QuickSpots Antibody Array Analysis Documentation
Index » Results TabNormalization 
Normalization is a computation used to make values from different data sets directly comparable to each other. Without normalization, differences in exposure or processing between arrays may affect all of the values on one or more arrays. For example, the values from one array may be 5% more intense than the values from another array simply because it was slightly overexposed at some point during processing. Therefore, direct comparison of the numbers between the data sets would be misleading.
Even without normalization, it is possible to discover the impact of the different treatments given to each array by observing how different spots vary in relation to each other. But in order to directly compare numbers between arrays, normalization is required. For normalization to be meaningful, be sure to keep Background Subtraction selected.
The documentation for each array model should indicate whether the reference spots (sometimes called positive reference spots, or some such thing) can be used for normalization. On some arrays, the reference spots are designed only to be used to determine proper orientation and alignment of the array, and not for normalization. On other arrays, the composition of the reference spots is carefully controlled so that the intensity of the reference spots can be used for normalization. We have attempted to ensure that this setting is initially turned off for arrays that do not support normalization using reference spots, and is initially turned on for arrays that do support it. If you have any question, please refer to the manufacturer’s documentation.
Even for arrays that do not support normalization using reference spots, you can choose to normalize using any spot pair on the array other than the negative controls. If the array contains multiple reference spots or pairs, they are all grouped together for this purpose and cannot be selected separately.
Normalization using spots other than the positive references can help you to see how different spots co-vary with each other. Within one experiment, you may wish to normalize with respect to multiple different spots in order to examine different relationships. It is up to you to determine whether and how to use normalization, based on the details of your experiments.