Remove unwanted variation retrieves unknown experimental designs

Abstract Remove unwanted variation (RUV) is an estimation and normalization system in which the underlying correlation structure of a multivariate dataset estimated from negative control measurements, typically gene expression values, are assumed to stay constant across experimental conditions. In this paper we derive weight matrix incorporated into generalized least squares estimates RUV‐inverse, show that average covariance measurements. RUV‐inverse can thus be viewed as method adjusting for unknown design. We balanced incomplete block design (BIBD), recovers intra‐ interblock relevant parameters combines them weighted sum just like best linear unbiased estimator (BLUE), except weights globally measurements instead being individually optimized each measurement classical, single BIBD BLUE.