Occasionally, you may see a message in the Details section of the Experiment List view notifying you that some genes are missing.
For example: "There are 516 genes missing values out of 6847 genes in this experiment. You may wish to create a replicate set to compensate for the missing data."
This message may appear for one of the following reasons:
•The annotation file imported for your project contains annotations for genes that are not present in your experiments. If this is unexpected, importing the correct annotation file should solve the problem. Or, if you imported annotations through the Data Import Wizard, first verify that you designated the correct columns during the import, and then re-import if necessary.
•Some of the chips in your project contain genes that are not present on other chips in the same project. If this is unexpected, importing the correct data file should solve the problem. Or, if you imported your data through the Data Import Wizard, first verify that you designated the correct columns during the import, and then re-import if necessary.
•Some of the genes in your project do not have signal values associated with them. You can look in the Gene Table to find genes that are missing values. Genes without signals will show --- in place of values. In this case, it may be beneficial to add more replicates if possible, and then group your replicates into replicate sets. The averaging method that you select for each replicate set (mean or median) will determine how the missing values are handled. In both cases, the missing replicates will not be counted for that gene. In mean averaging, any gene that has no value in one or more replicates will cause the other replicates to have more weight for that gene.