Missing values estimation in microarray data with partial least squares regression

Abstract

Microarray data usually contain missing values, thus estimating these missing values is an important preprocessing step. This paper proposes an estimation method of missing values based on Partial Least Squares (PLS) regression. The method is feasible for microarray data, because of the characteristics of PLS regression. We compared our method with three methods, including ROWaverage, KNNimpute and LLSimpute, on different data and various missing probabilities. The experimental results show that the proposed method is accurate and robust for estimating missing values. © Springer-Verlag Berlin Heidelberg 2006.