Quality control / Agilent


Creates some basic plots for Agilent quality control.



Three plots are created: boxplot, density plot and MA plot. Boxplot and density plot give an idea about the distribution of the data, and MA plot can be used for assessing the possible nonlinearity in the data caused by dye bias.


Three images visualizing the plots mentioned above.

The arrays should be centered around the same number in the boxplot. Deviant arrays can be identified by the deviation from the common median value or by a spread that is greater than for other chips.

The density plot visualizes the spread of expression values. Each lines corresponds to the smoothed histogram for one chip. Optimally, all lines should have a similar form, and have their peaks at the same location. Usually this is not reached, and the chips that deviate from others can be identified by their flatter line or a higher spread.

The MA plot rotates the usual scatterplot visualization 45 degress to the right, and plots an average of red and green channels on the x-axis, and the normalized red/green ratio on the y-axis. In addition, a lowess smoother line is plotted to every image with red. Optimally, the MA-plots for all chips should look similar, and the lowess line should show a flat line for chips.