The purpose of filtering is to increase the statistical power of differential expression analysis by removing any genes that are less likely to be informative. There are three common strategies:
Low variance filter: genes whose expression values do not change across different samples, and thus have very low variance. Genes are ranked by their variance from low to high, and you can exclude a certain percentile of genes with the lowest variance by adjusting the “Variance filter” slider.
Low abundance filter: genes with very low abundance are not measured reliably and may not be biologically important. You can exclude genes below a certain threshold by adjusting the “Low abundance” slider. The above referenced study has suggested 10% genes can be removed based on their abundance with improved results
Difficult-to-measure filter: some experiments including QC samples or technical replicates - features that changes a lot in those replicates cannot be measured reliably and should be excluded in downstream analysis.
Please refer to the paper Independent filtering increases detection power for high-throughput experiments for detailed discussion and benchmark tests. The study has suggested that up to 50% genes (!) can be removed based on their variance with improved results.