Baseline Quantification

Use this parameter to specify the statistics for quantifying the baseline across columns in the Input SAS Data Set (or the specified Baseline Reference SAS Data Set) for measurements with the same rank.

Statistics for quantifying the baseline are described in the table below:

Statistic	Description
Geometric Mean	Constructs the baseline using the geometric mean across all of the data columns. This option should be selected for non-log transformed count data.
Arithmetic Mean	Constructs the baseline using the arithmetic mean across all of the data columns. This is the default option of quantile normalization.
Median	Constructs the baseline using the median across all of the data columns. This option should be selected for count data1.
Upper Quartile	Constructs the baseline using the upper quartile across all of the data columns. This option should be selected for count data.

---

1

Bullard, JH., E. Purdom, KD Hansen, and S. Dudoit. 2010. Evaluation of statistical methods for normalization and differential expression in mRNA-Seq experiments. BMC Bioinformatics 11:9.