Hi,

I have a little confusion while running DESeq2 (I have 2 conditions and three biological replicates). some genes display NA for p and p adjusted even though other values such as base mean and log2 fold change are all given. When I run DESeq on the same sample I get a p value rounded to zero for the same genes with NA with DESeq2. Can you please explain me the reason behind this variation? how can I fix it?

Best,

Tima

Look at the help for 'results' and read through the detail section, particularly through these paragraphs:

 On p-values:

     By default, independent filtering is performed to select a set of
     genes for multiple test correction which will optimize the number
     of adjusted p-values less than a given critical value ‘alpha’ (by
     default 0.1). The adjusted p-values for the genes which do not
     pass the filter threshold are set to ‘NA’. By default, the mean of
     normalized counts is used to perform this filtering, though other
     statistics can be provided. Several arguments from the
     ‘filtered_p’ function of genefilter are provided here to control
     or turn off the independent filtering behavior.

     By default, ‘results’ assigns a p-value of ‘NA’ to genes
     containing count outliers, as identified using Cook's distance.
     See the ‘cooksCutoff’ argument for control of this behavior.
     Cook's distances for each sample are accessible as a matrix
     "cooks" stored in the ‘assays()’ list. This measure is useful for
     identifying rows where the observed counts might not fit to a
     Negative Binomial distribution.

In other words, set independentFiltering = FALSE, and cooksCutoff = Inf, and the missing values from the p-values should disappear.