Hi, I would advise against this. I understand that you want to highlight the most interesting genes by leaving out those whose expression is unaffected. However, people are used to see "classic" volcanoplot and would be rightfully confused if some genes where left out. It also has the advantage to provide context to the visualization – most genes are unaffected, but a few are strongly up- or down--regulated.

A possible solution that allows to keep all genes in the visualization while highlighting the differentially expressed genes would be to use different colors. For instance light-transparent grey for not differentially expressed genes (including those between -1 and 1 log2foldchange) and blue for the others, like this:

I second the opinion that cutting out those data points is probably not a good idea. Instead, you can color-code the points to distinguish those that meet both p-value and log2FC cutoffs vs those that only meet one or the other. I suggest the Enhanced Volcano package for that purpose. Lucky for you (and all of us), its author is very active on Biostars and may be able to help if needed.

• https://github.com/kevinblighe/EnhancedVolcano
• https://bioconductor.org/packages/release/bioc/vignettes/EnhancedVolcano/inst/doc/EnhancedVolcano.html

An example plot demonstrating what I suggested above: