How to use the OutlierD Package

It is important to preprocess high-throughput data generated from microarray or mass spectrometry experiments in order to obtain a successful analysis. Outlier detection is an important preprocessing step. For outlier detection, upper and lower fences (Q3+1.5IQR and Q1 − 1.5IQR) of the differences are often used in statistics, where Q1=lower 25% quantile, Q3=upper 25% quantile, and IQR = Q3 − Q1. However, heterogenous variability is often observed in high-throughput data. By ignoring heterogenous variability and using the fences that are constant over all values, we may often miss true outliers and detect false outliers. Therefore, the OutlierD package provides various quantile regression techniques (constant, linear, nonlinear, and nonparametric quantile estimators) on the M-A scatterplots, accounting for heterogenous variability to detect outliers with low false positive and negative rates in high-throughput data. The OutlierD package employs libraries quantreg and Biobase, which must be installed in advance.