: 33KB Download file

This file can be viewed with: Microsoft Excel Viewer

Additional File 1:

Description of the 3 array formats used in the comparisons. The numbers represent the counts made of total number of targets represented on the respective microarrays. ¹ The number of unique public sequence identifiers (ID) represented on the respective microarrays. These counts are based on the annotation table provided at the time the experiments were carried out (March – June, 2004). Control targets were excluded from the counts.

Format: XLS Size: 20KB Download file

This file can be viewed with: Microsoft Excel Viewer

Additional File 3:

Primer/Probe information for the 15 genes analyzed by QRT-PCR. Shown in the table are the 15 genes (Genes Names), accession number, forward and reverse primers and probes with associated 5'-3' sequence, 5' position and size analyzed by QRT-PCR.

Format: XLS Size: 29KB Download file

This file can be viewed with: Microsoft Excel Viewer