my aim is find genes not differentially expressed in several RNA-Seq experiments. First i calculated coefficients of determination to see if genes expression value were correlate, i've also calculate regression line interval of confidence of angular coefficient and intercept.
Ideally these intervals must contain value 1 for angular coefficient, 0 for intercept, but these requirements were satisfied only in a few models.
The expression value were calculeted in the same way in every GSE, so my question is: there's a normalization method or something similar to improve data fitting?
Any suggestion is welcome.