| Â | ensEMBL exon | ensEMBL gene | ensEMBL transcript_ | Entrez | RefSeq | UniGene |
|---|
MAS5
| -0.014 p = 0.094 | 0.013 p = 0.057 | 0.013 p = 0.0092 | 0.014 p = 0.056 | 0.019 p = 0.018 | 0.023 p = 0.052 |
RMA
| -0.035 p = 0.0041 | 0.053 p = 0.00025 | 0.030 p = 0.0071 | 0.059 p = 0.00011 | 0.045 p = 0.00045 | 0.047 p = 2.9E-06 |
GCRMA
| -0.11 p = 0.000051 | 0.023 p = 0.045 | -0.0063 p = 0.28 | 0.040 p = 0.019 | 0.031 p = 0.062 | 0.028 p = 0.0071 |
- We compared the precision in microarray experiments when using the updated probe set definitions in comparison with the original probe set definition (netAffx). We measured the precision as the Pearson correlation between two log2 ratios of expression levels comparing duplicate sample-vs-control measurements. The measurement indicates how well a fold change can be reproduced. Data from five different data sets generated using identical samples were averaged (mean) The table shows differences in means for updated probe set definitions compared to the original (NetAffx) definitions across the five different labs together with the p-value assessing the difference in means between each updated probe set definition compared to the original probe set definition (NetAffx) calculated using a paired t-test (two-tailed distribution). Three methods to generate expression estimates (Mas5, RMA and GCRMA) were compared.
