...
Numbered figure captions |
---|
Prefix | Table |
---|
SubtitleText | Regression of the observed transcript abundance on the expected transcript abundance, transcript length, GC content, and the expected fold change, subjecting all the variables to log2 transformation. We started with the full model containing four covariates and performed model selection based on two criteria: adjusted r^2 (computed for all possible models) and stepwise regression (with a cutoff p-value of 0.15). The assessment was performed on the Mix A of ERCC, using Partek’s modified expectation-maximization (EM) algorithm for transcript quantification. |
---|
AnchorName | table1 |
---|
|
Image Modified |
Numbered figure captions |
---|
Prefix | Table |
---|
SubtitleText | Regression of the observed transcript abundance on the expected transcript abundance and transcript length (the best model). We started with the full model containing four covariates and performed model selection based on two criteria: adjusted r^2 (computed for all possible models) and stepwise regression (with a cutoff p-value of 0.15). The assessment was performed on the Mix A of ERCC, using Partek’s modified expectation-maximization (EM) algorithm for transcript quantification. |
---|
AnchorName | table2 |
---|
|
|
Numbered figure captions |
---|
Prefix | Table |
---|
SubtitleText | Regression of the observed transcript abundance on the expected transcript abundance (the benchmark model). We started with the full model containing four covariates and performed model selection based on two criteria: adjusted r^2 (computed for all possible models) and stepwise regression (with a cutoff p-value of 0.15). The assessment was performed on the Mix A of ERCC, using Partek’s modified expectation-maximization (EM) algorithm for transcript quantification. |
---|
AnchorName | table3 |
---|
|
|
The full and the best models for the Mix B are shown in Tables 4 and 5 (respectively). Apparently, the regression failed to find evidence of any kind of bias.
...