The statistical analysis within Mass Dynamics is performed using the R package Limma (https://bioconductor.org/packages/release/bioc/html/limma.html), one of the most popular packages for this differential expression analysis.
Limma computes log fold changes, p-values, and adjusted p-values (q-values) for each protein and for each pairwise comparison using the conditions available in the experiment design and after log-transforming raw intensities. If missing values are detected, they will be imputed using the same imputation as adopted in Perseus (https://maxquant.net/perseus/) and for each comparison, only proteins with at least 50% available measurements are considered.
All the differential expression results are then made available in our platform to be explored interactively via volcano plots, summary tables and other visualizations to allow quick selection and further analysis of the differentially abundant proteins.
For the selection of differential abundant proteins, a user is able to chose a desired log fold change thresholds.
Currently, we set our adjusted p-value threshold to 5% but we are working to allow this threshold to also be dynamically chosen.
Regarding how we handle the statistics for the differential abundance analysis, there is a complete overview of how the analysis is performed following the workflow here - (Bloom et al. 2021).