Batch effects are the systematic errors caused when samples are processed in different batches. They represent the nonbiological differences between the samples in an experiment. The reason can be the difference in sample preparation or hybridization protocol, and so on. It can be reduced, to some extent, by careful, experimental design but cannot be eliminated completely unless the study is performed under a single batch. Batch effects render the task of combining data from different batches difficult. This ultimately reduces the power of statistical analysis of the data. This needs appropriate preprocessing before the batches are combined. This recipe will present these preprocessing techniques.