Table 3 |
|||
|
Improved accuracy using updated probe set definitions |
|||
|
RMA |
Mean Slope |
p-value |
Std |
|
|
|||
|
NetAffx (original) |
0.74 |
0.02 |
|
|
ensEMBL gene |
0.83 |
0.00040 |
0.01 |
|
ensEMBL transcript |
0.83 |
0.00053 |
0.01 |
|
Entrez |
0.78 |
0.00430 |
0.01 |
|
RefSeq |
0.78 |
0.00381 |
0.01 |
|
UniGene |
0.75 |
0.07085 |
0.01 |
|
|
|||
|
The accuracy between measured log2 ratios using RT-PCR and microarrays were compared for microarray measurements using the original (NetAffx) and updated probe set definitions across 16 genes. The accuracy was obtained by calculating the slope from a linear regression comparing the obtained log2 ratios estimates between the sample groups from the microarray to those estimates obtained using RT-PCR, where the microarray estimates were obtained using the different probe set definitions. A slope of 1 would indicate perfect accuracy. The table shows the mean of the obtained accuracy across the five labs. We performed a t-test to assess the difference in means of accuracy between the updated probe set definitions and the original probe set definition (NetAffx); p-values were calculated using a paired t-test (two-tailed distribution). The standard deviation for the mean is also shown. The data represents an assessment of RMA normalized data. |
|||
|
Sandberg and Larsson BMC Bioinformatics 2007 8:48 doi:10.1186/1471-2105-8-48 |
|||