Fig. 1From: quantro: a data-driven approach to guide the choice of an appropriate normalization methodA schematic of quantile normalization. Quantile normalization is a non-linear transformation that replaces each feature value (row) with the mean of the features across all the samples with the same rank or quantile. To quantile normalize a raw high-throughput data set with multiple samples: (1) order the feature values within each sample; (2) for each feature, average across the rows; (3) substitute the raw feature value with the average; (4) re-order the transformed values by placing in the original orderBack to article page